TrackSmart AI: Cumulative Performance Report - NYRA & Regional Tracks

Reporting Period: December 26, 2025 – June 21, 2026

Track: Belmont at the Big A (BAQ) / Finger Lakes (FL) / Aqueduct (AQU) / Laurel Park (LRL) / Santa Anita Park (SA) / Penn National (PEN) / Saratoga Harness (SARH) / Woodbine (WO) / Saratoga (SAR)

1. Executive Summary: June 21 Update

• Volume Expansion: The tracking dataset has grown to 1036 total races with the addition of 50 races run across Finger Lakes and Belmont at the Big A from June 15 through June
• Strike Rate Resilience: The cumulative Top 4 Strike Rate held perfectly steady at 77.6% (804 for 1036). The AI model turned in a spectacular week, highlighted by 87.5% hit rates at Belmont at the Big A on both June 18 and June 21, and an impressive 6-win top-choice performance on June 20.
• Value Normalization: The Average Payout for AI hits rebounded slightly to ~$7.73 (up minimally from ~$7.70). Heavy value identification anchored the week, catching substantial payouts like Sir Kartrite ($31.18) on June 16, Kunshan Bridge ($25.94) on June 20, and Smiling Rosie ($20.54) on June
• System Integrity and Updates: The automated scratch adjustment protocol was extensively validated this Moving up Alternate selections cleanly absorbed heavy scratch volume on the June 18, 19, and 21 cards, actively securing top-tier hits and keeping multi-race tickets alive with no drop in top-end confidence.

2. Daily Performance Breakdown

(Note: Prior totals reflect cumulative data up to June 14. Daily breakdowns from previous reports have been condensed into the Prev. Totals row. Live scratch protocols dynamically shifted runners up in rankings across multiple races this week, securing numerous top-tier hits.)

3. Winner Distribution by Rank (Cumulative)

4. Cumulative Statistics Summary

5. Aggregate Observations: June 15 - June 21 Insights

• Incredible Top-End Dominance: The AI engine delivered an incredibly sharp performance on June 20 at BAQ, securing six top-choice (1st) winners out of 10 races and returning highly generous overall payouts of $17.04,

$13.82, and $21.58 inside the top tier.

• Pace Projection & Track Bias Accuracy: The model expertly mapped race environments. It correctly isolated "Honest Pace" structures in turf routes and confidently predicted "Contested Duel" meltdowns (e.g., Race 5 at FL on June 15). The AI also successfully utilized track bias overlays ("E/P bias") to lock in rail/pressing
• Deep Alternate Values: The machine's alternate selection logic proved incredibly potent for identifying hidden value. By specifically flagging favorable pace dynamics and "ground-saving inside trips," the AI caught massive overlay winners like Smiling Rosie ($20.54) and Tapwrits Temper ($8.04).
• Chaos Immunity & Vulnerabilities: While the machine comfortably locked onto fair-value longshots, it proved vulnerable to extreme un-modeled Complete meltdowns and massive form reversals by horses like Bandmaster ($71.44) and Into the Light ($33.20) bypassed standard baseline metrics. The AI also operated completely blind to the tote board, missing heavily bet favorites (like Pencima and Probably Flattery) when it aggressively anticipated pace meltdowns that didn't materialize.
• Scratch Protocol Validation: Automated adjustment rules proved indispensable once again. Under heavy late scratch conditions across the week (particularly on June 18, 19, and 21), moving depth selections perfectly filled top-tier This logic successfully captured winners in the adjusted Top 4 every single time a scratch hit the top configurations.

TrackSmart AI: Cumulative Performance Report - NYRA & Regional Tracks

Reporting Period: December 26, 2025 – June 28, 2026

Track: Belmont at the Big A (BAQ) / Finger Lakes (FL) / Aqueduct (AQU) / Laurel Park (LRL) / Santa Anita Park (SA) / Penn National (PEN) / Saratoga Harness (SARH) / Woodbine (WO) / Saratoga (SAR)

1. Executive Summary: June 28 Update

• Volume Expansion: The tracking dataset has grown to 1088 total races with the addition of 52 races run across Finger Lakes and Belmont at the Big A from June 22 through June 28.
• Strike Rate Resilience: The cumulative Top 4 Strike Rate rose slightly to 78% (849 for 1088). The AI model delivered massive consistency throughout the week, highlighted by a flawless 100% Top 4 hit rate at Finger Lakes on June 24 and an astounding 90.9% strike rate at Belmont at the Big A on June 27.
• Value Normalization: The Average Payout for AI hits trended upwards to ~$7.81 (up from ~$7.73). Heavy value identification anchored the week, catching substantial payouts like Paint the Line ($59.24) and Stevie Wonderful ($23.28) on June 23, Open Soul Autism ($24.52) on June 27, and Honor for Mandin ($20.88) on June
• System Integrity and Updates: The automated scratch adjustment protocol was highly lucrative. Field alternates filling vacated Top 4 positions seamlessly absorbed heavy scratch volume, actively securing major overlays directly from bumped positions like Open Soul Autism ($24.52), Salty Ryan ($20.54), and Secured Landing ($6.68) with no drop in top-end confidence.

