Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
⚠ Concerns:(1) Has not raced in 101 days. (2) Moving from a route to a sprint today. (3) Poor record at this distance (0 wins from 4 recent starts). (4) Trip note from last race: 2p-ins; faded 1/8. (5) 0 wins from post 1 in 3 recent starts. (6) 0 wins from 9 starts on dirt in last 10 — repeated failure on today's surface. (7) First start at BTP — never raced here before (last seen at: TP, KEE).
Mod1(W): 100 (1/4) Mod2(W): 100 (1/4) Mod3(W): 100 (1/4) Mod1-LS: 100 (1/4) Mod2-LS: 100 (1/4) Mod3-LS: 100 (1/4) Mod4(Form): 100 (1/4) Win Prob: 52.7% Value Score: 2.1 ★ ML: 3.00 (2/1) Overlay %: 0.71 Fair Odds: 2.88 (9/5) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | A |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | A | B |
| Early Pace | A | A |
| Mid Pace | A | A |
| Late Pace | B | D |
| Finish Pos | A | F |
| Dirt Speed | A | A |
| Turf Speed | A | B |
| Distance Speed | A | B |
| Trainer Effectiveness | A | C |
| Jockey Effectiveness | A | C |
| 3rd Party Ratings | A | A |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | C |
| Early Pace | A | A |
| Mid Pace | A | A |
| Late Pace | B | D |
| Sprint FPS | A | B |
| Route FPS | A | A |
| Trainer Effectiveness | A | C |
| Jockey Effectiveness | A | C |
| 3rd Party Ratings | A | A |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | B |
| Early Pace | A | A |
| Mid Pace | A | A |
| Late Pace | B | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 44-day layoff with a recent workout. (2) Significantly better on off tracks (off: 50 vs fast: 41). (3) Big last-race effort (63) — 11 pts above recent average. (4) Equipment change today (code: 1). (5) Has a bullet work (fastest at track that day) — sharp conditioning signal. (6) Returning from 44-day layoff with a bullet workout — trainer signaling extra readiness for today's spot. (7) Carries 9 lbs less than field median (116 lbs vs median 125 lbs). (8) Speed improved +7 pts then +7 pts in last 2 starts (49 → 56 → 63) — trainer is progressively building fitness. (9) Had documented trouble last race (codes: ['D', 'L', 'O', 'S', 'T']) yet still finished 3rd — a clean trip today sets up a form reversal.
⚠ Concerns:(1) Has not raced in 44 days. (2) Speed average (47) well below field median (60). (3) Moving from a route to a sprint today. (4) Poor record at this distance (0 wins from 6 recent starts). (5) Tends to fade in the stretch at similar distances (5 of 6 starts). (6) Trip note from last race: Ins trio5/16;naild plc. (7) Last race was career-best speed (63) — bounce risk after peak effort. (8) 0 wins from 6 starts on dirt in last 10 — repeated failure on today's surface. (9) Two consecutive 5+ pt speed jumps (49 → 56 → 63) and last race speed is well above career average (47) — high bounce risk after back-to-back peak efforts.
Mod1(W): 2 (3/4) Mod2(W): 20 (4/4) Mod3(W): 55 (3/4) Mod1-LS: 45 (4/4) Mod2-LS: 30 (4/4) Mod3-LS: 71 (3/4) Mod4(Form): 80 (2/4) Win Prob: 51.6% Value Score: 2.31 ★ ML: 3.50 (5/2) Overlay %: -2.98 Fair Odds: 4.18 (3/1) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | C |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | C | F |
| Early Pace | A | B |
| Mid Pace | A | C |
| Late Pace | C | D |
| Finish Pos | A | F |
| Dirt Speed | B | D |
| Turf Speed | A | C |
| Distance Speed | B | C |
| Trainer Effectiveness | B | C |
| Jockey Effectiveness | D | F |
| 3rd Party Ratings | C | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | C | D |
| Early Pace | B | B |
| Mid Pace | B | B |
| Late Pace | C | D |
| Sprint FPS | A | B |
| Route FPS | A | B |
| Trainer Effectiveness | B | C |
| Jockey Effectiveness | D | F |
| 3rd Party Ratings | C | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | C | F |
| Early Pace | A | B |
| Mid Pace | A | C |
| Late Pace | B | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 44-day layoff with a recent workout. (2) Hot jockey — high ITM rate (33%). (3) Late pace significantly above field median (LP: 71 vs field 67). (4) Had documented trouble last race (codes: ['D', 'L', 'O', 'S', 'T']) yet still finished 4th — a clean trip today sets up a form reversal.
⚠ Concerns:(1) Has not raced in 44 days. (2) Limited experience at today's distance (2 starts). (3) Unproven on dirt surface (2 starts). (4) Trip note from last race: 4-5p;stayed on mildly.
Mod1(W): 2 (4/4) Mod2(W): 25 (3/4) Mod3(W): 51 (4/4) Mod1-LS: 52 (3/4) Mod2-LS: 33 (3/4) Mod3-LS: 62 (4/4) Mod4(Form): 60 (3/4) Win Prob: 53.9% Value Score: 2.69 ★ ML: 4.00 (3/1) Overlay %: -0.85 Fair Odds: 4.21 (3/1) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | B | C |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | B | D |
| Early Pace | C | C |
| Mid Pace | D | C |
| Late Pace | A | C |
| Finish Pos | C | F |
| Dirt Speed | D | F |
| Turf Speed | D | D |
| Distance Speed | A | C |
| Trainer Effectiveness | C | D |
| Jockey Effectiveness | A | D |
| 3rd Party Ratings | B | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | D |
| Early Pace | C | C |
| Mid Pace | D | C |
| Late Pace | A | C |
| Sprint FPS | C | C |
| Route FPS | B | D |
| Trainer Effectiveness | C | D |
| Jockey Effectiveness | A | D |
| 3rd Party Ratings | B | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | A |
| Early Pace | C | B |
| Mid Pace | B | C |
| Late Pace | B | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 36-day layoff with a recent workout. (2) Hot trainer - high ITM rate (40%). (3) Big last-race effort (74) — 13 pts above recent average. (4) Closer/presser in a race with pace pressure — ideal race setup for late runners. (5) Has won at today's distance (1830 yds) within the last 10 starts — proven ability at this trip. (6) Returning from 36-day layoff with a bullet workout — trainer signaling extra readiness for today's spot. (7) Career-best figure (81) buried in races 4–10 back — 12 pts above race median and 6 pts above recent form — market has forgotten the peak effort. (8) Had documented trouble last race (codes: ['D', 'F', 'L', 'O', 'S', 'T', 'U']) yet still finished 1st — a clean trip today sets up a form reversal.
⚠ Concerns:(1) Has not raced in 36 days. (2) Trip note from last race: 3-4p;led 3/16;shft out. (3) Speed figures are highly erratic over last 5 starts (σ = 11.5 pts: [74, 50, 75, 59, 75]) — unpredictable performer. (4) First start at BTP — never raced here before (last seen at: TP).
Mod1(W): 88 (2/4) Mod2(W): 55 (3/4) Mod3(W): 9 (3/4) Mod1-LS: 66 (4/4) Mod2-LS: 67 (2/4) Mod3-LS: 0 Mod4(Form): 75 (2/4) Win Prob: 58.9% Value Score: 2.71 ★ ML: 3.50 (5/2) Overlay %: -2.27 Fair Odds: 4.13 (3/1) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | C | C |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | C | C |
| Early Pace | B | D |
| Mid Pace | A | C |
| Late Pace | C | C |
| Finish Pos | B | C |
| Dirt Speed | D | B |
| Turf Speed | A | A |
| Distance Speed | C | B |
| Trainer Effectiveness | A | C |
| Jockey Effectiveness | C | B |
| 3rd Party Ratings | C | C |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | C |
| Early Pace | A | D |
| Mid Pace | A | D |
| Late Pace | D | D |
| Sprint FPS | B | D |
| Route FPS | A | C |
| Trainer Effectiveness | A | C |
| Jockey Effectiveness | C | B |
| 3rd Party Ratings | C | C |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | C | C |
| Early Pace | B | D |
| Mid Pace | A | C |
| Late Pace | B | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 50-day layoff with a recent workout. (2) Finish positions improving over last 4 starts. (3) Big last-race effort (78) — 10 pts above recent average. (4) Hot jockey — high ITM rate (46%). (5) Late pace significantly above field median (LP: 86 vs field 74). (6) Closer/presser in a race with pace pressure — ideal race setup for late runners. (7) Has won at today's distance (1830 yds) within the last 10 starts — proven ability at this trip. (8) Has won at BTP on dirt in the last 10 starts — proven at today's exact track/surface combo. (9) Had documented trouble last race (codes: ['D', 'L', 'S']) yet still finished 2nd — a clean trip today sets up a form reversal.
⚠ Concerns:(1) Has not raced in 50 days. (2) Trip note from last race: 5-6p2nd;rallied;missed.
Mod1(W): 81 (3/4) Mod2(W): 100 (1/4) Mod3(W): 100 (1/4) Mod1-LS: 87 (2/4) Mod2-LS: 100 (1/4) Mod3-LS: 77 (2/4) Mod4(Form): 100 (1/4) Win Prob: 60.4% Value Score: 2.47 ★ ML: 3.00 (2/1) Overlay %: -1.39 Fair Odds: 3.33 (5/2) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | B |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | A | C |
| Early Pace | D | F |
| Mid Pace | C | D |
| Late Pace | A | A |
| Finish Pos | A | C |
| Dirt Speed | C | B |
| Turf Speed | A | B |
| Distance Speed | B | B |
| Trainer Effectiveness | C | D |
| Jockey Effectiveness | A | B |
| 3rd Party Ratings | B | C |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | B |
| Early Pace | C | F |
| Mid Pace | C | D |
| Late Pace | A | A |
| Sprint FPS | C | D |
| Route FPS | D | D |
| Trainer Effectiveness | C | D |
| Jockey Effectiveness | A | B |
| 3rd Party Ratings | B | C |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | C |
| Early Pace | D | D |
| Mid Pace | D | D |
| Late Pace | B | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 36-day layoff with a recent workout. (2) Early pace significantly above field median (E1: 80 vs field 76). (3) Late pace significantly above field median (LP: 77 vs field 74). (4) Has won at today's distance (1830 yds) within the last 10 starts — proven ability at this trip.
⚠ Concerns:(1) Has not raced in 36 days. (2) Consider there will be likely pace pressure in this race - multiple E types entered. (3) Trip note from last race: 2-3p;5p3/16;yield btw. (4) Has not won in last 10 starts despite career wins.
