What a week to be a Pick Pony subscriber! Our AI Picks system fired on all cylinders across more than 720 races this week, delivering consistent, competitive results from coast to coast. Across all six algorithm variants, our top picks landed in-the-money (win, place, or show) at rates ranging from 60.4% to 63.2% — meaning nearly two out of every three top selections finished in the top three. Leading the charge was Algo2 and its variants, posting a top-pick win rate of 29.0–29.1%, while Algo3 Am edged ahead with the week's best individual mark at 29.9%. It's worth keeping the benchmark in mind: the morning-line favorite — the horse the industry collectively points to as the most likely winner — won just 34.1% of comparable races this week. Our algorithms came within striking distance of that chalk baseline while consistently surfacing horses at much more attractive odds, which is exactly where the value lives. With 16 races still pending settlement across all algorithms, those final numbers have only one direction to go.
Standout Wins
If you love a longshot, this week was made for you. The headliner of the week has to be SMART UNCLE at Saratoga on August 5th — a horse our Algo1 and Algo1 Am variants both tabbed as their top pick at a generous 10-1 morning line, and that horse went out and won. That's the kind of overlay that makes a player's week. But the value didn't stop there. Algo2 and Algo3 both nailed DIVISIBLE BY (DMR, Aug 9) and HENNA GAL (WOX, Aug 6), both at 8-1, while Algo2 added BOURBON TIGER at 6-1 (TIL, Aug 5) to the highlight reel. Meanwhile, Pennsylvania players had reason to celebrate two days running, as BLAST OF MISCHIEF and WHISPER ROAR — both at 5-1 — came home winners at PEN on August 5th, hits confirmed across multiple algorithms. And let's not overlook SMALLTOWNSMOKESHOW at Mountaineer on August 4th and ESOOOO at CNL on August 8th, each a 5-1 winner that rewarded players who trusted the process. On the exotic side, Algo3 led the way with 55 exacta hits and 13 trifecta hits, giving multi-race players plenty of reasons to smile.
Key Insights
This week's results offer a few compelling patterns worth noting. First, our algorithms showed exceptional range, producing winners at tracks as varied as Saratoga, Del Mar, Evangeline Downs, and smaller venues like Tillamook and Wyoming — a sign that the models are calibrated broadly, not just optimized for the big stages. Second, the 5-1 morning-line price point was a particularly sweet spot this week, with a remarkable number of winners clustered right there — horses the public wasn't fully embracing, but our AI clearly identified as undervalued. Third, Algo3 and Algo3 Am's strong trifecta conversion rate (13 hits each) suggests the models are doing excellent work not just picking winners, but sequencing the order of finish — a crucial edge for players building exotic tickets. Finally, with ITM rates holding above 60% across every single algorithm this week, the depth of the picks held up beautifully, giving place and show bettors consistent returns even on days when the win column was competitive.
Looking Ahead
The ponies don't rest, and neither does our AI — with 16 races still pending settlement, this week's already impressive numbers could tick even higher, so stay tuned for final updates. Next week, keep your eyes on how our models handle any late-summer track bias shifts, and make sure you're loading up on those 5-1 to 10-1 selections — because as this week proved, that's where Pick Pony's AI is doing its best work! 🏇
Weekly Performance Summary
Algorithm Algo1 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
721
Top Pick Wins (1st)
173
27.9%
Top Pick Places (2nd)
114
18.4%
Top Pick Shows (3rd)
93
15.0%
Top Pick ITM
380
61.4%
Pending
16
Exotic Wager Hits (Algorithm Algo1 Winner)
Wager Type
Hits
Description
Exacta
47
Picks 1 & 2 finished 1st & 2nd in exact order
Boxed Exacta
36
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
6
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
10
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Boxed Trifecta
40
Picks 1, 2 & 3 all finished in top 3, any order
Algorithm Algo3 Winner
Metric
Count
Rate
Total Races
721
Top Pick Wins (1st)
185
30.1%
Top Pick Places (2nd)
110
17.9%
Top Pick Shows (3rd)
92
15.0%
Top Pick ITM
387
62.9%
Pending
16
Exotic Wager Hits (Algorithm Algo3 Winner)
Wager Type
Hits
Description
Exacta
55
Picks 1 & 2 finished 1st & 2nd in exact order
Boxed Exacta
31
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
38
Picks 1, 2 & 3 all finished in top 3, any order
Algorithm Algo3 Am Winner
Metric
Count
Rate
Total Races
721
Top Pick Wins (1st)
186
30.3%
Top Pick Places (2nd)
109
17.8%
Top Pick Shows (3rd)
91
14.8%
Top Pick ITM
386
62.9%
Pending
16
Exotic Wager Hits (Algorithm Algo3 Am Winner)
Wager Type
Hits
Description
Exacta
55
Picks 1 & 2 finished 1st & 2nd in exact order
Boxed Exacta
31
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
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.
Metric
Count
Rate
Longshot Selections
636
Races Covered
452
Wins (1st)
25
5.0%
Places (2nd)
33
6.6%
Shows (3rd)
40
8.0%
ITM
98
19.7%
Pending
18
Longshot Betting Strategy Simulator
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: 5803; wider 10/1-and-longer universe available to the edge rows: 14036, of which 9252 could be assigned a calibrated edge. Calibrator: out-of-sample tiered calibrator on the race-normalised longshot score (meta_ls_rank_score_fwd), trained through 2026-06-26; edge rows replay only the 45 days after that date, so no graded race was ever seen by the calibrator. The edge filters use only leak-free forward scores (meta_ls_rank_score_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 45 days (10492 runners), because everything before 2026-06-26 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.
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
366
Trifectas hit
39 (10.7%)
Total staked
$1,404.60
Total returned
$665.35
Profit / loss
−$739.25 (-52.6% ROI)
⭐ Best Bet races only
9 plays, 3 hit (33.3%), −$35.50 (-65.7% ROI)
341 race(s) skipped for scratches inside the top four; 14 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
598
Favorite Wins (1st)
204
34.1%
Favorite Places (2nd)
121
20.2%
Favorite Shows (3rd)
70
11.7%
Favorite ITM
395
66.1%
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.
Metric
Count
Rate
Total Races
616
Top Pick Wins (1st)
188
36.7%
Top Pick Places (2nd)
106
20.7%
Top Pick Shows (3rd)
55
10.7%
Top Pick ITM
349
68.2%
Pending
104
Exotic Wager Hits (Consensus Top 3 Picks)
Wager Type
Hits
Description
Exacta
51
Picks 1 & 2 finished 1st & 2nd in exact order
Boxed Exacta
22
Picks 1 & 2 both finished in top 2, reversed order
Trifecta
16
Picks 1, 2 & 3 finished 1st, 2nd & 3rd in exact order
Boxed Trifecta
34
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
721
173 (28%)
114 (18%)
93 (15%)
380 (61%)
47
36
6
58
Algo 2 Winner
721
181 (29%)
120 (20%)
93 (15%)
394 (64%)
52
39
10
41
Algo 3 Winner
721
185 (30%)
110 (18%)
92 (15%)
387 (63%)
55
31
13
38
AI Longshot
633
25 (5%)
33 (7%)
40 (8%)
98 (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.
