TrackSmart AI: Cumulative Performance Report - NYRA Tracks
Reporting Period: December 26, 2025 – March 15, 2026
Track: Aqueduct (AQU)
- Executive Summary: March 15 Update
- Volume Expansion: The tracking dataset has grown to 335 total races with the addition of 32 races run across four cards from March 12 to March 15.
- Strike Rate Resilience: The cumulative Top 4 Strike Rate held strong at 78.8%. The model was incredibly consistent traversing difficult weather and track conditions this week, hitting on 78.1% of its races overall (25 of 32), and peaking at an exceptional 87.5% strike rate on March 15 and 85.7% on March 12.
- Value Normalization: The Average Payout for AI hits saw a slight downward shift from ~$7.87 to ~$7.75. This was driven by a heavy concentration of logical favorites hitting on the off-tracks on March 12 and March 13, though it was offset by catching double-digit upsets later in the week, including a massive $24.00 winner on March 14 and a
$16.02 winner on March 15.
- Alternate/Scratch Protection Strength: The AI's alternate picks and scratch logic were the unequivocal MVPs this week. Late scratches mechanically elevated several winners into the active selections, seamlessly catching K Gun ($24.00) as an alternate, and automatically moving horses like Celestial Body, Sharp Spark, Monte Avi, and Ez Roll into upgraded winning positions.
- 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 |
78.9% |
303 |
98 |
55 |
54 |
32 |
~$7.87 |
|
Mar 12 |
AQU |
85.7% |
7 |
1 |
3 |
1 |
1 |
$5.56 |
|
Mar 13 |
AQU |
75.0% |
8 |
2 |
3 |
1 |
0 |
$4.63 |
|
Mar 14 |
AQU |
66.7% |
9 |
3 |
1 |
2 |
0 |
$8.62 |
|
Mar 15 |
AQU |
87.5% |
8 |
1 |
3 |
3 |
0 |
$7.28 |
|
NEW TOTALS |
AQU |
78.8% |
335 |
105 |
65 |
61 |
33 |
~$7.75 |
- Winner Distribution by Rank (Cumulative)
|
AI Rank Wins % of Trend Note Hits |
|||
|
Top Pick (#1) |
105 |
39.8% |
Volume Leader: The top pick cooled slightly compared to last week's absolute dominance (contributing 7 winners), allowing secondary selections to pick up the slack. Still accounts for nearly 40% of all AI hits. |
|
2nd Selection |
65 |
24.6% |
Secondary Surge: A massive week for the second choice, picking up 10 winners across the four cards, showcasing the model's depth in successfully isolating the top two highest-probability race flows. |
|
3rd Selection |
61 |
23.1% |
Alternate Elevation: Bounced back aggressively after a quiet prior week, adding 7 winners. This slot was significantly bolstered by the dynamic scratch protocol pulling deep value (like $24.00 K Gun) into the active tier. |
|
|
|
|
|
|
4th Selection |
33 |
12.5% |
The Safety Net: Contributed only 1 winner this week (Natural Hunk), reflecting that the top 3 selections captured the vast majority of predictable race flows without needing the basement picks. |
- Cumulative Statistics Summary
|
Metric Stat Description |
||
|
Overall Accuracy |
78.8% |
(264/335 Races) The AI has successfully placed the winner in its Top 4 selections in 264 of the 335 recorded races. |
|
Top 2 Accuracy |
50.7% |
(170/335 Races) The primary or secondary choice continues to win over half of all races modeled, proving exceptional long-term reliability. |
|
Avg Payout |
~$7.75 |
The average payout adjusted slightly downward as chalky/logical runners held an advantage over off-track conditions early in the week. |
- Aggregate Observations: Mar 12 - Mar 15 Insights
- Scratch Protocol Dominance (All Week): The dynamic scratch and alternate adaptation logic was By mechanically moving the next highest True Pace Number (TPN) horses up to replace scratched runners, the AI salvaged cards that were decimated by late changes. This successfully nailed a $24.00 longshot (K Gun) as a promoted alternate, while directly leading to top-tier hits on Celestial Body, Monte Avi, Apalta, and Ez Roll.
- Off-Track Model Robustness (Mar 12 & Mar 15): The Bris-Core pacing models adjusted exceptionally well to muddy/sealed anomalous conditions. The AI effectively mapped early-speed track biases, prioritizing "Lone Speed" and "Alpha Clearance" setups to record an 7% accuracy rate on Thursday and 87.5% on Sunday.
- Over-Projecting Pace Meltdowns (Mar 13): The system outsmarted itself on Friday by aggressively anticipating destructive pace It penalized speed heavily on what turned out to be a speed-carrying "Good" track, leading to complete misses on horses like Artemis Sound and Nabokov. Corrective Action: Scale back "meltdown" certainty when speed horses draw favorable outside tracking paths or when returning off layoffs for elite trainers.
- Tactical Pressers Deep Closers (Mar 14): The AI occasionally overvalued deep closers at the expense of tactical mid-pack stalkers in routes where hot paces were projected (resulting in misses like Centavo and Three Little Birds). Corrective Action: Refine stamina algorithms for 7-furlong to 1-mile setups to better upgrade outside-drawn mid-pack runners who can avoid interior speed duels without dropping to the deep rear.