Mod1(W): 100 (1/4) Mod2(W): 55 (2/4) Mod3(W): 41 (2/4) Mod1-LS: 100 (1/4) Mod2-LS: 61 (3/4) Mod3-LS: 49 (3/4) Mod4(Form): 47 (3/4) Win Prob: 49.2% Value Score: 2.51 ★ ML: 4.00 (3/1) Overlay %: 1.60 Fair Odds: 3.72 (5/2) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | B |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | A | C |
| Early Pace | A | C |
| Mid Pace | A | C |
| Late Pace | A | B |
| Finish Pos | C | D |
| Dirt Speed | A | B |
| Turf Speed | D | D |
| Distance Speed | D | C |
| Trainer Effectiveness | B | D |
| Jockey Effectiveness | B | B |
| 3rd Party Ratings | A | C |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | C |
| Early Pace | A | C |
| Mid Pace | A | B |
| Late Pace | B | C |
| Sprint FPS | A | B |
| Route FPS | B | C |
| Trainer Effectiveness | B | D |
| Jockey Effectiveness | B | B |
| 3rd Party Ratings | A | C |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | C |
| Early Pace | A | C |
| Mid Pace | A | C |
| Late Pace | B | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 58-day layoff with a recent workout. (2) Speed average (72) well above field median (65). (3) Hot trainer - high ITM rate (56%). (4) Career-best speed figure (59) recorded in last 1-2 starts — peaking form. (5) Hot jockey — high ITM rate (55%). (6) Early pace significantly above field median (E1: 84 vs field 78). (7) Late pace significantly above field median (LP: 74 vs field 71). (8) Returning from 58-day layoff with a bullet workout — trainer signaling extra readiness for today's spot. (9) Career-best figure (84) buried in races 4–10 back — 19 pts above race median and 12 pts above recent form — market has forgotten the peak effort.
⚠ Concerns:(1) Has not raced in 58 days. (2) Was claimed from its most recent race - new connections. (3) Moving from a route to a sprint today. (4) Poor record at this distance (0 wins from 9 recent starts). (5) Tends to fade in the stretch at similar distances (6 of 9 starts). (6) Trip note from last race: Ins;4p btw 2nd;mildly. (7) Has not won in last 10 starts despite career wins. (8) Speed figures are highly erratic over last 5 starts (σ = 9.4 pts: [66, 66, 72, 59, 84]) — unpredictable performer. (9) 0 wins from 8 starts on dirt in last 10 — repeated failure on today's surface. (10) First start at BTP — never raced here before (last seen at: TP).
Mod1(W): 100 (1/4) Mod2(W): 100 (1/4) Mod3(W): 100 (1/4) Mod1-LS: 100 (1/4) Mod2-LS: 100 (1/4) Mod3-LS: 100 (1/4) Mod4(Form): 100 (1/4) Win Prob: 60.6% Value Score: 2.08 ★ ML: 2.50 (3/2) Overlay %: -5.62 Fair Odds: 3.25 (2/1) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | A |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | A | B |
| Early Pace | A | C |
| Mid Pace | A | B |
| Late Pace | A | B |
| Finish Pos | C | F |
| Dirt Speed | D | D |
| Turf Speed | A | A |
| Distance Speed | A | A |
| Trainer Effectiveness | A | A |
| Jockey Effectiveness | A | A |
| 3rd Party Ratings | A | B |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | C |
| Early Pace | A | B |
| Mid Pace | A | B |
| Late Pace | B | C |
| Sprint FPS | A | C |
| Route FPS | A | A |
| Trainer Effectiveness | A | A |
| Jockey Effectiveness | A | A |
| 3rd Party Ratings | A | B |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | C |
| Early Pace | A | B |
| Mid Pace | A | B |
| Late Pace | B | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 38-day layoff with a recent workout. (2) Hot trainer - high ITM rate (34%). (3) Hot jockey — high ITM rate (46%). (4) Had documented trouble last race (codes: ['B', 'L', 'T', 'U']) yet still finished 2nd — a clean trip today sets up a form reversal.
⚠ Concerns:(1) Has not raced in 38 days. (2) Trip note from last race: 5p;kept at bay;up plc. (3) 0 wins from 8 starts on dirt in last 10 — repeated failure on today's surface. (4) First start at BTP — never raced here before (last seen at: TP).
Mod1(W): 90 (2/4) Mod2(W): 76 (2/4) Mod3(W): 62 (3/4) Mod1-LS: 66 (3/4) Mod2-LS: 75 (2/4) Mod3-LS: 14 (4/4) Mod4(Form): 87 (2/4) Win Prob: 51.6% Value Score: 2.02 ★ ML: 3.00 (2/1) Overlay %: -4.36 Fair Odds: 3.73 (5/2) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | B | B |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | B | D |
| Early Pace | C | D |
| Mid Pace | C | D |
| Late Pace | C | C |
| Finish Pos | A | F |
| Dirt Speed | A | C |
| Turf Speed | A | A |
| Distance Speed | D | F |
| Trainer Effectiveness | A | D |
| Jockey Effectiveness | A | B |
| 3rd Party Ratings | A | C |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | C |
| Early Pace | B | C |
| Mid Pace | B | C |
| Late Pace | A | C |
| Sprint FPS | A | B |
| Route FPS | A | C |
| Trainer Effectiveness | A | D |
| Jockey Effectiveness | A | B |
| 3rd Party Ratings | A | C |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | D |
| Early Pace | D | D |
| Mid Pace | B | C |
| Late Pace | B | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 120-day layoff with a recent workout. (2) Dropping in class today ($10,000 → $5,000). (3) Dropping in class ($5,000 vs last $10,000) after a 120-day freshening — trainer setup-for-win angle. (4) Averages 62.6 speed pts in today's distance/surface configuration vs 56.8 in other configs (5 qualifying starts) — returning to ideal setup.
⚠ Concerns:(1) Has not raced in 120 days. (2) Poor recent form - recent speed ratings well below career norm. (3) Moving from a route to a sprint today. (4) 0 wins from 4 career starts on off tracks. (5) Poor record at this distance (0 wins from 5 recent starts). (6) Trip note from last race: Bump st;4-5p;done erly. (7) Speed figures are highly erratic over last 5 starts (σ = 10.5 pts: [39, 55, 56, 47, 67]) — unpredictable performer. (8) 0 wins from 10 starts on dirt in last 10 — repeated failure on today's surface. (9) First start at BTP — never raced here before (last seen at: TP, IND).
Mod1(W): 0 Mod2(W): 0 Mod3(W): 0 Mod1-LS: 0 Mod2-LS: 0 Mod3-LS: 25 (3/4) Mod4(Form): 1 (4/4) Win Prob: 28.7% Value Score: 2.54 ★ ML: 8.00 (7/1) Overlay %: 4.38 Fair Odds: 5.36 (9/2) O/U: Overlay
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | D | C |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | D | D |
| Early Pace | B | D |
| Mid Pace | B | D |
| Late Pace | D | D |
| Finish Pos | B | F |
| Dirt Speed | C | C |
| Turf Speed | C | B |
| Distance Speed | C | C |
| Trainer Effectiveness | B | F |
| Jockey Effectiveness | C | F |
| 3rd Party Ratings | D | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | D | F |
| Early Pace | D | D |
| Mid Pace | D | D |
| Late Pace | D | D |
| Sprint FPS | C | C |
| Route FPS | B | C |
| Trainer Effectiveness | B | F |
| Jockey Effectiveness | C | F |
| 3rd Party Ratings | D | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | D | D |
| Early Pace | C | D |
| Mid Pace | C | C |
| Late Pace | B | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Hot trainer - high ITM rate (38%). (2) Career-best speed figure (59) recorded in last 1-2 starts — peaking form. (3) Big last-race effort (63) — 10 pts above recent average. (4) Hot jockey — high ITM rate (41%). (5) Late pace significantly above field median (LP: 69 vs field 64). (6) Trainer 'WALSH RYAN D' has 2 horses in today's card all showing last-race improvement — barn appears sharp right now. (7) Career-best figure (71) buried in races 4–10 back — 14 pts above race median and 8 pts above recent form — market has forgotten the peak effort.
⚠ Concerns:(1) Has not raced in 38 days. (2) Limited experience at today's distance (1 start). (3) Moving from a route to a sprint today. (4) Trip note from last race: 2wd btwn; no threat. (5) Consistently runs wide at the first call (3 of last 5 starts). (6) No workout in last 30 days despite 38 days since last race. (7) Speed figures are highly erratic over last 5 starts (σ = 9.2 pts: [63, 62, 56, 42, 64]) — unpredictable performer. (8) First start at BTP — never raced here before (last seen at: TP).
Mod1(W): 68 (2/5) Mod2(W): 90 (2/5) Mod3(W): 77 (2/5) Mod1-LS: 100 (1/5) Mod2-LS: 88 (2/5) Mod3-LS: 46 (3/5) Mod4(Form): 100 (1/5) Win Prob: 69.8% Value Score: 3.16 ★ ML: 3.00 (2/1) Overlay %: -2.73 Fair Odds: 4.17 (3/1) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | B |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | A | D |
| Early Pace | D | D |
| Mid Pace | B | D |
| Late Pace | A | C |
| Finish Pos | A | D |
| Dirt Speed | D | D |
| Turf Speed | A | C |
| Distance Speed | A | C |
| Trainer Effectiveness | A | C |
| Jockey Effectiveness | A | C |
| 3rd Party Ratings | A | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | D |
| Early Pace | D | D |
| Mid Pace | C | D |
| Late Pace | A | C |
| Sprint FPS | D | D |
| Route FPS | A | D |
| Trainer Effectiveness | A | C |
| Jockey Effectiveness | A | C |
| 3rd Party Ratings | A | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | D |
| Early Pace | C | D |
| Mid Pace | B | D |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Finish positions improving over last 4 starts. (2) Big last-race effort (75) — 22 pts above recent average. (3) Early pace significantly above field median (E1: 81 vs field 78). (4) Late pace significantly above field median (LP: 68 vs field 64). (5) Has won at today's distance (1320 yds) within the last 10 starts — proven ability at this trip. (6) Outstanding wet-track record: 1W from 4 starts (25%) — thrives when the going is off. (7) Carries 5 lbs less than field median (118 lbs vs median 123 lbs). (8) Had documented trouble last race (codes: ['B', 'D', 'F', 'S', 'T', 'U']) yet still finished 1st — a clean trip today sets up a form reversal. (9) Averages 61.7 speed pts in today's distance/surface configuration vs 52.7 in other configs (7 qualifying starts) — returning to ideal setup.
