TrackSmart AI: Cumulative Performance Report - NYRA & Regional Tracks

Reporting Period: December 26, 2025 – May 31, 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)

  1. Executive Summary: May 31 Update
  • Volume Expansion: The tracking dataset has grown to 840 total races with the addition of 76 races run across Finger Lakes, Penn National, Santa Anita Park, Saratoga Harness, and Woodbine from May 27 through May
  • Strike Rate Resilience: The cumulative Top 4 Strike Rate held impressively steady at 78.6% (660 for 840). The model secured another flawless 100% accuracy card at Finger Lakes (May 27) and hit dominant ~90% cards at both Santa Anita Park and Saratoga Harness on May 31, successfully translating its logic across dirt, turf, synthetic, and harness racing.
  • Value Normalization: The Average Payout for AI hits stabilized to ~$7.59 (down slightly from ~$7.72). While payouts were depressed in the middle of several chalky cards where logical favorites delivered, excellent value retention was achieved by capitalizing on pace meltdowns to catch overlays like Sinclairity ($11.90) at WO, Quantum Innergy ($10.80) at SA, and Dressed To Kill ($6.60) at
  • System Integrity and Updates: Execution of the dynamic scratch adjustment protocol was highly lucrative this week. The system flawlessly promoted Alternate (4th) selections into the primary Top 3 tiers across multiple slates, securing outright wins from horses like Sunset Storm, Miso Phansy, and Strawberry Treat that would have otherwise fallen off the board.
  1. 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

 

78.5%

 

764

 

221

 

160

 

144

 

75

 

~$7.72

May 27

FL

100.0%

9

4

1

2

2

$5.58

May 29

PEN

70.0%

10

1

2

3

1

$6.63

May 30

SA

66.7%

9

2

1

3

0

$7.73

May 30

SARH

70.0%

10

3

2

2

0

$5.76

May 30

WO

70.0%

10

3

0

1

3

$7.66

May 31

SA

88.9%

9

2

2

3

1

$7.28

May 31

SARH

90.0%

10

5

3

0

1

$4.26

May 31

WO

77.8%

9

5

2

0

0

$5.80

NEW TOTALS

 

MULTI

 

78.6%

 

840

 

246

 

173

 

158

 

83

 

~$7.59

 

(Note: Prior totals reflect cumulative data up to May 25. 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. Finger Lakes May 27 includes a Dead Heat adjustment.)

  1. Winner Distribution by Rank (Cumulative)

 

 

AI Rank

 

Wins

% of Hits

 

Trend Note

 

Top Pick

 

246

 

37.3%

Pace Flow Precision: Saw a massive surge in accuracy on synthetic (Woodbine) and harness tracks (Saratoga) by successfully locking onto undisputed inside speed and wire-to-wire profiles.

 

 

2nd Selection

 

173

 

26.2%

Class Elevation & Adaptability: Maintained consistency, particularly excelling at finding pocket trips ("The Golden Chair") and logical stalkers that capitalized on front-end pressure.

3rd Selection

 

158

 

23.9%

Alternate Protection: Routinely caught off-the-pace sweepers and highly contested dirt sprint survivors, specifically diagnosing destructive early duels.

 

4th Selection

 

83

 

12.6%

Rebounding Safety Net: Provided critical exotic value and outright wins this week as the built-in scratch protection vaulted unexposed runners and algorithmic alternates into the matrix.

 

  1. Cumulative Statistics Summary

 

Metric         Stat         Description

Overall Accuracy

 

78.6%

(660/840 Races) The AI has successfully placed the winner in its Top 4 selections in 660 of the 840 recorded races.

Top 2 Accuracy

 

49.9%

(419/840 Races) The primary or secondary choice remains highly reliable, winning in nearly half of all modeled races.

 

Avg Payout

 

~$7.59

The average payout trended slightly down this week due to chalky cards. However, value retention remained strong via mid-price chaos saves and excellent lone-speed reads on all-weather and harness surfaces.

 

  1. Aggregate Observations: May 27 - May 31 Insights
  • The Power of the Alternate (Scratch Protection): The automated scratch adjustment rule proved massively valuable this As massive favorites or top selections scratched out, the AI vaulted algorithmic alternates into the Top 3 slots. This mechanical promotion directly salvaged multi-race exotics by catching winners like Dressed To Kill ($6.60), Strawberry Treat ($6.80), Sunset Storm ($3.40), and Miso Phansy ($4.80).
  • Track Bias & Geometry Capitalization: The model was incredibly dialed in on specific track geometries this At Saratoga Harness, the AI heavily rewarded inside draws and pocket trips ("The Golden Chair"), accurately predicting wire-to-wire runs. At Woodbine, the system successfully weaponized the all-weather synthetic bias, locking onto horses holding a raw algorithmic speed advantage from inside posts.
  • Pace Meltdowns and Duel Projections: The flow algorithms performed exceptionally well when reading dirt and synthetic sprints. The system perfectly diagnosed contested early fractions and destructive early duels, successfully cashing on off-the-pace stalkers and sweepers like Sinclairity ($11.90) at Woodbine and various late-kick beneficiaries at Santa Anita.
  • Turf Route Blind Spots: While the model hit high-priced turf winners like Bandiagara (GB) for $14.40 at Penn National, 1-mile turf routes remain a notable The AI struggled by aggressively favoring deep closers with high stamina ratings on firm turf that ultimately held early and tactical speed far better than historical pars suggested. Re-evaluation of the "turf route penalty" algorithm is needed so inside speed and outside-stalking profiles are not automatically downgraded for distance.

THE MACHINE (AKA) TrackSmart AI isn’t a tip sheet or a picks website — it’s a professional-grade handicapping engine that works directly with the Past Performances (PPs) you already use from DRF, Brisnet, and Australia's Racing and Sports. Each version of TrackSmart AI — DRF, Brisnet, Harness, and International — is custom-engineered to read and interpret every printed detail in those PPs. No guessing. No databases. No shortcuts. The AI performs the deep work for you — comparing figures, detecting trainer intent, modeling pace, and uncovering hidden value plays — with the speed and accuracy of a professional analytics team. You stay in control of your handicapping while AI handles the heavy lifting. Pace projections, class and form analysis, TPN rankings, and betting strategy — all built from the same data you’d normally spend hours studying.