2. Daily Performance Breakdown

(Note: Prior totals reflect cumulative data up to June 21. Daily breakdowns from previous reports have been condensed into the Prev. Totals row. Live scratch protocols dynamically shifted runners up in rankings across multiple races this week, securing numerous top-tier hits.)

3. Winner Distribution by Rank (Cumulative)

4. Cumulative Statistics Summary

5. Aggregate Observations: June 22 - June 28 Insights

• Flawless & Near-Flawless Consistency: The AI engine delivered incredibly sharp performances all week, securing a perfect 100% Top-4 hit rate at Finger Lakes on June 24 and an outstanding 90.9% on June 27 at BAQ, proving exceptional baseline contender modeling.
• Scratch Adjustments Paid Massive Dividends: Automatically bumping up alternate selections cleanly absorbed extreme late scratch This user-defined logic actively secured major overlays like Open Soul Autism ($24.52), Salty Ryan ($20.54), and Secured Landing ($6.68) directly from bumped positions without missing a beat.
• Potent Value Discovery via Pace & Bias: The model continually extracted high value using structural overlays—identifying extreme 'E' runstyle advantages for Paint the Line ($59.24) and aggressively identifying class plunge overrides for Honor for Mandin ($20.88). The "3YO Phantom Figure Protocol" accurately identified uncapped potential to upset older horses like Stevie Wonderful ($23.28).
• The Power of the 2nd Choice: The AI's #2 ranked selection was the absolute sweet spot on certain cards, striking for 6 wins on June 27 It comfortably faded vulnerable favorites and extracted better value in the secondary slot.
• Chaos Vulnerabilities & Blind Spots: The model remained slightly vulnerable to unmodeled FTS (First-Time Starter) strength from elite barns (e.g., Consensus View missing due to over-penalization) and unexpected off-the-pace meltdowns when front-end chaos materialized outside the model’s thresholds (e.g., Sin Nombre, Roseberns Dream). Expanding early speed/need-the-lead variance rules is recommended to better capture these outliers.

TrackSmart AI: Cumulative Performance Report - NYRA & Regional Tracks

Reporting Period: December 26, 2025 – July 5, 2026

Track: Belmont at the Big A (BAQ) / Finger Lakes (FL) / Aqueduct (AQU) / Laurel Park (LRL) / Santa Anita Park (SA) / Penn National (PEN) / Saratoga Harness (SARH) / Woodbine (WO) / Saratoga (SAR)

1. Executive Summary: July 5 Update

• Volume Expansion: The tracking dataset has grown to 1137 total races with the addition of 49 races run across Finger Lakes and Saratoga from June 29 through July 5.
• Strike Rate Resilience: The cumulative Top 4 Strike Rate maintained a highly consistent 77.9% (886 for 1137). The AI model delivered exceptional precision throughout the week, highlighted by a flawless 100% Top 4 hit rate at Saratoga on July 5 (9 for 9) and at Finger Lakes on July 1 (weather-shortened).
• Value Normalization: The Average Payout for AI hits remained perfectly stabilized at ~$7.81. Deep value identification anchored the week, catching substantial payouts like Neigh Baby ($31.72) on July 5, Heeere's Johnny ($32.20) on July 3, Bettrluckythangood ($16.82) on July 5, and Montauk Surf ($13.66) on June 30.
• System Integrity and Updates: The automated scratch adjustment protocol continued to be extremely Field alternates filling vacated Top 4 positions seamlessly absorbed scratch volume, actively securing major overlays directly from bumped positions like Heeere's Johnny ($32.20), McTigue ($13.24), and Tapizar's Temper ($9.32) with no drop in top-end confidence.

2. Daily Performance Breakdown

(Note: Prior totals reflect cumulative data up to June 28. Daily breakdowns from previous reports have been condensed into the Prev. Totals row. Live scratch protocols dynamically shifted runners up in rankings across multiple races this week, securing numerous top-tier hits.)

