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
AI Model Score History
AI Model History Guide
Lines
Each solid line shows this horse’s model score over its last 10 races: blue = Mod1, green = Mod2, amber = Mod3.
Dashed Lines
Dashed lines represent the current race’s field average for each model — a score below the dashed line indicates the horse ranked below average in this model’s assessment.
Higher is Stronger
All three model scores are 0–100: a higher number means the model rates this horse more favorably among the field.
Example preview — subscribe to view AI Model Score HistorySubscribe now →
Example preview — upgrade to view AI Model Score BadgesUpgrade now →
Track Bias by Lane
Example preview — subscribe to view Track Bias by LaneSubscribe now →
Race Pace Bias
Example preview — subscribe to view Race Pace BiasSubscribe now →
Total Pace
Example preview — subscribe to view Total PaceSubscribe now →
E1 / E2 / LP Pace Positions
E1 — Early
E2 — Mid
LP — Stretch
Example preview — subscribe to view E1 / E2 / LP PositionsSubscribe now →
Performance vs. Peers
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
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
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.
Example preview — subscribe to view Winning ProfilesSubscribe now →
AI Model Comparison
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.
Example preview — subscribe to view AI Model ComparisonSubscribe now →
Race Pace Simulation
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.
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.
E1 — EarlyE2 — MidLP — Stretch
🏁 GateFinish 🏁
Track Bias◀ CloserSpeed ▶0
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
B
A
Dirt Speed
A
B
Turf Speed
C
D
Distance Speed
A
B
Trainer Effectiveness
D
C
Jockey Effectiveness
D
F
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
A
A
Late Pace
B
D
Sprint FPS
A
A
Route FPS
A
A
Trainer Effectiveness
D
C
Jockey Effectiveness
D
F
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
What a week to be a Pick Pony subscriber! Across 748 races covered, our AI Picks suite delivered outstanding results from coast to coast, with every algorithm finishing the week strong. Leading the charge was the Algorithm Ensemble, our top performer this period, posting an eye-popping 38.2% win rate on top picks — a full two percentage points ahead of the morning-line chalk baseline of 36.5% — while landing in-the-money an impressive 70.9% of the time. Right alongside it, the Ai Exacta model matched that elite level with a 36.2% win rate and a chart-topping 71.2% ITM rate, meaning nearly three out of every four top picks hit the board. Even our individual algorithm models — Algo1, Algo2, and Algo3, along with their Americanized variants — all posted solid win rates in the 29–32% range and ITM rates above 63%, consistently outpacing random chance across hundreds of settled races. With 27 races still pending across all models, those numbers have nowhere to go but up.
Standout Wins
The highlight reel this week is packed with value plays that remind us exactly why smart handicapping matters. On the exotic side, Algorithm Ensemble led all models with 67 exacta hits, 41 boxed exactas, 24 trifectas, and 58 boxed trifectas — numbers that translate to serious ticket cashing opportunities for our players. Algo1 also dazzled with one of the week's most exciting longshot calls, flagging Cali Cream (ML 15-1) at RPX on September 3rd as a top pick — and that horse delivered in style. Add NICKY'S LITTLE SIS (ML 10-1) at Canterbury Park on August 29th, a win shared across Algo1, Algo2, Algo3, and their Am variants, and you have a pick that nearly every algorithm agreed on — and every bettor who followed it was rewarded handsomely. Other standout value wins included SHOPPING ACCIDENT (ML 8-1) at Elko (Algo1), FIVE CHARLIE (ML 8-1) at Prairie Meadows (Algo2), and — turning heads at Del Mar — MARLA HOOCH (ML 5-1), flagged by both Ai Exacta and Algorithm Ensemble, proving our models are dialed in at one of the sport's marquee summer venues. CAPTURED (ML 5-1) at Indiana Downs on September 1st was another consensus gem, appearing across Ai Exacta, Algo2, Algo3, and Ensemble — a true cross-model conviction play that paid off beautifully.