⚠ Concerns:(1) Trip note from last race: Caused traffic break. (2) Last race was career-best speed (75) — bounce risk after peak effort. (3) No workout in last 30 days despite 19 days since last race. (4) Speed figures are highly erratic over last 5 starts (σ = 9.8 pts: [75, 53, 55, 50, 57]) — unpredictable performer.
Mod1(W): 100 (1/5) Mod2(W): 100 (1/5) Mod3(W): 100 (1/5) Mod1-LS: 87 (2/5) Mod2-LS: 100 (1/5) Mod3-LS: 27 (4/5) Mod4(Form): 72 (3/5) Win Prob: 45.3% Value Score: 2.56 ★ ML: 4.00 (3/1) Overlay %: 2.30 Fair Odds: 4.01 (3/1) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | B |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | A | D |
| Early Pace | A | C |
| Mid Pace | A | C |
| Late Pace | A | D |
| Finish Pos | B | D |
| Dirt Speed | A | C |
| Turf Speed | B | C |
| Distance Speed | A | B |
| Trainer Effectiveness | B | F |
| Jockey Effectiveness | F | F |
| 3rd Party Ratings | A | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | D |
| Early Pace | A | B |
| Mid Pace | A | B |
| Late Pace | D | D |
| Sprint FPS | A | B |
| Route FPS | B | D |
| Trainer Effectiveness | B | F |
| Jockey Effectiveness | F | F |
| 3rd Party Ratings | A | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | D |
| Early Pace | A | C |
| Mid Pace | A | C |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 153-day layoff with a recent workout. (2) 3 workouts in last 30 days — horse appears sharp and fit. (3) Has a bullet work (fastest at track that day) — sharp conditioning signal. (4) Strong speed 2 back (52), modest dip last out (46) — form reversal angle, ready to bounce back to top form. (5) Has won at today's distance (1320 yds) within the last 10 starts — proven ability at this trip. (6) Returning from 153-day layoff with a bullet workout — trainer signaling extra readiness for today's spot. (7) Averages 46.0 speed pts in today's distance/surface configuration vs 37.5 in other configs (8 qualifying starts) — returning to ideal setup.
⚠ Concerns:(1) Has not raced in 153 days. (2) Repeatedly beaten at this class level (3 of last 5 starts). (3) Speed average (44) well below field median (57). (4) Poor record at this track (0 wins from 7 starts). (5) Trip note from last race: Played no factor. (6) Dropping in class in consecutive starts ($10,000 → $5,000 → $4,000). (7) Speed figures are highly erratic over last 5 starts (σ = 11.9 pts: [46, 52, 50, 60, 28]) — unpredictable performer.
Mod1(W): 48 (4/5) Mod2(W): 25 (4/5) Mod3(W): 31 (4/5) Mod1-LS: 4 (5/5) Mod2-LS: 25 (4/5) Mod3-LS: 100 (1/5) Mod4(Form): 57 (4/5) Win Prob: 50.2% Value Score: 3.13 ★ ML: 4.50 (7/2) Overlay %: -0.86 Fair Odds: 5.58 (9/2) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | C | C |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | D | F |
| Early Pace | A | C |
| Mid Pace | D | D |
| Late Pace | D | D |
| Finish Pos | A | D |
| Dirt Speed | C | D |
| Turf Speed | B | C |
| Distance Speed | D | D |
| Trainer Effectiveness | C | F |
| Jockey Effectiveness | C | D |
| 3rd Party Ratings | D | F |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | D | F |
| Early Pace | F | D |
| Mid Pace | D | D |
| Late Pace | A | C |
| Sprint FPS | B | B |
| Route FPS | D | D |
| Trainer Effectiveness | C | F |
| Jockey Effectiveness | C | D |
| 3rd Party Ratings | D | F |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | D | F |
| Early Pace | A | C |
| Mid Pace | F | D |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 65-day layoff with a recent workout. (2) Big last-race effort (56) — 9 pts above recent average. (3) One gate workout in recent training — sharpness signal from the gate. (4) Has won at today's distance (1320 yds) within the last 10 starts — proven ability at this trip. (5) Returning from 65-day layoff with a bullet workout — trainer signaling extra readiness for today's spot. (6) Has won at BTP on dirt in the last 10 starts — proven at today's exact track/surface combo. (7) Had documented trouble last race (codes: ['D', 'F', 'L', 'S', 'T', 'V']) yet still finished 4th — a clean trip today sets up a form reversal. (8) Averages 61.0 speed pts in today's distance/surface configuration vs 16.5 in other configs (3 qualifying starts) — returning to ideal setup.
⚠ Concerns:(1) Has not raced in 65 days. (2) Speed average (43) well below field median (57). (3) Trip note from last race: 2p;vie shw;left ld;mss. (4) Speed figures are highly erratic over last 5 starts (σ = 24.9 pts: [56, 58, 69, 15, 18]) — unpredictable performer.
Mod1(W): 61 (3/5) Mod2(W): 45 (3/5) Mod3(W): 56 (3/5) Mod1-LS: 64 (4/5) Mod2-LS: 46 (3/5) Mod3-LS: 19 (5/5) Mod4(Form): 84 (2/5) Win Prob: 59.4% Value Score: 3.03 ★ ML: 3.50 (5/2) Overlay %: -2.02 Fair Odds: 4.68 (7/2) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | B | B |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | F | F |
| Early Pace | B | D |
| Mid Pace | A | D |
| Late Pace | F | D |
| Finish Pos | F | F |
| Dirt Speed | A | C |
| Turf Speed | D | C |
| Distance Speed | B | C |
| Trainer Effectiveness | F | F |
| Jockey Effectiveness | B | C |
| 3rd Party Ratings | C | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | D |
| Early Pace | A | B |
| Mid Pace | A | B |
| Late Pace | C | D |
| Sprint FPS | A | B |
| Route FPS | A | C |
| Trainer Effectiveness | F | F |
| Jockey Effectiveness | B | C |
| 3rd Party Ratings | C | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | F | F |
| Early Pace | B | C |
| Mid Pace | A | D |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 218-day layoff with a recent workout. (2) In the money in last 3 consecutive starts. (3) Track specialist — 4 wins from 15 starts here (27%). (4) Strong speed 2 back (71), modest dip last out (62) — form reversal angle, ready to bounce back to top form. (5) Has won at today's distance (1320 yds) within the last 10 starts — proven ability at this trip. (6) Has won at BTP on dirt in the last 10 starts — proven at today's exact track/surface combo. (7) Had documented trouble last race (codes: ['D', 'L']) yet still finished 3rd — a clean trip today sets up a form reversal.
⚠ Concerns:(1) Has not raced in 218 days. (2) Consider there will be likely pace pressure in this race - multiple E types entered. (3) Trip note from last race: In range; lacked rally.
Mod1(W): 17 (6/6) Mod2(W): 8 (6/6) Mod3(W): 13 (6/6) Mod1-LS: 20 (6/6) Mod2-LS: 10 (6/6) Mod3-LS: 0 Mod4(Form): 48 (5/6) Win Prob: 60.5% Value Score: 4.84 ★ ML: 4.00 (3/1) Overlay %: 0.01 Fair Odds: 7.00 (6/1) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | F | D |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | D | D |
| Early Pace | D | D |
| Mid Pace | C | D |
| Late Pace | C | B |
| Finish Pos | D | C |
| Dirt Speed | D | B |
| Turf Speed | C | C |
| Distance Speed | C | B |
| Trainer Effectiveness | C | D |
| Jockey Effectiveness | F | D |
| 3rd Party Ratings | D | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | D | C |
| Early Pace | C | C |
| Mid Pace | C | C |
| Late Pace | B | B |
| Sprint FPS | B | A |
| Route FPS | A | B |
| Trainer Effectiveness | C | D |
| Jockey Effectiveness | F | D |
| 3rd Party Ratings | D | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | D | D |
| Early Pace | C | C |
| Mid Pace | C | C |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 287-day layoff with a recent workout. (2) Hot jockey — high ITM rate (46%). (3) In the money in last 5 consecutive starts. (4) Track specialist — 2 wins from 6 starts here (33%). (5) Late pace significantly above field median (LP: 81 vs field 78). (6) Closer/presser in a race with pace pressure — ideal race setup for late runners. (7) Has won at today's distance (1320 yds) within the last 10 starts — proven ability at this trip. (8) Has won at BTP on dirt in the last 10 starts — proven at today's exact track/surface combo. (9) Had documented trouble last race (codes: ['D', 'F', 'L', 'O']) yet still finished 2nd — a clean trip today sets up a form reversal.
⚠ Concerns:(1) Has not raced in 287 days. (2) Moving from a route to a sprint today. (3) Trip note from last race: 2w; rallied for place.
Mod1(W): 24 (5/6) Mod2(W): 30 (4/6) Mod3(W): 15 (5/6) Mod1-LS: 25 (4/6) Mod2-LS: 18 (5/6) Mod3-LS: 49 (3/6) Mod4(Form): 74 (3/6) Win Prob: 52.8% Value Score: 3.38 ★ ML: 3.00 (2/1) Overlay %: -2.50 Fair Odds: 6.62 (6/1) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | C | D |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | C | C |
| Early Pace | C | C |
| Mid Pace | D | D |
| Late Pace | A | A |
| Finish Pos | A | B |
| Dirt Speed | C | B |
| Turf Speed | B | B |
| Distance Speed | D | C |
| Trainer Effectiveness | C | B |
| Jockey Effectiveness | A | B |
| 3rd Party Ratings | C | C |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | B |
| Early Pace | F | D |
| Mid Pace | F | D |
| Late Pace | A | A |
| Sprint FPS | F | C |
| Route FPS | C | B |
| Trainer Effectiveness | C | B |
| Jockey Effectiveness | A | B |
| 3rd Party Ratings | C | C |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | C | C |
| Early Pace | D | C |
| Mid Pace | D | D |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 129-day layoff with a recent workout. (2) Speed average (77) well above field median (68). (3) Hot trainer - high ITM rate (56%). (4) Track specialist — 3 wins from 5 starts here (60%). (5) Early pace significantly above field median (E1: 93 vs field 81). (6) Has won at today's distance (1320 yds) within the last 10 starts — proven ability at this trip. (7) Has won at BTP on dirt in the last 10 starts — proven at today's exact track/surface combo. (8) Career-best figure (84) buried in races 4–10 back — 16 pts above race median and 6 pts above recent form — market has forgotten the peak effort.
