What a week to be a Pick Pony subscriber! Across 846 races covered between September 4–10, our AI suite delivered some truly eye-popping numbers. Leading the charge was Algo 2 (AI Ensemble), which posted the week's best win rate at 37.5% on full-day picks and an even hotter 38.7% on morning updates — right at the chalk baseline of 38.2% and in some cases edging past it. More impressively, Algo 2 also topped the board in In-The-Money rate at 71.4%, meaning nearly 3 out of every 4 top picks finished on the board. Our flagship AI Exacta model was equally strong, hitting 37.0% wins and a 69.7% ITM rate across 733 settled races, with a remarkable 72 exacta hits and 71 boxed trifecta hits on the week — the kind of exotic production that turns good weeks into great paydays. Algo 3 (AI Win Probability) matched that 37.0% win clip in its full-day set and delivered a stellar 69.5% ITM rate, demonstrating that our probability engine is firing on all cylinders. With 35 races still pending settlement, those numbers have only one direction to go: up.
Standout Wins
The highlight reel this week is loaded. On the longshot front, Algo 1 lit up the board on September 6 at Emerald Downs when POP D'ORO romped home at 15-1 morning line odds, a tremendous value call that rewarded patient bettors handsomely. Right behind it, STYLISH GENT (8-1 ML) at Finger Lakes on September 7 and GIRLS DON'T CRY (8-1 ML) at Albuquerque on September 4 both delivered for Algo 1 followers, proving the model's knack for finding winners the crowd underestimates. The AI Exacta model had its own banner moments, with SARAH'S RUNNER kicking things off at Evangeline Downs on September 5 and INDIAN PHAROAH closing out the stretch run at Delaware on September 9 — both at 4-1 ML and both first across the wire. Meanwhile, Algo 2 (AI Ensemble) showed its range by nailing back-to-back wins at Charles Town on September 7 — SILVER ROAD in Race 3 and STORM'S WAKE in Race 8, both at 5-1 ML — a devastating one-two punch at the same track on the same card. Algo 3 deserves a tip of the cap for FROST MOUNTAIN at 5-1 ML at Parx on September 9, a pick that appeared across multiple model variants and rewarded the conviction.
Key Insights
A few clear patterns emerged from this week's data worth tucking into your handicapping toolkit. First, our AI models showed exceptional strength in mid-tier claiming and allowance races at regional tracks, with standout performances at Albuquerque, Charles Town, Parx, and Emerald Downs suggesting the algorithms are particularly well-calibrated for those competitive but often overlooked circuits. Second, the morning-update versions of our models continued to sharpen results — Algo 2's morning update nudged its win rate to 38.7% and ITM to 71.6%, confirming that our real-time data ingestion and late-scratch adjustments are adding genuine edge. Third, the exotic hit volume was exceptional across the board: combined, our models generated over 350 exacta, trifecta, and boxed exotic hits in a single week, underscoring that Pick Pony isn't just picking winners — it's structuring tickets intelligently. The Weight Set 3 models also quietly put up the week's best show rate at 13.7–13.8%, making them a reliable anchor for trifecta and superfecta wheels heading into next week.
Looking Ahead
With 35 races still pending settlement, this week's final numbers will only look better — and given the momentum our AI suite is carrying, next week's cards across the full fall meet schedule set up as a prime opportunity to build on what's been a genuinely impressive stretch. Stay locked in to Pick Pony, trust the models, and let the data do the work — the horses are running our way. 🏇
Weekly Performance Summary
How each pick source's top selection finished across every graded race this week.
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).
The top two horses in our exotics ordering, played as a $1 straight exacta.
Metric
Count
Rate
Total Races
846
Top Pick Wins (1st)
271
37.0%
Top Pick Places (2nd)
149
20.3%
Top Pick Shows (3rd)
91
12.4%
Top Pick ITM
511
69.7%
Pending
35
Exotic Wager Hits (AI Exacta Winner)
How often the source's top three finished 1-2-3, counted as straight and boxed exactas and trifectas.