Key Insights
A few patterns stood out clearly this week. First, cross-model consensus picks proved especially powerful — horses like CAPTURED, NICKY'S LITTLE SIS, SEASIDE STARTUP, JUNIOR BATTLE, and ENJOY THE MUSIC appeared on multiple algorithm top-pick lists simultaneously, and all of them won. When our models agree, the signal is worth paying close attention to. Second, mid-range overlay horses in the 4-1 to 6-1 range were the sweet spot this week, delivering strong returns without the variance of bigger longshots — names like ALMIGHTY ONE, KIRBY DERBY, PRISONER, BLAST FURNACE, and CANDY ROCKETTE all fit this profile perfectly. Finally, the Ensemble model's trifecta total of 24 was the week's best across all algorithms, suggesting that when you lean on the combined intelligence of our full model suite for multi-leg exotic construction, the results speak for themselves.
Looking Ahead
With 27 races still pending settlement across all models, this week's final numbers may climb even higher — so stay tuned for updates as those results roll in. Next week, keep a close eye on the Ensemble and Ai Exacta models for your top-pick selections, and watch for cross-algorithm consensus plays that could give you another edge heading into the heart of the fall racing season. The horses are running, the models are firing, and Pick Pony is just getting warmed up! 🏇
Weekly Performance Summary
Ai Exacta Winner
Pick Pony's AI Winner models rank every horse in each race using a weighted combination of speed figures, pace analysis, trainer/jockey statistics, track bias, and dozens of other handicapping factors. Each algorithm uses a different set of optimized weights to capture different angles on the same race data. Top Pick Wins measures how often the model's #1 ranked horse won the race outright. Places and Shows count 2nd and 3rd place finishes respectively. ITM (In The Money) is the combined total of Win + Place + Show. Rates are calculated against settled races only (pending results excluded). The Exotic Wager Hits table shows how often the model's top 3 picks finished in the correct order (Exacta/Trifecta) or in any order within the top positions (Boxed Exacta/Boxed Trifecta).
Metric
Count
Rate
Total Races
748
Top Pick Wins (1st)
232
36.3%
Top Pick Places (2nd)
127
19.9%
Top Pick Shows (3rd)
97
15.2%
Top Pick ITM
456
71.4%
Pending
27
Exotic Wager Hits (Ai Exacta Winner)
Wager Type
Hits
Description
Exacta
63
Picks 1 & 2 finished 1st & 2nd in exact order
Boxed Exacta
44
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
18
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
19
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Boxed Trifecta
48
Picks 1, 2 & 3 all finished in top 3, any order
Algorithm Ensemble Winner
Metric
Count
Rate
Total Races
748
Top Pick Wins (1st)
243
38.2%
Top Pick Places (2nd)
124
19.5%
Top Pick Shows (3rd)
84
13.2%
Top Pick ITM
451
70.9%
Pending
27
Exotic Wager Hits (Algorithm Ensemble Winner)
Wager Type
Hits
Description
Exacta
67
Picks 1 & 2 finished 1st & 2nd in exact order
Boxed Exacta
41
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
24
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Boxed Trifecta
58
Picks 1, 2 & 3 all finished in top 3, any order
AI Longshot Picks
Longshot picks are individual horses flagged at publication time: their morning-line odds were 8-1 or greater — prices the betting public considers unlikely to win — and the model's calibrated win probability said the price overpaid. Unlike the winner models above, a longshot is not a race's top-ranked horse, so these tables count selections, not races: every flagged horse is one bet, graded on its own finish, and a single race can contribute more than one. (For the same reason there is no exotic-hits table here — an exacta or trifecta needs a ranked 1-2-3 ticket, which a set of longshot flags does not define.) There is also one longshot table rather than one per algorithm: the flag depends only on the horse's price and its calibrated edge, never on where an algorithm ranked it, so all three models publish exactly the same longshot list. Because longshots win far less frequently by definition, win rates in this section are expected to be substantially lower than the winner models above. That is normal and by design — even a modest longshot win rate of 5-10% can be highly profitable because the payoffs are dramatically larger. A $2 win bet on an 8-1 longshot returns $18. Focus on ITM rate and on the realised ROI in the Longshot Betting Strategy Simulator below rather than on raw win percentage when evaluating longshot model performance.
Metric
Count
Rate
Longshot Selections
384
Races Covered
185
Wins (1st)
9
3.8%
Places (2nd)
21
9.0%
Shows (3rd)
17
7.3%
ITM
47
20.1%
Pending
74
Trifecta Structure Plays
Each AI Predictions post suggests one trifecta structure per race: the Algo 3 top pick to win, with its next three ranked horses boxed for 2nd and 3rd (6 combinations). Races our race-shape model rates most favorable are tagged ⭐ Best Bet in the post. This table grades last week's published plays against the actual trifecta payoffs at each track's real wager minimum — no assumptions, no hindsight.
