TrackSmart AI: Cumulative Performance Report - NYRA Tracks
Reporting Period: December 26, 2025 – March 29, 2026
Track: Aqueduct (AQU)
- Executive Summary: March 29 Update
- Volume Expansion: The tracking dataset has grown to 392 total races with the addition of 25 races run across the March 27, March 28, and March 29 cards.
- Strike Rate Resilience: The cumulative Top 4 Strike Rate settled at 77.6%. After struggling with speed-biased tracks and unexpected pace flows on Friday (50.0%) and Saturday (55.6%), the model rebounded beautifully on Sunday with a 75.0% strike rate.
- Value Normalization: The Average Payout for AI hits adjusted slightly downward from ~$7.79 to ~$7.73. The weekend saw a mix of chalkier 1st-pick hits alongside excellent mid-tier value finds like Sheriff Bianco ($16.12) and Radio Red ($11.32).
- Down-Ballot / Alternate Strength: Because the AI cannot alter its rankings once scratches are announced, bettors rely heavily on the model's static alternates (5th/6th rankings) to slide into the active Top 4. This structural depth proved highly valuable this weekend, catching several winners who were elevated by end-users following late
- 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 |
AQU |
78.7% |
367 |
110 |
73 |
69 |
37 |
~$7.79 |
|
Mar 27 |
AQU |
50.0% |
8 |
1 |
1 |
2 |
0 |
$5.60 |
|
Mar 28 |
AQU |
55.6% |
9 |
4 |
0 |
1 |
0 |
$6.80 |
|
Mar 29 |
AQU |
75.0% |
8 |
2 |
3 |
1 |
0 |
$6.93 |
|
NEW TOTALS |
AQU |
77.6% |
392 |
117 |
77 |
73 |
37 |
~$7.73 |
- Winner Distribution by Rank (Cumulative)
|
AI Rank |
Wins |
% of Hits |
Trend Note |
|
Top Pick (#1) |
117 |
38.5% |
Pace Flow Precision: Dominated the early card on Saturday (4 wins) and added 2 wins on Sunday. This included a direct hit on Pretty Boy Miah ($10.04), aided by a highly accurate "Honest Pace" algorithmic projection. |
|
|
|
|
Sunday Rebound: After a quiet Friday/Saturday (only 1 win), this tier bounced back |
|
2nd Selection |
77 |
25.3% |
with 3 wins on Mar 29, heavily benefiting from a fair track profile that didn't artificially penalize stalkers. |
|
3rd Selection |
73 |
24.0% |
Scratch Beneficiaries: Continues to be a haven for deep-value hits. Snagged Sheriff Bianco ($16.12) on Saturday and Radio Red ($11.32) on Sunday when bettors utilized the AI's alternates to fill Top 4 gaps left by scratches. |
|
4th Selection |
37 |
12.2% |
Cooled Off: Yielded 0 winners over the three-day weekend stretch. The chaotic, deep-closing race flows that this tier usually captures failed to materialize, as front-runners and tactical pressers held firm. |
- Cumulative Statistics Summary
|
Metric Stat Description |
||
|
Overall Accuracy |
77.6% |
(304/392 Races) The AI has successfully placed the winner in its Top 4 selections in 304 of the 392 recorded races. |
|
Top 2 Accuracy |
49.5% |
(194/392 Races) The primary or secondary choice remains highly reliable, continuing to win in roughly half of all modeled races. |
|
Avg Payout |
~$7.73 |
The average payout adjusted downward slightly due to an influx of shorter-priced favorites dominating the early half of the Saturday card. |
- Aggregate Observations: Mar 27 - Mar 29 Insights
- Static Pace Vulnerability (The Scratch Factor): The AI evaluates and projects pace based on the field exactly as it is On Mar 27 (Race 5), the model completely missed a $21.46 winner (Solomini's World) because the primary "Need-the-Lead" horse scratched. The AI could not dynamically recalculate the sudden lack of pace pressure, failing to foresee the winner inheriting an uncontested lead. Corrective Action: Develop an automated secondary "If Primary Speed Scratches" pace scenario flag for end-users. This will allow bettors to manually pivot their strategy when a race's fundamental flow is compromised by a scratch.
- Stalker Penalties Meltdown Projections: The AI heavily downgraded tactical mid-pack horses in anticipated hot paces, missing winners like Hedge Ratio ($7.20) and Master of Arms ($13.52) because it rigidly favored perfect "lone speed" or "deep closer" profiles. Corrective Action: Increase the raw situational weighting for outside pressers/stalkers. Horses that can dictate their own engagement distance should not be as severely penalized in meltdown scenarios as rail-bound stalkers.
- Age-to-Upside Multipliers: On Sunday, the AI missed Tapit Back ($7.30), leaning heavily on established older class-droppers and undervaluing a lightly raced 3-year-old filly facing heavily exposed older Corrective Action: Integrate a stronger age-to-upside developmental multiplier for lightly raced 3-year-olds facing strictly older, exposed maiden claimers.