Wager Type
Hits
Description
Exacta
72
Picks 1 & 2 finished 1st & 2nd in exact order
Boxed Exacta
49
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
71
Picks 1, 2 & 3 all finished in top 3, any order
Algo 1 Winner
Our weighted rating: every horse is scored on speed, pace, class, form and connections, and the top score is the pick.
Metric
Count
Rate
Total Races
846
Top Pick Wins (1st)
232
31.8%
Top Pick Places (2nd)
135
18.5%
Top Pick Shows (3rd)
89
12.2%
Top Pick ITM
456
62.6%
Pending
35
Exotic Wager Hits (Algo 1 Winner)
How often the source's top three finished 1-2-3, counted as straight and boxed exactas and trifectas.
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
17
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Boxed Trifecta
64
Picks 1, 2 & 3 all finished in top 3, any order
Algo 1, morning update Winner
Our weighted rating: every horse is scored on speed, pace, class, form and connections, and the top score is the pick. This is the morning-update version: the same picks re-ranked after the official scratches and surface changes.
Metric
Count
Rate
Total Races
846
Top Pick Wins (1st)
233
31.7%
Top Pick Places (2nd)
136
18.5%
Top Pick Shows (3rd)
92
12.5%
Top Pick ITM
461
62.7%
Pending
35
Exotic Wager Hits (Algo 1, morning update Winner)
How often the source's top three finished 1-2-3, counted as straight and boxed exactas and trifectas.
Wager Type
Hits
Description
Exacta
56
Picks 1 & 2 finished 1st & 2nd in exact order
Boxed Exacta
44
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
17
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Boxed Trifecta
68
Picks 1, 2 & 3 all finished in top 3, any order
Weight Set 2 (shadow) Winner
An experimental variant of the Algo 1 rating with a different weight profile, graded here to see whether it earns a place in the published ladder.
Metric
Count
Rate
Total Races
846
Top Pick Wins (1st)
252
35.5%
Top Pick Places (2nd)
113
15.9%
Top Pick Shows (3rd)
95
13.4%
Top Pick ITM
460
64.8%
Pending
35
Exotic Wager Hits (Weight Set 2 (shadow) Winner)
How often the source's top three finished 1-2-3, counted as straight and boxed exactas and trifectas.
Wager Type
Hits
Description
Exacta
54
Picks 1 & 2 finished 1st & 2nd in exact order
Boxed Exacta
39
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
11
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Boxed Trifecta
52
Picks 1, 2 & 3 all finished in top 3, any order
Weight Set 2, morning update Winner
An experimental variant of the Algo 1 rating with a different weight profile, graded here to see whether it earns a place in the published ladder. This is the morning-update version: the same picks re-ranked after the official scratches and surface changes.
Metric
Count
Rate
Total Races
846
Top Pick Wins (1st)
256
35.7%
Top Pick Places (2nd)
114
15.9%
Top Pick Shows (3rd)
96
13.4%
Top Pick ITM
466
64.9%
Pending
35
Exotic Wager Hits (Weight Set 2, morning update Winner)
How often the source's top three finished 1-2-3, counted as straight and boxed exactas and trifectas.
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
12
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Boxed Trifecta
53
Picks 1, 2 & 3 all finished in top 3, any order
Weight Set 3 (shadow) Winner
An experimental variant of the Algo 1 rating with a different weight profile, graded here to see whether it earns a place in the published ladder.
Metric
Count
Rate
Total Races
846
Top Pick Wins (1st)
252
34.6%
Top Pick Places (2nd)
131
18.0%
Top Pick Shows (3rd)
100
13.7%
Top Pick ITM
483
66.3%
Pending
35
Exotic Wager Hits (Weight Set 3 (shadow) Winner)
How often the source's top three finished 1-2-3, counted as straight and boxed exactas and trifectas.
Wager Type
Hits
Description
Exacta
57
Picks 1 & 2 finished 1st & 2nd in exact order
Boxed Exacta
47
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
13
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Boxed Trifecta
63
Picks 1, 2 & 3 all finished in top 3, any order
Weight Set 3, morning update Winner
An experimental variant of the Algo 1 rating with a different weight profile, graded here to see whether it earns a place in the published ladder. This is the morning-update version: the same picks re-ranked after the official scratches and surface changes.