⚠ Concerns:(1) Has not raced in 129 days. (2) Consider there will be likely pace pressure in this race - multiple E types entered. (3) 0 wins from 5 career starts on off tracks. (4) Trip note from last race: Outside;3p;no bid.
Mod1(W): 25 (4/6) Mod2(W): 35 (3/6) Mod3(W): 32 (3/6) Mod1-LS: 45 (3/6) Mod2-LS: 43 (3/6) Mod3-LS: 37 (5/6) Mod4(Form): 20 (6/6) Win Prob: 46.8% Value Score: 4.49 ★ ML: 5.00 (4/1) Overlay %: 3.91 Fair Odds: 5.98 (5/1) O/U: Overlay
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | C | C |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | A | B |
| Early Pace | A | A |
| Mid Pace | A | A |
| Late Pace | C | B |
| Finish Pos | B | B |
| Dirt Speed | B | B |
| Turf Speed | A | A |
| Distance Speed | B | A |
| Trainer Effectiveness | A | A |
| Jockey Effectiveness | C | B |
| 3rd Party Ratings | A | B |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | B |
| Early Pace | A | A |
| Mid Pace | B | B |
| Late Pace | C | B |
| Sprint FPS | C | B |
| Route FPS | D | C |
| Trainer Effectiveness | A | A |
| Jockey Effectiveness | C | B |
| 3rd Party Ratings | A | B |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | B |
| Early Pace | A | A |
| Mid Pace | B | A |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 150-day layoff with a recent workout. (2) Speed average (77) well above field median (68). (3) Hot trainer - high ITM rate (56%). (4) Late pace significantly above field median (LP: 81 vs field 78). (5) Closer/presser in a race with pace pressure — ideal race setup for late runners. (6) Outstanding wet-track record: 5W from 15 starts (33%) — thrives when the going is off. (7) Career-best figure (83) buried in races 4–10 back — 15 pts above race median and 8 pts above recent form — market has forgotten the peak effort. (8) Had documented trouble last race (codes: ['D', 'L', 'O']) yet still finished 1st — a clean trip today sets up a form reversal.
⚠ Concerns:(1) Has not raced in 150 days. (2) Moving from a route to a sprint today. (3) Poor record at this distance (0 wins from 4 recent starts). (4) Trip note from last race: Loom3p1/4;edged clear.
Mod1(W): 44 (2/6) Mod2(W): 47 (2/6) Mod3(W): 47 (2/6) Mod1-LS: 53 (2/6) Mod2-LS: 54 (2/6) Mod3-LS: 42 (4/6) Mod4(Form): 86 (2/6) Win Prob: 64.1% Value Score: 3.59 ★ ML: 2.50 (3/2) Overlay %: -2.50 Fair Odds: 5.51 (9/2) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | C |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | B | B |
| Early Pace | C | C |
| Mid Pace | C | C |
| Late Pace | A | A |
| Finish Pos | C | B |
| Dirt Speed | A | A |
| Turf Speed | A | B |
| Distance Speed | A | A |
| Trainer Effectiveness | A | A |
| Jockey Effectiveness | C | B |
| 3rd Party Ratings | B | B |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | B |
| Early Pace | C | C |
| Mid Pace | C | C |
| Late Pace | C | B |
| Sprint FPS | D | C |
| Route FPS | C | B |
| Trainer Effectiveness | A | A |
| Jockey Effectiveness | C | B |
| 3rd Party Ratings | B | B |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | B |
| Early Pace | C | C |
| Mid Pace | C | B |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 159-day layoff with a recent workout. (2) Speed average (86) well above field median (68). (3) Hot trainer - high ITM rate (49%). (4) Big last-race effort (99) — 9 pts above recent average. (5) Hot jockey — high ITM rate (45%). (6) Won last 3 consecutive starts. (7) 3 workouts in last 30 days — horse appears sharp and fit. (8) Early pace significantly above field median (E1: 94 vs field 81). (9) Has won at today's distance (1320 yds) within the last 10 starts — proven ability at this trip. (10) Had documented trouble last race (codes: ['B', 'D', 'L', 'O', 'S']) yet still finished 1st — a clean trip today sets up a form reversal. (11) Averages 90.4 speed pts in today's distance/surface configuration vs 81.4 in other configs (5 qualifying starts) — returning to ideal setup.
⚠ Concerns:(1) Has not raced in 159 days. (2) Consider there will be likely pace pressure in this race - multiple E types entered. (3) Trip note from last race: Broke in;2-3p;hld sway. (4) Last race was career-best speed (99) — bounce risk after peak effort.
Mod1(W): 100 (1/6) Mod2(W): 100 (1/6) Mod3(W): 100 (1/6) Mod1-LS: 100 (1/6) Mod2-LS: 100 (1/6) Mod3-LS: 100 (1/6) Mod4(Form): 100 (1/6) Win Prob: 45.7% Value Score: 2.19 ★ ML: 2.00 (1/1) Overlay %: -2.68 Fair Odds: 4.51 (7/2) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | B |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | A | A |
| Early Pace | A | A |
| Mid Pace | A | A |
| Late Pace | B | B |
| Finish Pos | A | B |
| Dirt Speed | A | A |
| Turf Speed | C | C |
| Distance Speed | A | A |
| Trainer Effectiveness | B | A |
| Jockey Effectiveness | B | B |
| 3rd Party Ratings | A | A |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | A |
| Early Pace | A | A |
| Mid Pace | A | A |
| Late Pace | A | A |
| Sprint FPS | A | A |
| Route FPS | A | B |
| Trainer Effectiveness | B | A |
| Jockey Effectiveness | B | B |
| 3rd Party Ratings | A | A |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | A |
| Early Pace | A | A |
| Mid Pace | A | A |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Dropping in class today ($7,500 → $5,000). (2) Finish positions improving over last 4 starts. (3) Big last-race effort (66) — 16 pts above recent average. (4) Closer/presser in a race with pace pressure — ideal race setup for late runners. (5) Dropping in class ($5,000 vs last $7,500) after a 49-day freshening — trainer setup-for-win angle. (6) Returning from 49-day layoff with a bullet workout — trainer signaling extra readiness for today's spot. (7) Had documented trouble last race (codes: ['B', 'D', 'F', 'L', 'S', 'T']) yet still finished 4th — a clean trip today sets up a form reversal. (8) Averages 55.0 speed pts in today's distance/surface configuration vs 43.5 in other configs (8 qualifying starts) — returning to ideal setup.
⚠ Concerns:(1) Has not raced in 49 days. (2) Poor record at this distance (0 wins from 8 recent starts). (3) Trip note from last race: 5p;stayed btw;left ld. (4) No workout in last 30 days despite 49 days since last race. (5) 0 wins from 10 starts on dirt in last 10 — repeated failure on today's surface.
Mod1(W): 100 (1/7) Mod2(W): 100 (1/7) Mod3(W): 100 (1/7) Mod1-LS: 100 (1/7) Mod2-LS: 100 (1/7) Mod3-LS: 100 (1/7) Mod4(Form): 92 (2/7) Win Prob: 55% Value Score: 3.21 ★ ML: 4.00 (3/1) Overlay %: -2.27 Fair Odds: 5.73 (9/2) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | A |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | A | F |
| Early Pace | B | C |
| Mid Pace | A | C |
| Late Pace | B | D |
| Finish Pos | A | D |
| Dirt Speed | A | C |
| Turf Speed | B | C |
| Distance Speed | A | C |
| Trainer Effectiveness | C | D |
| Jockey Effectiveness | D | D |
| 3rd Party Ratings | A | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | F |
| Early Pace | A | B |
| Mid Pace | A | B |
| Late Pace | B | D |
| Sprint FPS | C | C |
| Route FPS | A | C |
| Trainer Effectiveness | C | D |
| Jockey Effectiveness | D | D |
| 3rd Party Ratings | A | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | F |
| Early Pace | B | C |
| Mid Pace | B | C |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 37-day layoff with a recent workout. (2) Dropping in class today ($7,500 → $5,000). (3) Hot trainer - high ITM rate (44%). (4) Career-best speed figure (36) recorded in last 1-2 starts — peaking form. (5) Has a bullet work (fastest at track that day) — sharp conditioning signal. (6) Early pace significantly above field median (E1: 88 vs field 82). (7) Dropping in class ($5,000 vs last $7,500) after a 37-day freshening — trainer setup-for-win angle. (8) Returning from 37-day layoff with a bullet workout — trainer signaling extra readiness for today's spot.
⚠ Concerns:(1) Has not raced in 37 days. (2) Consider there will be likely pace pressure in this race - multiple E types entered. (3) Speed average (38) well below field median (50). (4) Poor record at this distance (0 wins from 8 recent starts). (5) Trip note from last race: Trck 2-3w;empty lane. (6) Speed figures are highly erratic over last 5 starts (σ = 10.2 pts: [36, 28, 54, 43, 32]) — unpredictable performer. (7) 0 wins from 7 starts on dirt in last 10 — repeated failure on today's surface. (8) First start at BTP — never raced here before (last seen at: TP).
Mod1(W): 40 (6/7) Mod2(W): 34 (6/7) Mod3(W): 41 (6/7) Mod1-LS: 96 (2/7) Mod2-LS: 40 (6/7) Mod3-LS: 45 (7/7) Mod4(Form): 33 (6/7) Win Prob: 26.3% Value Score: 3.38 ★ ML: 10.00 (9/1) Overlay %: 4.05 Fair Odds: 7.45 (6/1) O/U: Overlay
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | D | B |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | D | F |
| Early Pace | A | A |
| Mid Pace | A | C |
| Late Pace | F | F |
| Finish Pos | F | F |
| Dirt Speed | C | D |
| Turf Speed | A | C |
| Distance Speed | C | D |
| Trainer Effectiveness | A | B |
| Jockey Effectiveness | F | F |
| 3rd Party Ratings | B | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | D | F |
| Early Pace | A | A |
| Mid Pace | A | C |
| Late Pace | F | F |
| Sprint FPS | A | A |
| Route FPS | A | B |
| Trainer Effectiveness | A | B |
| Jockey Effectiveness | F | F |
| 3rd Party Ratings | B | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | D | F |
| Early Pace | A | B |
| Mid Pace | A | B |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Hot jockey — high ITM rate (45%). (2) Has a bullet work (fastest at track that day) — sharp conditioning signal. (3) Returning from 58-day layoff with a bullet workout — trainer signaling extra readiness for today's spot. (4) Career-best figure (59) buried in races 4–10 back — 9 pts above race median and 10 pts above recent form — market has forgotten the peak effort.
