TrackSmart AI Weekly Report Card: NYRA & Regional Tracks

Cumulative Performance Scorecard — 2,146 Races and a 91.3% Week

TrackSmart AI | Prepared for Saratogabets | Week Ending September 27, 2026

1. Overview & Methodology

This report card tracks every race TrackSmart AI has handicapped across the NYRA circuit and the regional tracks in the Saratogabets coverage footprint, from December 26, 2025 through September 27, 2026. The tracks represented in the cumulative dataset are 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), Del Mar (DMR), Hollywood Casino at Charles Town Races (CT) and Kentucky Downs (KD).

The headline metric is the Top 4 Strike Rate: the share of races in which the eventual winner was one of the AI’s four ranked selections. Each hit is also credited to the rank the winner held (1st through 4th pick), and the average $2 win payout on those hits is tracked alongside it so accuracy can be read against price.

This week added 69 races run from September 21 through September 27 at Finger Lakes, Belmont at the Big A and Saratoga Harness, bringing the dataset to 2,146 races.

2. This Week at a Glance

Volume Expansion

The 69 new races came from three Finger Lakes cards (September 21–23), two Belmont at the Big A cards (September 24–25) and two Saratoga Harness programs (September 26–27). The dataset now stands at 2,146 total races.

Strike Rate Resilience

The cumulative Top 4 Strike Rate rose again, from 77.5% to 78.0% (1,673 for 2,146). The week itself was the strongest stretch of the season: the winner landed in the Top 4 in 63 of 69 races (91.3%). Finger Lakes produced two perfect cards (September 21 and 23), and Saratoga Harness closed the week at 84.6% and 92.9%.

Value Normalization

Average payout on AI hits moved up from ~$7.92 to ~$7.98. The week’s price horses did the lifting: Wildcat Antonia ($33.08), Baseball Lady ($32.06), Signal Chat ($30.68), Counter Move ($23.30), Signifying Nothing ($20.10) and Double Shot ($17.50) were all captured inside the Top 4.

System Integrity & Scratch Protocol

The automated scratch adjustment was the defining mechanic of the week. Sliding "Next-Man-Up" alternates into vacated Top-4 slots placed Wildcat Antonia, Signal Chat and Signifying Nothing directly into playable structures — three of the week’s six biggest prices.

3. Daily Performance Breakdown

Prior totals reflect cumulative data through September 20; earlier daily breakdowns are condensed into the Prev. Totals row. Late scratches shifted runners up in the rankings on every card this week.

4. Winner Distribution by Rank (Cumulative)

The distribution continues to step down cleanly by rank — 37.5%, 26.5%, 21.9%, 14.0% — which is the shape a well-calibrated ranking should produce. Together the top two picks account for 1,071 wins, or 64.0% of all hits. What stands out this week is where the prices came from: the biggest payouts were spread across all four slots rather than concentrated at the top, and three of them reached the Top 4 only because the scratch protocol promoted them.

5. Cumulative Statistics Summary

6. This Week’s Structural Findings

The Next-Man-Up Rule Pays for Itself

TrackSmart AI posted two perfect Top-4 cards at Finger Lakes this week — 9-for-9 on September 21 and 8-for-8 on September 23. Holding to the mechanical rule of sliding alternates up into the Top 4 after late scratches was directly responsible for catching Wildcat Antonia ($33.08), Signal Chat ($30.68) and Signifying Nothing ($20.10). A simple structural adjustment, applied without second-guessing, turned scratches from a disruption into an edge.

Fair Odds Found the Overlays

The AI’s Fair Odds line identified real value across all three tracks. Even where the model assigned modest absolute win probabilities, it isolated horses that dramatically outran their Morning Lines — Baseball Lady ($32.06), Counter Move ($23.30) and Ideal Amy ($14.02) among them.

Half-Mile Geometry at Saratoga Harness

The model’s emphasis on half-mile track geometry — rail protection and blast-off protocols — was validated across both Saratoga Harness programs. Inside draws and early speed controlled the cards, lining up with the "Golden Chair" and "Engine Seal" projections and producing 84.6% on September 26 and 92.9% on September 27.

Reading the Meltdown at Finger Lakes

At Finger Lakes the AI read pace setups well, anticipating "Meltdown" scenarios and self-destructive speed duels. By elevating horses flagged as chaos beneficiaries or ground-saving trips, it set up wins for off-the-pace closers such as Kozy Bee ($17.32) and Double Shot ($17.50).

7. AI’s Final Analysis — Building the Path Forward

Five takeaways from this week argue for keeping, or leaning harder into, in the weeks ahead:

1. Keep the scratch protocol fully mechanical. Three of the week’s six biggest payouts reached the Top 4 only through the Next-Man-Up slide. The value comes from applying it every time, without manual overrides after scratches are posted.
2. Treat Fair Odds overlays as live Top-4 candidates, not afterthoughts. Baseball Lady, Counter Move and Ideal Amy all won at prices far beyond their Morning Lines. A modest win probability paired with a big overlay is exactly the profile that lifts average payout without costing strike rate.
3. Weight rail and early speed heavily on half-mile harness ovals. The Golden Chair and Engine Seal projections held up across both Saratoga Harness programs; the same geometry logic should carry to other half-mile tracks in the coverage footprint.
4. Trust the Meltdown flag when the pace map shows multiple committed speed horses. At Finger Lakes the chaos-beneficiary and ground-saving closers were where the double-digit prices came from. When the duel is projected, the closer deserves promotion.
5. Expect regression toward the season mark. A 91.3% week sits well above the 78.0% cumulative rate. The right read is that the process is working, not that 90%+ is the new normal — subscribers should be framed the same way.

8. Data Notes & Caveats

• Reporting period: December 26, 2025 – September 27, 2026. This week covers September 21–27 only (69 races at FL, BAQ and SARH).
• Daily breakdowns for weeks before September 21 are condensed into the Prev. Totals row (1,983 races, 77.5%, ~$7.92).
• The 91.3% weekly figure is a Top-4 strike rate (winner among the four selections), not a top-pick win rate. The week’s top pick won 22 of 69 races.
• Average payout is the mean $2 win price on AI hits; the cumulative figure is shown as approximate (~) in the source data.
• Rankings reflect post-scratch order: late scratches moved alternates up on every card this week, and hits are credited to the rank held after those adjustments.
• The dataset mixes thoroughbred and harness racing across multiple circuits; weekly results are sensitive to the track mix and field sizes of the cards run.