Metric
Count
Rate
Total Races
846
Top Pick Wins (1st)
255
34.7%
Top Pick Places (2nd)
131
17.8%
Top Pick Shows (3rd)
102
13.9%
Top Pick ITM
488
66.4%
Pending
35
Exotic Wager Hits (Weight Set 3, morning update Winner)
How often the source's top three finished 1-2-3, counted as straight and boxed exactas and trifectas.
An AI estimate of each horse's chance of winning, blended with the morning line; the most likely winner is the pick. This is the morning-update version: the same picks re-ranked after the official scratches and surface changes.
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
20
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Boxed Trifecta
71
Picks 1, 2 & 3 all finished in top 3, any order
Algo 2 (AI Ensemble), morning update Winner
Several independent AI models each rank the race; the horse they agree on most is the pick. This is the morning-update version: the same picks re-ranked after the official scratches and surface changes.
How often the source's top three finished 1-2-3, counted as straight and boxed exactas and trifectas.
Wager Type
Hits
Description
Exacta
39
Picks 1 & 2 finished 1st & 2nd in exact order
Boxed Exacta
21
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
11
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Boxed Trifecta
37
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.
Horses at 10-1 or higher on the morning line that our models rated a value play; counted per horse, not per race.
Metric
Count
Rate
Longshot Selections
437
Races Covered
230
Wins (1st)
16
4.6%
Places (2nd)
32
9.2%
Shows (3rd)
26
7.5%
ITM
74
21.3%
Pending
12
Longshot Betting Strategy Simulator
What a flat win bet on every flagged longshot would have returned this week at each minimum-edge threshold.
Each strategy below replays the past 60 days of AI longshot picks as actual $2 mutuel bets and reports the realised return. A longer window than the rest of this report is used here on purpose: longshots are rare and extremely high-variance, so a single week's results are dominated by luck — only over a larger sample do the calibrated edge strategies separate from simply betting every longshot. Bets are deduplicated by (track, race, horse) so a horse picked by all three algorithms generates one bet, not three. Edge strategies use a calibrated win probability (from the longshot meta-learner) and bet only when P(win) × decimal odds − 1 exceeds the listed threshold — these target overlays rather than long prices for their own sake. Hit% is the share of tickets that cashed any leg. Published longshots graded: 2842; wider 10/1-and-longer universe available to the edge rows: 14162, of which 10195 could be assigned a calibrated edge. Calibrator: out-of-sample tiered calibrator on the cross-race win probability (lgbm_win_prob_fwd), trained through 2026-07-30; edge rows replay only the 42 days after that date, so no graded race was ever seen by the calibrator. The edge filters use only leak-free forward scores (lgbm_win_prob_fwd — captured before each race ran), so the ROIs are a genuine out-of-sample estimate — never the in-sample re-scored values that previously inflated this section. The two families are measured differently — read them separately. The raw-filter rows grade the longshots we actually published, over the full 60 days. The edge rows sweep a different population: every thoroughbred runner priced at 10/1 or longer, published or not. That matters, because being published as a longshot already requires passing an edge test — so re-applying an edge threshold to the published picks would just re-filter an already-filtered group and tell you nothing. The edge rows also cover only the most recent 42 days (10195 runners), because everything before 2026-07-30 was needed to fit the calibrator and grading it would be marking its own homework. That window widens on its own as forward scores accumulate.