⚠ Concerns:(1) Has not raced in 58 days. (2) Consider there will be likely pace pressure in this race - multiple E types entered. (3) Poor record at this distance (0 wins from 5 recent starts). (4) Trip note from last race: Brshd btw st;ins;outrn. (5) 0 wins from post 4 in 3 recent starts. (6) 0 wins from 10 starts on dirt in last 10 — repeated failure on today's surface. (7) First start at BTP — never raced here before (last seen at: TP, CT).
Mod1(W): 66 (5/7) Mod2(W): 71 (4/7) Mod3(W): 70 (5/7) Mod1-LS: 60 (5/7) Mod2-LS: 77 (4/7) Mod3-LS: 52 (6/7) Mod4(Form): 74 (4/7) Win Prob: 42.2% Value Score: 2.96 ★ ML: 5.00 (4/1) Overlay %: -0.75 Fair Odds: 6.39 (5/1) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | C | B |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | B | F |
| Early Pace | D | D |
| Mid Pace | D | D |
| Late Pace | C | D |
| Finish Pos | A | F |
| Dirt Speed | A | D |
| Turf Speed | D | D |
| Distance Speed | B | D |
| Trainer Effectiveness | B | C |
| Jockey Effectiveness | A | B |
| 3rd Party Ratings | D | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | F |
| Early Pace | D | D |
| Mid Pace | D | D |
| Late Pace | C | D |
| Sprint FPS | D | D |
| Route FPS | D | D |
| Trainer Effectiveness | B | C |
| Jockey Effectiveness | A | B |
| 3rd Party Ratings | D | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | F |
| Early Pace | C | D |
| Mid Pace | D | D |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 35-day layoff with a recent workout. (2) Dropping in class today ($7,500 → $5,000). (3) Hot trainer - high ITM rate (38%). (4) Big last-race effort (58) — 9 pts above recent average. (5) Late pace significantly above field median (LP: 65 vs field 60). (6) Dropping in class ($5,000 vs last $7,500) after a 35-day freshening — trainer setup-for-win angle.
⚠ Concerns:(1) Has not raced in 35 days. (2) Limited experience at today's distance (2 starts). (3) Trip note from last race: Vied ins pair;fade5/16. (4) Dropping in class in consecutive starts ($30,000 → $15,000 → $7,500). (5) First start at BTP — never raced here before (last seen at: TP).
Mod1(W): 87 (2/7) Mod2(W): 87 (2/7) Mod3(W): 86 (2/7) Mod1-LS: 90 (3/7) Mod2-LS: 88 (3/7) Mod3-LS: 55 (5/7) Mod4(Form): 100 (1/7) Win Prob: 51.9% Value Score: 2.73 ★ ML: 3.50 (5/2) Overlay %: -4.25 Fair Odds: 5.76 (5/1) O/U: Underlay
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | A |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | A | F |
| Early Pace | C | C |
| Mid Pace | B | D |
| Late Pace | A | D |
| Finish Pos | D | F |
| Dirt Speed | F | F |
| Turf Speed | D | D |
| Distance Speed | F | F |
| Trainer Effectiveness | A | C |
| Jockey Effectiveness | C | D |
| 3rd Party Ratings | A | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | F |
| Early Pace | C | C |
| Mid Pace | B | D |
| Late Pace | A | D |
| Sprint FPS | B | B |
| Route FPS | B | D |
| Trainer Effectiveness | A | C |
| Jockey Effectiveness | C | D |
| 3rd Party Ratings | A | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | C | F |
| Early Pace | B | C |
| Mid Pace | B | D |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Has a bullet work (fastest at track that day) — sharp conditioning signal. (2) One gate workout in recent training — sharpness signal from the gate.
⚠ Concerns:(1) Consider there will be likely pace pressure in this race - multiple E types entered. (2) Speed average (20) well below field median (50). (3) Trip note from last race: Folded abruptly turn.
Mod1(W): 7 (7/7) Mod2(W): 12 (7/7) Mod3(W): 12 (7/7) Mod1-LS: 0 Mod2-LS: 17 (7/7) Mod3-LS: 57 (4/7) Mod4(Form): 17 (7/7) Win Prob: 20.1% Value Score: 3.05 ★ ML: 12.00 (10/1) Overlay %: 3.08 Fair Odds: 9.34 (8/1) O/U: Overlay
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | D | D |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | F | F |
| Early Pace | A | B |
| Mid Pace | D | D |
| Late Pace | D | F |
| Finish Pos | D | F |
| Dirt Speed | C | D |
| Turf Speed | D | D |
| Distance Speed | C | D |
| Trainer Effectiveness | D | F |
| Jockey Effectiveness | B | D |
| 3rd Party Ratings | D | F |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | F | F |
| Early Pace | B | B |
| Mid Pace | D | D |
| Late Pace | D | F |
| Sprint FPS | A | B |
| Route FPS | B | D |
| Trainer Effectiveness | D | F |
| Jockey Effectiveness | B | D |
| 3rd Party Ratings | D | F |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | F | F |
| Early Pace | F | D |
| Mid Pace | F | D |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Dropping in class today ($7,500 → $5,000). (2) Hot jockey — high ITM rate (46%). (3) Dropping in class ($5,000 vs last $7,500) after a 50-day freshening — trainer setup-for-win angle.
⚠ Concerns:(1) Has not raced in 50 days. (2) Poor record at this distance (0 wins from 4 recent starts). (3) Trip note from last race: Trailed 4p; outrun. (4) Post 7 in a 8-horse dirt sprint — wide draw forces extra ground and a difficult first-turn path. (5) First start at BTP — never raced here before (last seen at: TP, MVR).
Mod1(W): 68 (4/7) Mod2(W): 65 (5/7) Mod3(W): 73 (4/7) Mod1-LS: 42 (6/7) Mod2-LS: 74 (5/7) Mod3-LS: 65 (3/7) Mod4(Form): 72 (5/7) Win Prob: 39.1% Value Score: 2.74 ★ ML: 5.00 (4/1) Overlay %: -0.96 Fair Odds: 6.51 (6/1) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | B | A |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | C | F |
| Early Pace | B | C |
| Mid Pace | C | D |
| Late Pace | A | D |
| Finish Pos | C | F |
| Dirt Speed | B | D |
| Turf Speed | D | D |
| Distance Speed | B | C |
| Trainer Effectiveness | D | D |
| Jockey Effectiveness | A | B |
| 3rd Party Ratings | C | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | C | F |
| Early Pace | B | C |
| Mid Pace | B | D |
| Late Pace | A | D |
| Sprint FPS | D | C |
| Route FPS | D | D |
| Trainer Effectiveness | D | D |
| Jockey Effectiveness | A | B |
| 3rd Party Ratings | C | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | D |
| Early Pace | A | C |
| Mid Pace | C | D |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 50-day layoff with a recent workout. (2) Dropping in class today ($7,500 → $5,000). (3) Hot trainer - high ITM rate (38%). (4) Hot jockey — high ITM rate (42%). (5) Closer/presser in a race with pace pressure — ideal race setup for late runners. (6) Dropping in class ($5,000 vs last $7,500) after a 50-day freshening — trainer setup-for-win angle. (7) Returning from 50-day layoff with a bullet workout — trainer signaling extra readiness for today's spot. (8) Career-best figure (63) buried in races 4–10 back — 13 pts above race median and 6 pts above recent form — market has forgotten the peak effort. (9) Averages 55.8 speed pts in today's distance/surface configuration vs 41.5 in other configs (6 qualifying starts) — returning to ideal setup.
⚠ Concerns:(1) Has not raced in 50 days. (2) Poor record at this distance (0 wins from 6 recent starts). (3) Trip note from last race: Saved ground; no avail. (4) Post 8 in a 8-horse dirt sprint — wide draw forces extra ground and a difficult first-turn path. (5) Speed figures are highly erratic over last 5 starts (σ = 8.1 pts: [44, 57, 44, 59, 60]) — unpredictable performer. (6) 0 wins from 9 starts on dirt in last 10 — repeated failure on today's surface. (7) First start at BTP — never raced here before (last seen at: TP).
Mod1(W): 85 (3/7) Mod2(W): 86 (3/7) Mod3(W): 82 (3/7) Mod1-LS: 78 (4/7) Mod2-LS: 91 (2/7) Mod3-LS: 70 (2/7) Mod4(Form): 81 (3/7) Win Prob: 49.6% Value Score: 3.18 ★ ML: 4.50 (7/2) Overlay %: -1.75 Fair Odds: 6.23 (5/1) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | B | A |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | B | F |
| Early Pace | D | C |
| Mid Pace | B | D |
| Late Pace | B | D |
| Finish Pos | B | F |
| Dirt Speed | F | F |
| Turf Speed | B | C |
| Distance Speed | A | C |
| Trainer Effectiveness | B | C |
| Jockey Effectiveness | B | B |
| 3rd Party Ratings | B | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | F |
| Early Pace | D | C |
| Mid Pace | C | D |
| Late Pace | B | D |
| Sprint FPS | B | C |
| Route FPS | C | D |
| Trainer Effectiveness | B | C |
| Jockey Effectiveness | B | B |
| 3rd Party Ratings | B | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | F |
| Early Pace | D | D |
| Mid Pace | A | B |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 192-day layoff with a recent workout. (2) Dropping in class today ($12,500 → $5,000). (3) Hot trainer - high ITM rate (56%). (4) Closer/presser in a race with pace pressure — ideal race setup for late runners. (5) Switching turf-to-dirt with a strong turf pedigree rating (94) — pedigree suggests main-track improvement is likely. (6) Dropping in class ($5,000 vs last $12,500) after a 192-day freshening — trainer setup-for-win angle. (7) Career-best figure (73) buried in races 4–10 back — 14 pts above race median and 7 pts above recent form — market has forgotten the peak effort. (8) Had documented trouble last race (codes: ['B', 'D', 'F', 'L', 'S', 'T']) yet still finished 4th — a clean trip today sets up a form reversal.
⚠ Concerns:(1) Has not raced in 192 days. (2) Consider there will be likely pace pressure in this race - multiple E types entered. (3) Limited experience at today's distance (1 start). (4) Moving from a route to a sprint today. (5) Switching from turf to dirt today. (6) Trip note from last race: Inside;bid1/4;faltered. (7) First start at BTP — never raced here before (last seen at: IND, CNL).
