TrackSmart AI: Cumulative Performance Report - NYRA & Regional Tracks
Reporting Period: December 26, 2025 – May 10, 2026
Track: Belmont at the Big A (BAQ) / Finger Lakes (FL) / Aqueduct (AQU)
- Executive Summary: May 10 Update
- Volume Expansion: The tracking dataset has grown to 610 total races with the addition of 44 races run across Finger Lakes and Belmont at the Big A from May 5 through May
- Strike Rate Resilience: The cumulative Top 4 Strike Rate held strong at 78.9% (481 for 610), following a solid 75% accuracy week (33 for 44). The model successfully absorbed variable track conditions and heavy scratching, highlighting an 87.5% strike rate on the May 7 card.
- Value Normalization: The Average Payout for AI hits settled slightly to ~$7.55 (down from ~$7.65). Formful speed bias at Finger Lakes and logical payouts at BAQ slightly depressed the daily averages, though the model still comfortably isolated value hits like Superpower ($13.48) and Irish Maxima ($14.44).
- Weather and Pace Adjustments: Significant mid-week weather shifts revealed an algorithmic over-sensitivity to "Pace Meltdowns" on downgraded/sealed surfaces. Adjustments to tactical speed weighting and class-drop constraints have been noted for future patches to combat chaotic turf sprints and moisture-retaining dirt
- Daily Performance Breakdown
|
Date |
Track |
Top 4 Strike Rate |
Total Races |
1st Pick |
2nd Pick |
3rd Pick |
4th Pick |
Avg Payout (AI Hits) |
|
Prev. Totals |
MULTI |
79.2% |
566 |
171 |
118 |
106 |
53 |
~$7.65 |
|
May 5 |
FL |
75.0% |
8 |
3 |
2 |
1 |
0 |
$5.42 |
|
May 7 |
BAQ |
87.5% |
8 |
2 |
2 |
1 |
2 |
$6.20 |
|
May 8 |
BAQ |
75.0% |
8 |
1 |
2 |
2 |
1 |
$6.00 |
|
May 9 |
BAQ |
72.7% |
11 |
2 |
4 |
0 |
2 |
$7.10 |
|
May 10 |
BAQ |
66.7% |
9 |
2 |
2 |
1 |
1 |
$5.68 |
|
NEW TOTALS |
MULTI |
78.9% |
610 |
181 |
130 |
111 |
59 |
~$7.55 |
(Note: Prior totals reflect cumulative data up to May 3. 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.)
- Winner Distribution by Rank (Cumulative)
|
AI Rank Wins % of Trend Note Hits |
|||
|
Top Pick (#1) |
181 |
37.6% |
Pace Flow Precision: Maintained consistency, perfectly identifying forwardly placed runners (e.g., Finger Lakes "Lone Speed" profiles) and capitalizing on logical overlays and sharp class drops. |
|
2nd Selection |
130 |
27.0% |
Class Elevation & Adaptability: Showed immense strength this week, securing 12 wins. Continues to capitalize on class plungers and tactical stalkers when the primary pace setup softens. |
|
3rd Selection |
111 |
23.1% |
Alternate Protection: Highly consistent as a safety net. Automated scratch adjustments mechanically moved alternates directly into winning slots for crucial hits. |
|
4th Selection |
59 |
12.3% |
Rebounding Safety Net: Provided massive value consistency, accurately catching chaotic pace-meltdown beneficiaries and keeping long-layoff survivors on the fringe of multi-race exotics. |
Cumulative Statistics Summary
|
Metric Stat Description |
||
|
Overall Accuracy |
78.9% |
(481/610 Races) The AI has successfully placed the winner in its Top 4 selections in 481 of the 610 recorded races. |
|
Top 2 Accuracy |
51.0% |
(311/610 Races) The primary or secondary choice remains highly reliable, winning in more than half of all modeled races. |
|
Avg Payout |
~$7.55 |
The average payout remained steady. While chalk-heavy predictable cards and formful speed profiles slightly depressed daily averages, finding chaotic winners protected strong long-term ROI. |
Aggregate Observations: May 5 - May 10 Insights
-
- Pace Meltdown & Weather Bias Conflicts: The AI systematically down-ranked front-runners on May 9 anticipating pace collapses, but the mid-card weather shift to a Good/Sealed track allowed early and outside speed to carry wire-to-wire.
- Tactical Mid-Pack Stalkers vs. Deep Closers: When predicting pace meltdowns or early speed duels, the AI occasionally undervalued the tactical mid-pack stalkers (e.g., Remi's Moon, Grace and Grit) that got the first jump before the deep closers could Corrective Action: Adjust pace meltdown algorithms to afford more weight to tactical off-the-pace stalkers when track friction allows them to sustain early moves.
- "3YO Phantom Protocol" Reliability: Upgrading lightly-raced 3YOs against older horses paid excellent dividends (e.g., Mermaid paying $11.02), proving this remains a highly profitable angle However, severe rigid penalty constraints for extreme layoff horses (e.g., 0-for-16 older maidens like Noble One) need slight softening when they are dropping into bottom-tier claiming company, as weak fields can occasionally trigger sudden wake-ups.
- Scratch Adjustments Functioned Smoothly: Automated contingency rules to bump alternate selections up following late scratches remained entirely It effectively elevated alternate selections (e.g., Soaring Spirit, Point of Reference) into secondary matrix slots to salvage exotic ticket payouts.