Strategy
Bets
Tickets Cashed
Stake
Payout
Profit
ROI
$2 WIN — all longshots
2842
159 (5.6%)
$5684.00
$3208.00
$-2476.00
-43.6%
$2 PLACE — all longshots
2842
418 (14.7%)
$5684.00
$3919.44
$-1764.56
-31.0%
$2 SHOW — all longshots
2842
687 (24.2%)
$5684.00
$3829.84
$-1854.16
-32.6%
$2 WIN + $2 PLACE
2842
418 (14.7%)
$11368.00
$7127.44
$-4240.56
-37.3%
$2 WIN — ML ≥ 15/1
1490
62 (4.2%)
$2980.00
$1528.44
$-1451.56
-48.7%
$2 WIN — ML ≥ 20/1
893
26 (2.9%)
$1786.00
$686.16
$-1099.84
-61.6%
$2 WIN — field ≥ 8
1843
93 (5.0%)
$3686.00
$2077.38
$-1608.62
-43.6%
$2 WIN — edge ≥ 5%
5840
342 (5.9%)
$11680.00
$7590.16
$-4089.84
-35.0%
$2 WIN — edge ≥ 10%
4666
277 (5.9%)
$9332.00
$6108.72
$-3223.28
-34.5%
$2 WIN — edge ≥ 20%
4514
273 (6.0%)
$9028.00
$5975.70
$-3052.30
-33.8%
$2 WIN — edge ≥ 30%
2891
204 (7.1%)
$5782.00
$3639.04
$-2142.96
-37.1%
$2 WIN — edge ≥ 40%
1630
126 (7.7%)
$3260.00
$2208.56
$-1051.44
-32.3%
$2 WIN — edge ≥ 50%
123
10 (8.1%)
$246.00
$150.44
$-95.56
-38.8%
$2 WIN — edge ≥ 60%
0
—
$0.00
$0.00
$+0.00
—
$2 WIN — edge ≥ 70%
0
—
$0.00
$0.00
$+0.00
—
$2 WIN — edge ≥ 80%
0
—
$0.00
$0.00
$+0.00
—
$2 WIN — edge ≥ 90%
0
—
$0.00
$0.00
$+0.00
—
$2 WIN — edge ≥ 100%
0
—
$0.00
$0.00
$+0.00
—
$2 WIN — edge ≥ 10% AND field ≥ 8
2628
145 (5.5%)
$5256.00
$3600.96
$-1655.04
-31.5%
Trifecta Structure Plays
Key the top pick to win over the next three horses (six combinations per race); the tag shows whether the race shape looked playable.
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
426
Trifectas hit
52 (12.2%)
Total staked
$1,476.00
Total returned
$994.61
Profit / loss
−$481.39 (-32.6% ROI)
⭐ Best Bet races only
9 plays, 1 hit (11.1%), −$23.90 (-72.4% ROI)
398 race(s) skipped for scratches inside the top four; 22 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
What betting the morning-line favorite in every race would have returned: the baseline every pick source has to beat.
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
613
Favorite Wins (1st)
234
38.2%
Favorite Places (2nd)
103
16.8%
Favorite Shows (3rd)
91
14.8%
Favorite ITM
428
69.8%
Consensus Expert Picks
Picks from more than twenty-five published handicappers, weighted into one consensus selection per race.
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.
Metric
Count
Rate
Total Races
662
Top Pick Wins (1st)
209
37.3%
Top Pick Places (2nd)
103
18.4%
Top Pick Shows (3rd)
82
14.6%
Top Pick ITM
394
70.4%
Pending
102
Exotic Wager Hits (Consensus Top 3 Picks)
How often the source's top three finished 1-2-3, counted as straight and boxed exactas and trifectas.
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
11
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Boxed Trifecta
56
Picks 1, 2 & 3 all finished in top 3, any order
Performance by Breed
The same results split by thoroughbred and quarter horse races, which are handicapped differently.
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
846
232 (32%)
135 (19%)
89 (12%)
456 (63%)
56
44
17
64
Algo 2 Winner
846
277 (38%)
159 (22%)
91 (12%)
527 (71%)
69
54
20
71
Algo 3 Winner
555
176 (37%)
81 (17%)
74 (16%)
331 (70%)
50
33
12
41
Weight Set 2 Winner
846
252 (35%)
113 (16%)
95 (13%)
460 (65%)
54
39
11
52
Weight Set 3 Winner
846
252 (35%)
131 (18%)
100 (14%)
483 (66%)
57
47
13
63
AI Longshot
437
16 (5%)
32 (9%)
26 (7%)
74 (21%)
—
—
—
—
Performance by Track
The same results split by track, to show where each source is strongest and weakest.
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. Algo 1 is the weighted rating, Algo 2 the AI Ensemble and Algo 3 the AI Win Probability model — ordered by measured performance, so a higher number should be the stronger source; the Weight Set rows are experimental profiles graded as shadows. Comparing sources at a specific track helps identify which 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.