Mod1(W): 67 (4/7) Mod2(W): 71 (3/7) Mod3(W): 80 (3/7) Mod1-LS: 72 (3/7) Mod2-LS: 66 (3/7) Mod3-LS: 2 (7/7) Mod4(Form): 86 (3/7) Win Prob: 57.1% Value Score: 3.34 ★ ML: 3.00 (2/1) Overlay %: -4.60 Fair Odds: 7.01 (6/1) O/U: Underlay
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | B | B |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | A | D |
| Early Pace | D | D |
| Mid Pace | D | D |
| Late Pace | C | D |
| Finish Pos | D | F |
| Dirt Speed | D | F |
| Turf Speed | A | B |
| Distance Speed | F | F |
| Trainer Effectiveness | A | A |
| Jockey Effectiveness | B | B |
| 3rd Party Ratings | C | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | D |
| Early Pace | D | C |
| Mid Pace | D | D |
| Late Pace | D | D |
| Sprint FPS | D | D |
| Route FPS | C | D |
| Trainer Effectiveness | A | A |
| Jockey Effectiveness | B | B |
| 3rd Party Ratings | C | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | D |
| Early Pace | D | D |
| Mid Pace | D | D |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 52-day layoff with a recent workout. (2) Dropping in class today ($7,500 → $5,000). (3) Early pace significantly above field median (E1: 89 vs field 83). (4) Dropping in class ($5,000 vs last $7,500) after a 52-day freshening — trainer setup-for-win angle. (5) Returning from 52-day layoff with a bullet workout — trainer signaling extra readiness for today's spot. (6) Last race E1 pace (96.0) was 6.7 pts above personal avg (89.3) and horse still finished 4th — pace collapse victim; expect bounce if today's fractions are softer. (7) Had documented trouble last race (codes: ['D', 'S', 'V']) yet still finished 4th — a clean trip today sets up a form reversal. (8) Cutting back to sprint (1320 yds) from routes — sprint speed figures (51.5) are 5.7 pts better than route figures (45.8).
⚠ Concerns:(1) Has not raced in 52 days. (2) Consider there will be likely pace pressure in this race - multiple E types entered. (3) Moving from a route to a sprint today. (4) Poor record at this distance (0 wins from 5 recent starts). (5) Tends to fade in the stretch at similar distances (3 of 5 starts). (6) Trip note from last race: Ins;headd 1/4;gave way. (7) Last race was career-best speed (69) — bounce risk after peak effort. (8) Speed figures are highly erratic over last 5 starts (σ = 9.1 pts: [69, 71, 66, 64, 48]) — unpredictable performer. (9) 0 wins from 7 starts on dirt in last 10 — repeated failure on today's surface. (10) First start at BTP — never raced here before (last seen at: TP).
Mod1(W): 95 (2/7) Mod2(W): 86 (2/7) Mod3(W): 90 (2/7) Mod1-LS: 89 (2/7) Mod2-LS: 87 (2/7) Mod3-LS: 0 Mod4(Form): 93 (2/7) Win Prob: 63.3% Value Score: 3.24 ★ ML: 2.50 (3/2) Overlay %: -4.24 Fair Odds: 5.54 (9/2) O/U: Underlay
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | A |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | B | D |
| Early Pace | A | A |
| Mid Pace | A | B |
| Late Pace | D | D |
| Finish Pos | B | F |
| Dirt Speed | A | D |
| Turf Speed | D | D |
| Distance Speed | D | D |
| Trainer Effectiveness | F | D |
| Jockey Effectiveness | D | D |
| 3rd Party Ratings | A | C |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | C |
| Early Pace | A | A |
| Mid Pace | A | B |
| Late Pace | A | C |
| Sprint FPS | A | A |
| Route FPS | A | B |
| Trainer Effectiveness | F | D |
| Jockey Effectiveness | D | D |
| 3rd Party Ratings | A | C |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | D |
| Early Pace | A | A |
| Mid Pace | A | B |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 63-day layoff with a recent workout. (2) Dropping in class today ($7,500 → $5,000). (3) Speed average (68) well above field median (59). (4) Hot jockey — high ITM rate (49%). (5) Early pace significantly above field median (E1: 91 vs field 83). (6) Closer/presser in a race with pace pressure — ideal race setup for late runners. (7) Dropping in class ($5,000 vs last $7,500) after a 63-day freshening — trainer setup-for-win angle. (8) Career-best figure (79) buried in races 4–10 back — 20 pts above race median and 10 pts above recent form — market has forgotten the peak effort. (9) Had documented trouble last race (codes: ['B', 'D', 'L', 'S', 'T']) yet still finished 2nd — a clean trip today sets up a form reversal.
⚠ Concerns:(1) Has not raced in 63 days. (2) Poor record at this distance (0 wins from 10 recent starts). (3) Trip note from last race: Bpd btw st;3w;held plc. (4) 0 wins from 9 starts on dirt in last 10 — repeated failure on today's surface. (5) First start at BTP — never raced here before (last seen at: TP).
Mod1(W): 100 (1/7) Mod2(W): 100 (1/7) Mod3(W): 100 (1/7) Mod1-LS: 100 (1/7) Mod2-LS: 100 (1/7) Mod3-LS: 100 (1/7) Mod4(Form): 100 (1/7) Win Prob: 56.6% Value Score: 2.48 ★ ML: 2.00 (1/1) Overlay %: -8.44 Fair Odds: 5.97 (5/1) O/U: Underlay
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | A |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | A | C |
| Early Pace | A | A |
| Mid Pace | A | A |
| Late Pace | A | D |
| Finish Pos | C | F |
| Dirt Speed | A | C |
| Turf Speed | A | B |
| Distance Speed | A | B |
| Trainer Effectiveness | B | C |
| Jockey Effectiveness | A | A |
| 3rd Party Ratings | A | C |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | D |
| Early Pace | A | A |
| Mid Pace | A | A |
| Late Pace | B | D |
| Sprint FPS | B | B |
| Route FPS | A | B |
| Trainer Effectiveness | B | C |
| Jockey Effectiveness | A | A |
| 3rd Party Ratings | A | C |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | C |
| Early Pace | A | A |
| Mid Pace | A | A |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 36-day layoff with a recent workout. (2) Dropping in class today ($7,500 → $5,000). (3) Hot trainer - high ITM rate (56%). (4) Career-best speed figure (55) recorded in last 1-2 starts — peaking form. (5) Hot jockey — high ITM rate (45%). (6) Closer/presser in a race with pace pressure — ideal race setup for late runners. (7) Dropping in class ($5,000 vs last $7,500) after a 36-day freshening — trainer setup-for-win angle. (8) Cutting back to sprint (1320 yds) from routes — sprint speed figures (48.6) are 6.2 pts better than route figures (42.4).
⚠ Concerns:(1) Has not raced in 36 days. (2) Moving from a route to a sprint today. (3) Poor record at this distance (0 wins from 5 recent starts). (4) Trip note from last race: Stalk 3p;faded btw5/16. (5) Post 7 in a 8-horse dirt sprint — wide draw forces extra ground and a difficult first-turn path. (6) Speed figures are highly erratic over last 5 starts (σ = 9.5 pts: [55, 69, 42, 55, 55]) — unpredictable performer. (7) 0 wins from 7 starts on dirt in last 10 — repeated failure on today's surface. (8) First start at BTP — never raced here before (last seen at: TP, TAM).
Mod1(W): 62 (5/7) Mod2(W): 56 (5/7) Mod3(W): 77 (5/7) Mod1-LS: 50 (4/7) Mod2-LS: 53 (5/7) Mod3-LS: 52 (2/7) Mod4(Form): 47 (5/7) Win Prob: 39.5% Value Score: 4.04 ★ ML: 6.00 (5/1) Overlay %: 3.49 Fair Odds: 6.54 (6/1) O/U: Overlay
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | C | B |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | C | D |
| Early Pace | C | C |
| Mid Pace | B | C |
| Late Pace | B | D |
| Finish Pos | A | F |
| Dirt Speed | B | D |
| Turf Speed | B | C |
| Distance Speed | A | C |
| Trainer Effectiveness | A | A |
| Jockey Effectiveness | B | B |
| 3rd Party Ratings | B | C |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | D | F |
| Early Pace | B | B |
| Mid Pace | B | C |
| Late Pace | C | D |
| Sprint FPS | A | B |
| Route FPS | B | C |
| Trainer Effectiveness | A | A |
| Jockey Effectiveness | B | B |
| 3rd Party Ratings | B | C |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | C | D |
| Early Pace | C | C |
| Mid Pace | B | C |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Hot trainer - high ITM rate (49%). (2) Hot jockey — high ITM rate (51%). (3) Early pace significantly above field median (E1: 89 vs field 84). (4) Last race E1 pace (95.0) was 6.0 pts above personal avg (89.0) and horse still finished 6th — pace collapse victim; expect bounce if today's fractions are softer.
⚠ Concerns:(1) Has not raced in 45 days. (2) Consider there will be likely pace pressure in this race - multiple E types entered. (3) Poor record at this distance (0 wins from 5 recent starts). (4) Trip note from last race: Duel;svd gnd;wkn. (5) 0 wins from post 1 in 3 recent starts. (6) 0 wins from 5 starts on dirt in last 10 — repeated failure on today's surface. (7) First start at BTP — never raced here before (last seen at: TAM).
Mod1(W): 84 (3/7) Mod2(W): 100 (1/7) Mod3(W): 100 (2/7) Mod1-LS: 53 (3/7) Mod2-LS: 100 (1/7) Mod3-LS: 100 (1/7) Mod4(Form): 100 (1/7) Win Prob: 61.1% Value Score: 2.68 ★ ML: 2.00 (1/1) Overlay %: -7.52 Fair Odds: 5.54 (9/2) O/U: Underlay
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | A |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | C | F |
| Early Pace | A | A |
| Mid Pace | A | B |
| Late Pace | C | D |
| Finish Pos | A | D |
| Dirt Speed | A | D |
| Turf Speed | F | D |
| Distance Speed | A | C |
| Trainer Effectiveness | A | A |
| Jockey Effectiveness | A | A |
| 3rd Party Ratings | B | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | C | F |
| Early Pace | A | A |
| Mid Pace | A | B |
| Late Pace | C | D |
| Sprint FPS | A | B |
| Route FPS | B | A |
| Trainer Effectiveness | A | A |
| Jockey Effectiveness | A | A |
| 3rd Party Ratings | B | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | C | F |
| Early Pace | B | B |
| Mid Pace | A | B |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 37-day layoff with a recent workout. (2) Finish positions improving over last 4 starts. (3) Hot trainer - high ITM rate (40%). (4) Late pace significantly above field median (LP: 67 vs field 62). (5) Closer/presser in a race with pace pressure — ideal race setup for late runners. (6) Strong speed 2 back (66), modest dip last out (58) — form reversal angle, ready to bounce back to top form. (7) Had documented trouble last race (codes: ['D', 'O', 'T', 'U']) yet still finished 3rd — a clean trip today sets up a form reversal.