Plays graded
360
Trifectas hit
39 (10.8%)
Total staked
$1,275.60
Total returned
$1,200.98
Profit / loss
−$74.62 (-5.8% ROI)
⭐ Best Bet races only
9 plays, 2 hit (22.2%), −$19.01 (-56.6% ROI)
365 race(s) skipped for scratches inside the top four; 23 skipped where the track's trifecta minimum wasn't recorded. Per-race wager minimums have been captured since July 4, 2026, so this ledger grows more complete each week.
Benchmark: The Morning-Line Favorite
The simplest possible strategy is to always bet the morning-line favorite — the “chalk.” This table shows how that baseline performed across the same thoroughbred races this week, so you can judge the AI's win rates against what the public's top choice delivered. Beating the chalk consistently is the real test of any handicapping model.
Metric
Count
Rate
Settled Races
564
Favorite Wins (1st)
206
36.5%
Favorite Places (2nd)
126
22.3%
Favorite Shows (3rd)
77
13.7%
Favorite ITM
409
72.5%
Consensus Expert Picks
Consensus Expert Picks are derived from multiple independent handicapping sources. Each source submits their top picks per race, and the consensus ranking is built by scoring each horse: 4 points for a 1st-place pick, 3 for 2nd, 2 for 3rd, and 1 for 4th. The horse with the highest total score becomes the consensus top pick. Agreement % measures what share of experts included a horse in their picks — higher agreement signals stronger conviction across sources. The Win/Place/Show/ITM and Exotic stats below measure how the consensus top pick actually performed on race day.
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
15
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Boxed Trifecta
46
Picks 1, 2 & 3 all finished in top 3, any order
Performance by Breed
Top-pick win rates vary dramatically by breed. Thoroughbred (TB) racing is the most form-driven and predictable, while Quarter Horse (QH) and Arabian (AR) sprints are decided in a few hundred yards and are far more chaotic — they naturally produce lower win rates. Because the summary above blends all breeds together, this table separates them so each breed's true performance is visible. W/P/S are wins, places (2nd), and shows (3rd), with the percentage of settled races in parentheses. ITM is the In-The-Money total (Win + Place + Show). EX/BEX/TRI/BTRI are exacta, boxed-exacta, trifecta, and boxed-trifecta hits on the top picks. Longshot rows count individual flagged selections rather than races (a race can contribute more than one), and show dashes for the exotic columns because those wagers need a ranked 1-2-3 ticket.
Thoroughbred
Source
Races
W
P
S
ITM
EX
BEX
TRI
BTRI
Algo 1 Winner
748
202 (32%)
118 (19%)
82 (13%)
402 (64%)
58
39
17
42
Algo 2 Winner
748
190 (30%)
119 (19%)
102 (16%)
411 (65%)
54
30
18
47
Algo 3 Winner
748
190 (30%)
117 (18%)
101 (16%)
408 (64%)
54
28
19
46
AI Longshot
384
9 (4%)
21 (9%)
17 (7%)
47 (20%)
—
—
—
—
Performance by Track
The tables below break down each pick source's performance at individual tracks. W/P/S show the number of wins, places (2nd), and shows (3rd) with the percentage of settled races in parentheses. ITM is the combined In The Money total (Win + Place + Show). EX (Exacta) counts races where the top 2 picks finished 1st and 2nd in exact order. BEX (Boxed Exacta) counts races where the top 2 picks both finished in the top 2 in either order. TRI (Trifecta) and BTRI (Boxed Trifecta) apply the same logic to the top 3 picks finishing in the top 3 positions. Longshot rows count individual flagged selections rather than races (a race can contribute more than one) and show dashes in the exotic columns, because those wagers need a ranked 1-2-3 ticket that a per-horse longshot flag does not define. Each algorithm (Algo 1, 2, 3) uses different optimized weights, so their performance varies by track — comparing across algorithms at a specific track helps identify which model is strongest for that venue. The ML Favorite (chalk) row shows how the morning-line favorite performed in that track's thoroughbred races — the baseline any pick source needs to beat to be adding value at that venue.
Mod1(W): 82 (1/8) Mod2(W): 75 (2/8) Mod3(W): 79 (3/8) Mod1-LS: 55 (6/8) Win Prob: 18.4% Value Score: 1.35 ★