3. Winner Distribution by Rank (Cumulative)

4. Cumulative Statistics Summary

5. Aggregate Observations: June 29 - July 5 Insights

• Flawless & Near-Flawless Consistency: The AI engine delivered spectacular performances, securing a perfect 100% Top-4 hit rate at Saratoga on July 5 (9-for-9) and at Finger Lakes on July 1 (2-for-2 before weather cancellation), proving exceptional baseline contender modeling.
• Scratch Failsafe Protocol Perfection: The AI's Alternate Selection protocol was flawless this week. Automatically bumping up alternate selections cleanly absorbed late scratch volume and directly secured massive late-card hits, moving Heeere's Johnny ($32.20), Tapizar's Temper ($9.32), and McTigue ($13.24) into winning spots without missing a beat.
• Pace Mapping & Track Geometry: The AI's flow analysis was remarkably dialed The model confidently mapped "Contested Duels" and "Honest Paces," leaning heavily into ground-saving trips due to marathon inner-turf dynamics (e.g., Bettrluckythangood, $16.82).
• Finding Mid-Tier Value (Chaos Survivors): The model effectively identified outliers mathematically upgraded strictly on their dry dirt TS Speed figures or bypassing the Elite Bench Waiver for extreme layoffs, capturing deep prices like Neigh Baby ($31.72) and Montauk Surf ($13.66).
• Chaos Vulnerabilities & Blind Spots: The AI struggled occasionally on turf sprints when predicting pace meltdowns that failed to materialize, and heavily penalized inside-drawn runners who successfully held position against algorithms (e.g., Allura). Unpredictable multi-horse spills (Saratoga July 3) and unmodeled European shipper gate intent (Kensington Lane) temporarily disrupted Re-evaluating early speed baselines for inside posts and international shippers is recommended to better capture these outliers.

TrackSmart AI: Cumulative Performance Report - NYRA & Regional Tracks Reporting Period: December 26, 2025 – July 12, 2026

Track: Belmont at the Big A (BAQ) / Finger Lakes (FL) / Aqueduct (AQU) / Laurel Park (LRL) / Santa Anita Park (SA) / Penn National (PEN) / Saratoga Harness (SARH) / Woodbine (WO) / Saratoga (SAR)

1. Executive Summary: July 12 Update

• Volume Expansion: The tracking dataset has grown to 1199 total races with the addition of 62 races run across Finger Lakes and Saratoga from July 6 through July
• Strike Rate Resilience: The cumulative Top 4 Strike Rate shifted slightly but maintained a highly consistent 77.4% (928 for 1199). The AI model delivered an exceptional early week at Finger Lakes, highlighted by a flawless 100% Top 4 hit rate on July 8 (8 for 8), while managing extreme variance and chaos results during Saratoga's highly competitive late-week cards.
• Value Normalization: The Average Payout for AI hits remained perfectly stabilized at ~$7.79. Deep value identification continued to anchor the week, capturing substantial payouts like Malibu Smart ($28.20) on July 12, After Taxes (IRE) ($18.72) on July 9, Love Thyself ($14.00) on July 8, and Movin' On Up ($12.62) on July
• System Integrity and Updates: The automated scratch adjustment protocol continued to be extremely lucrative. Field alternates filling vacated positions seamlessly absorbed scratch volume, directly bumping up alternates into winning positions and securing payouts like Movin' On Up ($12.62) on July 11, Karey ($9.44) on July 9, and Carol Said No ($5.20) on July 8 with no drop in top-end confidence.

2. Daily Performance Breakdown

(Note: Prior totals reflect cumulative data up to July 5. Daily breakdowns from previous reports have been condensed into the Prev. Totals row. Live scratch protocols dynamically shifted runners up in rankings across multiple races this week, securing numerous top-tier hits.)

3. Winner Distribution by Rank (Cumulative)

4. Cumulative Statistics Summary

5. Aggregate Observations: July 6 - July 12 Insights

• Finger Lakes Dominance: The AI engine delivered spectacular, near-flawless execution at Finger Lakes, successfully identifying 7-of-8 winners on July 6 and sweeping the card with a perfect 8-for-8 performance on July

8. The model accurately recognized fast dirt profiles and early presser biases while masterfully navigating basement class drops (TPN Prime dominance).

• Scratch Failsafe Protocol Brilliance: The AI's Alternate Selection protocol proved its immense value again this week. Automatically bumping up alternates to fill gaps created by late scratches cleanly absorbed the volume and secured wins for horses like Carol Said No (July 8), Karey (July 9), and Movin' On Up (July 11), maximizing value even when morning-line favorites vanished.
• Pace Mapping & Track Geometry: The AI's flow analysis was incredibly spot-on. "Lone Speed" and "Bias Survivor" models successfully identified extreme inside biases (e.g., Cold Spell and Soaring High on July 9). The pace meltdown logic correctly shielded off-the-pace contenders and successfully identified vulnerable early speed that succumbed to duels throughout the week.
• Saratoga Chaos & Volatility: The model encountered high-variance chaos fields at Saratoga later in the week, particularly with 2-year-old maiden sprints and unexposed turf shippers. While race shape projections (like clustered front flights hitting a stamina wall) were often highly accurate, the AI occasionally fell victim to selecting the wrong beneficiary when deep closers failed to fire and mid-pack stalkers took the prize instead.
• Model Adjustments & Blind Spots: To better capture late-week volatility, adjustments are recommended. Expanding the variance tolerance for mid-priced turf stalkers, relaxing True Speed (TS) ceiling restrictions for lightly raced improvers holding TPN Prime #1 shields, and heavily weighting outside "Clear Air" draws in juvenile sprints will help elevate fringe contenders into top-tier spots during extreme chaos scenarios.