⚠ Concerns:(1) Has not raced in 37 days. (2) Rising in class today ($7,500 → $15,000). (3) Was claimed from its most recent race - new connections. (4) Trip note from last race: Coaxed 3-4w;outkicked. (5) Rising in class ($7,500 → $15,000) after finishing 3 last out — double negative. (6) First start at BTP — never raced here before (last seen at: TP).
Mod1(W): 100 (1/7) Mod2(W): 87 (2/7) Mod3(W): 100 (1/7) Mod1-LS: 48 (4/7) Mod2-LS: 83 (2/7) Mod3-LS: 57 (5/7) Mod4(Form): 80 (2/7) Win Prob: 60.8% Value Score: 3.56 ★ ML: 3.00 (2/1) Overlay %: -2.70 Fair Odds: 5.94 (5/1) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | A | A |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | A | D |
| Early Pace | D | C |
| Mid Pace | C | D |
| Late Pace | A | D |
| Finish Pos | D | F |
| Dirt Speed | D | F |
| Turf Speed | A | B |
| Distance Speed | D | F |
| Trainer Effectiveness | B | C |
| Jockey Effectiveness | C | B |
| 3rd Party Ratings | A | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | D |
| Early Pace | D | C |
| Mid Pace | D | D |
| Late Pace | A | C |
| Sprint FPS | B | B |
| Route FPS | D | C |
| Trainer Effectiveness | B | C |
| Jockey Effectiveness | C | B |
| 3rd Party Ratings | A | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | D |
| Early Pace | F | D |
| Mid Pace | B | C |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 126-day layoff with a recent workout. (2) Equipment change today (code: 1). (3) Hot jockey — high ITM rate (49%). (4) Closer/presser in a race with pace pressure — ideal race setup for late runners. (5) Returning from 126-day layoff with a bullet workout — trainer signaling extra readiness for today's spot.
⚠ Concerns:(1) Has not raced in 126 days. (2) Speed average (49) well below field median (58). (3) Trip note from last race: Ins-2p; no factor. (4) First start at BTP — never raced here before (last seen at: TP, IND).
Mod1(W): 0 Mod2(W): 0 Mod3(W): 0 Mod1-LS: 22 (6/7) Mod2-LS: 0 Mod3-LS: 64 (4/7) Mod4(Form): 11 (6/7) Win Prob: 32.5% Value Score: 4.27 ★ ML: 8.00 (7/1) Overlay %: 3.25 Fair Odds: 8.21 (7/1) O/U: Overlay
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | D | C |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | D | F |
| Early Pace | F | D |
| Mid Pace | D | D |
| Late Pace | B | D |
| Finish Pos | F | F |
| Dirt Speed | D | F |
| Turf Speed | F | D |
| Distance Speed | B | D |
| Trainer Effectiveness | C | F |
| Jockey Effectiveness | A | A |
| 3rd Party Ratings | D | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | D | F |
| Early Pace | F | D |
| Mid Pace | D | D |
| Late Pace | B | D |
| Sprint FPS | F | D |
| Route FPS | F | D |
| Trainer Effectiveness | C | F |
| Jockey Effectiveness | A | A |
| 3rd Party Ratings | D | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | D | F |
| Early Pace | D | C |
| Mid Pace | D | D |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 206-day layoff with a recent workout. (2) Significantly better on off tracks (off: 41 vs fast: 32). (3) Career-best speed figure (32) recorded in last 1-2 starts — peaking form. (4) Returning from 206-day layoff with a bullet workout — trainer signaling extra readiness for today's spot. (5) Had documented trouble last race (codes: ['D', 'O', 'T', 'U']) yet still finished 4th — a clean trip today sets up a form reversal.
⚠ Concerns:(1) Has not raced in 206 days. (2) Consider there will be likely pace pressure in this race - multiple E types entered. (3) Speed average (42) well below field median (58). (4) Poor record at this distance (0 wins from 5 recent starts). (5) Trip note from last race: Weakened on the turn. (6) Post 7 in a 8-horse dirt sprint — wide draw forces extra ground and a difficult first-turn path. (7) Speed figures are highly erratic over last 5 starts (σ = 13.2 pts: [41, 56, 32, 54, 26]) — unpredictable performer.
Mod1(W): 77 (4/7) Mod2(W): 36 (5/7) Mod3(W): 52 (4/7) Mod1-LS: 5 (7/7) Mod2-LS: 20 (7/7) Mod3-LS: 30 (7/7) Mod4(Form): 70 (3/7) Win Prob: 52.1% Value Score: 3.43 ★ ML: 3.50 (5/2) Overlay %: -3.93 Fair Odds: 7.87 (7/1) O/U: Underlay
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | D | C |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | D | F |
| Early Pace | B | B |
| Mid Pace | F | D |
| Late Pace | F | F |
| Finish Pos | A | D |
| Dirt Speed | A | D |
| Turf Speed | D | C |
| Distance Speed | D | F |
| Trainer Effectiveness | D | F |
| Jockey Effectiveness | D | D |
| 3rd Party Ratings | F | F |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | F | F |
| Early Pace | B | B |
| Mid Pace | F | D |
| Late Pace | F | F |
| Sprint FPS | D | D |
| Route FPS | D | D |
| Trainer Effectiveness | D | F |
| Jockey Effectiveness | D | D |
| 3rd Party Ratings | F | F |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | D | F |
| Early Pace | B | B |
| Mid Pace | F | F |
| Late Pace | C | C |
Chart Guide (2.0.35)
- Bubbles
- Color = surface: green (turf), brown (dirt), blue (synthetic)
- Top number = finishing position; bottom number in parentheses = starting post position
- Height above line = beaten lengths — closer means a smaller margin
- Border = track conditions: green (FST/GD/FM), yellow (WF/SL/YLD), red (SY/MY/FRZ/SFT/HVY)
- Connecting Line
- Solid = good conditions; dashed = bad conditions
- Line Symbol
- ▲ Triangle = sprint (7f or less); ● Circle = route (more than 7f)
- Days Between Races
- Small number above each bubble (e.g. 19d) = days since the previous race — helps spot freshness, layoffs, and tight scheduling.
- Shaded Band (Bollinger)
- Highlights volatility — a wider band means more variation in recent performance
- Performance Trend
- Straight line showing the overall performance direction — angled up = improving, angled down = declining
- Dashed Reference Line
- AI Opponent = AI analysis class-level par for the race
✓ Positives:(1) Returning from 43-day layoff with a recent workout. (2) Hot trainer - high ITM rate (44%). (3) Equipment change today (code: 1). (4) 3 workouts in last 30 days — horse appears sharp and fit. (5) Strong speed 2 back (65), modest dip last out (57) — form reversal angle, ready to bounce back to top form. (6) Consistent workout spacing (7, 7 days) — disciplined training regimen. (7) Last race E1 pace (86.0) was 4.2 pts above personal avg (81.8) and horse still finished 4th — pace collapse victim; expect bounce if today's fractions are softer. (8) Had documented trouble last race (codes: ['B', 'D', 'F', 'S', 'T', 'U']) yet still finished 4th — a clean trip today sets up a form reversal.
⚠ Concerns:(1) Has not raced in 43 days. (2) Repeatedly beaten at this class level (3 of last 5 starts). (3) Consider there will be likely pace pressure in this race - multiple E types entered. (4) Poor record at this distance (0 wins from 5 recent starts). (5) Trip note from last race: Rush btw;prmpt 2w;fade. (6) Post 8 in a 8-horse dirt sprint — wide draw forces extra ground and a difficult first-turn path. (7) First start at BTP — never raced here before (last seen at: TP).
Mod1(W): 93 (2/7) Mod2(W): 68 (3/7) Mod3(W): 86 (3/7) Mod1-LS: 41 (5/7) Mod2-LS: 61 (3/7) Mod3-LS: 40 (6/7) Mod4(Form): 60 (4/7) Win Prob: 54.1% Value Score: 3.95 ★ ML: 4.00 (3/1) Overlay %: 0.23 Fair Odds: 6.19 (5/1) O/U: Fair
Score Definitions
- Mod1(W)
- Winner Model 1 score (0–100) — AI composite rating among field contenders (higher = stronger). The number in parentheses (e.g. 1/9) shows this horse’s rank out of the field size.
- Mod2(W)
- Winner Model 2 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod3(W)
- Winner Model 3 score (0–100) — AI rating among field contenders (higher = stronger). The number in parentheses shows rank within the field.
- Mod1-LS
- Longshot Model 1 score (0–100) — AI model trained to surface high-value longshots. Higher = stronger longshot profile. Rank in parentheses shows position within the field.
- Mod2-LS
- Longshot Model 2 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod3-LS
- Longshot Model 3 score (0–100) — AI longshot model. Higher = stronger longshot profile.
- Mod4(Form)
- Form Model score (0–100) — AI sequence model trained on recent form patterns to capture momentum and trajectory. Higher = stronger recent form profile. Rank in parentheses shows position within the field.
- Win Prob
- LightGBM estimated win probability expressed as a percentage. This is a raw model output, not normalized to sum to 100% across the field — use it for comparison rather than as an absolute probability.
- Value Score
- Win Prob divided by the market-implied probability (derived from morning line). Values above 1.0 mean the model sees more win chance than the odds imply — a ★ star appears when Value Score ≥ 1.3, indicating a 30%+ model edge over the market.
- ML
- Morning line odds set by the track handicapper before wagering opens
- Overlay %
- How much the fair odds exceed the morning line, expressed as a percentage — positive values indicate value
- Fair Odds
- Model-derived fair-value odds based on estimated win probability
- O/U
- Overlay or Underlay — whether the horse offers value (O) or is overbet (U) relative to fair odds
Lane Bias Guide
- Bar Direction
- Bars above the line = more favorable lane • Bars below = less favorable, relative to the average at this track.
- Color
- Green = above-average advantage • Yellow = near-average • Red = below-average • Blue = this horse’s lane.
- Bias Score
- Derived from weighted average finishing positions across historical races at this track. Shown in the tooltip on hover. To use the bias score, calculate your handicap for the horse (e.g. finish position), then add (detrimental bias) or subtract (favorable bias) the bias score.
- Races
- Number of historical races used to compute each lane’s bias score.
Race Pace Bias Guide
- Bar Height & Direction
- Positive (green) = pace favors this horse • Negative (red) = pace works against this horse • Near-zero (amber) = neutral.
- Race Shape
- Projected race shape based on running styles of all horses (e.g. EE = two early horses, EP = one early presser).
- Pace Category
- FAST = contested early pace • LONE EARLY = one speed horse controlling pace • HONEST = moderate early pressure • SLOW = closers’ race.
- Flag (★)
- A star above the bar means this horse’s pace bias figure is ≥ 1.5 — a meaningful pace advantage or disadvantage.
Total Pace Guide
- Bar Segments
- Green = Early Pace (E1) • Amber = Mid Pace (E2) • Red = Late Pace (LP)
- Bar Height
- Total height = sum of E1 + E2 + LP average pace figures. Taller bars indicate a faster-paced horse overall.
- Grades & Rankings
- Hover to see each horse’s letter grade and ranking vs. other horses in this race (e.g. 2/9 = 2nd best of 9).
- Border
- Thicker border = this horse’s post position.
Performance vs. Peers Guide
- Bars (per post)
- Blue = avg career performance • Amber = avg last 3 races • Green = last race (improved vs. avg) • Red = last race (declined vs. avg)
- Trend Line
- Angled line above the bars: rising = improving form • falling = declining form.
- Horizontal Line
- Field average — the mean career performance of all horses in this race.
- Color Circles
- Standard post-position colors below each lane — same as the Total Pace chart.
- Border
- Thicker border = this horse’s post position.
Figure Comparison
- Rating
- The three Pick Pony algorithm ratings (Algo 1, Algo 2, Algo 3) that make up the overall PP AI Rating — higher is a stronger contender rating.
- Speed
- Career average speed, recent 3-race average, and last-race speed figure — spot horses in improving or declining form.
- Pace
- Average Early (E1), Mid (E2), and Stretch (LP) pace figures — reveals pace-setters vs. closers vs. all-the-way types.
- This horse
- The horse whose stats card is open is highlighted with a darker column background.
Winning Profiles
- Axes
- X: Win % from last 10 starts — Y: In-the-money % (top-3 finishes) from last 10 starts.
- Quadrants
- Consistent Winners (top-right): high win & ITM — reliable performers.
Place Threats (top-left): boards often but doesn’t win often — good exacta/trifecta angles.
Streaky (bottom-right): wins occasionally, misses the board often — all-or-nothing.
Developing (bottom-left): still finding form — longshot territory. - Reference Lines
- Dashed lines mark field averages. Dots to the right win more; dots higher board more.
- This horse
- Highlighted with a gold border and larger dot.
AI Model Comparison
- Winner Models (green)
- Mod1(W), Mod2(W), and Mod3(W) trained to predict race winners. Higher score = stronger contender.
- Longshot Models (red)
- Mod1-LS, Mod2-LS, and Mod3-LS trained to find high-value longshots. Higher score = stronger longshot candidate.
- Form Model (blue)
- Mod4 sequence model trained on recent form patterns. Higher score = stronger recent form profile.
- This horse
- Highlighted with a darker column background.
Race Pace Simulation Guide
- Slider
- Drag left→right to advance from gate to finish. Drag all the way back to reset the race.
- Predicted Finish Order
- Factors in contested pace burn, each horse’s effort curve (peaked-early vs. peaked-late), and wide trip cost from outside posts. Updated live when Track Bias changes.
- Pace Scenario
- Lone Speed: single E horse likely rates and holds. Contested: two E horses burn each other — closers benefit. Hot Pace: speed meltdown, deep closers thrive. Soft: no early speed, moderate fractions throughout.
- Track Bias
- Slider below canvas. −2 = strong closer/wet bias; +2 = strong speed/rail bias. Recomputes the predicted finish order.
- Boxed Indicator
- Dashed orange ring = horse is surrounded (rival ahead, rival behind, wall outside). In a real race this horse must find a clear path to run.
- Running Styles
- E/EE = early speed (rushes to rail). P = presser (stalks the pace). S = stalker. SS = slow starter / closer.
| Stat | vs. Current Card | vs. All Horses |
|---|---|---|
| Averages
Career-average metrics across all AI race analysis. Performance and pace averages reflect estimated historical performance; trainer and jockey grades score in-the-money percentages. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Pick Pony AI Rating | B | A |
| Pace Trip Longshot | — | — |
| Connections Longshot | — | — |
| Situational Longshot | — | — |
| Speed | B | D |
| Early Pace | D | C |
| Mid Pace | B | C |
| Late Pace | B | D |
| Finish Pos | A | D |
| Dirt Speed | D | F |
| Turf Speed | C | C |
| Distance Speed | A | C |
| Trainer Effectiveness | A | B |
| Jockey Effectiveness | B | B |
| 3rd Party Ratings | B | D |
| Recent
Performance analysis from the horse's recent races, weighted to reflect current form. Sprint and Route FPS (Foot Per Second) measure pace efficiency over the horse's recent sprints and routes respectively. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | A | D |
| Early Pace | D | C |
| Mid Pace | B | C |
| Late Pace | A | D |
| Sprint FPS | B | B |
| Route FPS | A | A |
| Trainer Effectiveness | A | B |
| Jockey Effectiveness | B | B |
| 3rd Party Ratings | B | D |
| Predictions
AI-generated projections for the upcoming race, derived from prior AI analysis for speed and pace trends, surface and distance suitability, and current conditions. Grading scale: A = elite • B = above average • C = average • D = below average • F = poor • — = no data |
||
| Speed | B | D |
| Early Pace | D | C |
| Mid Pace | D | D |
| Late Pace | C | C |
TL;DR
Toggle
Get the full Handicapper Report, AI picks, expert analysis, and more with a Pick Pony subscription!
Belterra Park — May 2, 2026 | Quick Handicapper Reference
Fast-to-good dirt surface, inside posts favored in sprints, speed on the front end productive all day. Larry Smith barn is the dominant stable presence; J. McKee and Guillermo Rodriguez lead the jockey colony. Bet accordingly.
Race 1 — 1320f | Maiden Special Weight
Bob Mo (1) is the 3-1 morning line choice from the rail and is the clear top selection in a manageable five-horse field. Spinning Spirit (4) and Turn Him Loose (5) are the logical exotics fillers. Primary exacta: Bob Mo (1) over Spinning Spirit (4).
Race 2 — 1830f | Claiming $5,000
Basscoat (4) is the class of this five-horse route with a top jockey and ideal post position to rate and pounce. Garden Leave (2) stalks from the two-hole and is the primary exacta companion. Royally Bitter (5) is the value trifecta closer at 4-1 from the outside.
Race 3 — 1320f | Claiming $5,000 Mares
Silent Sky (1) at 2-1 from the rail for the Smith barn is among the strongest win singles on the card. Councilwoman Jilly (3) with McKee up is the natural exacta partner. Bisa (4) at 8-1 is worth trifecta coverage as an overlay.
Race 4 — 1320f | Claiming $5,000 Mares N2L
Sweet Lilibet (1) at 3-1 from the inside post is the top choice in a competitive n2l field. Joia Rara (3) and Calcetines (5) round out the primary exotics at matching 4-1 prices. Little Juanita (6) warrants a small trifecta spot from the outside.
Race 5 — 1320f | Allowance $26,500 (Feature)
Big Truck (6) at 2-1 for the Smith barn with Rodriguez is the top selection, sharing co-favorite billing with Superwolf (7) at 2-1 for Gorham and Correa. Obi Kenobi (2) at 3-1 with McKee stalking from post 2 is the best value play if either favorite is pressed early. Check The Facts (1) and Just Like Rambo (4) deserve trifecta inclusion; any drift to 3-1 or better on Superwolf (7) at the windows is a standalone win bet signal.
Race 6 — 1210f | Maiden Claiming $5,000 Mares
Blue Opal (1) at 4-1 from the inside post in the largest field of the day is the top selection. Victura (5) at 4-1 and Elevated Power (4) at 5-1 are the primary exotic partners. Bizy Dreamin (7) with McKee and Grab My Tab (8) with Rodriguez are live at 5-1 and 4-1 respectively; Mescool (2) at 10-1 and Corellia (6) at 12-1 are speculative trifecta and superfecta fillers.
Race 7 — 1320f | Maiden Claiming $5,000
Halfway Joking (5) at 2-1 is the top selection in this eight-horse co-favorite showdown, with Franklin Wilder (3) at 2-1 the natural exacta box partner. Grand Kingdom (2) at 3-1 for the Smith barn with Rodriguez is the key secondary threat. Iknowyoucanwait (7) at 6-1, also for Smith, is the sleeper trifecta play in a dual-entry scenario.
Race 8 — 1320f | Maiden Claiming $15,000 Mares
Sisterlithic (1) at 2-1 from the rail is the second-strongest win single of the day and a logical parlay target off Silent Sky (1) in Race 3. Always In Vogue (4) at 3-1 with Rodriguez is the primary exacta threat. La Heroica (7) and Rosenvold (8) at 4-1 each are trifecta closers; No Solution (5) at 8-1 is the preferred option among Monday's dual entry in the nightcap.
Before wagering, check the Latest Race Scratches, Changes, and Conditions post for any last-minute scratches or race changes.
Unlock the full Handicapper Report and enjoy all Pick Pony membership perks!
✓ Positives:(1) Returning from 101-day layoff with a recent workout. (2) Speed average (74) well above field median (60). (3) Hot trainer - high ITM rate (43%). (4) Big last-race effort (80) — 9 pts above recent average. (5) Equipment change today (code: 2). (6) Hot jockey — high ITM rate (41%). (7) Early pace significantly above field median (E1: 89 vs field 86). (8) Speed improved +10 pts then +8 pts in last 2 starts (62 → 72 → 80) — trainer is progressively building fitness. (9) Last race E1 pace (96.0) was 6.8 pts above personal avg (89.2) and horse still finished 6th — pace collapse victim; expect bounce if today's fractions are softer. (10) Career-best figure (86) buried in races 4–10 back — 26 pts above race median and 6 pts above recent form — market has forgotten the peak effort. (11) Averages 79.0 speed pts in today's distance/surface configuration vs 71.0 in other configs (4 qualifying starts) — returning to ideal setup.