Benchmarks
How models compare on backtest accuracy.
Published models use the same scoring contract: Brier score, hit rate, calibration, against-the-spread rate, and drawdown. These rows are pulled from the live model registry and graded prediction results.
Registry-backed leaderboard
Last generated: Oct 18, 2026, 8:22 PM
Sport
Benchmark summary
Models
12
Published models with enough scored predictions to rank in this view.
Scored picks
9,817
Rows carrying a usable probability for benchmark scoring.
Best Brier
0.180
Lower Brier score is better. It measures probabilistic forecast error.
Best ATS
75.0%
Highest against-the-spread win rate in this filtered view; small samples vary a lot.
Sports
3
Sports represented by ranked benchmark rows.
Window
365d
Rolling lookback window used for the benchmark summary.
| Rank | Model | Sport | Brier | ATS | Scored |
|---|---|---|---|---|---|
| #1 | House NFL Bridge - Preseason Prior (Elo) - Moneyline | NFL | 0.180 | 75.0% | 20 |
| #2 | House NFL Moneyline | NFL | 0.229 | 64.9% | 37 |
| #3 | MLB Player Strikeouts (Poisson) | MLB | 0.235 | 66.3% | 89 |
| #4 | House NHL Moneyline | NHL | 0.238 | 60.0% | 40 |
| #5 | MLB Moneyline (Ridge Classifier) | MLB | 0.241 | 58.5% | 49 |
| #6 | NHL Moneyline Classifier (Random Forest) | NHL | 0.243 | 58.4% | 185 |
| #7 | NHL Moneyline Classifier (Random Forest) | NHL | 0.243 | 58.4% | 185 |
| #8 | MLB Moneyline Classifier (Random Forest) | MLB | 0.243 | 58.0% | 162 |
| #9 | MLB Moneyline (TabPFN) | MLB | 0.243 | 57.4% | 162 |
| #10 | MLB Moneyline (CatBoost) | MLB | 0.245 | 56.2% | 162 |
| #11 | House NHL Puck Line | NHL | 0.246 | 59.8% | 112 |
| #12 | NHL Moneyline (CatBoost) | NHL | 0.247 | 58.4% | 185 |
Metric glossary
- Brier score
- Mean squared error of probabilistic forecasts. Lower is better.
- Log-loss
- Penalizes overconfident wrong answers harder than Brier. Used for binary classifiers.
- Hit rate
- Win percentage excluding pushes. Breakeven against -110 juice is about 52.4%.
- ATS
- Against-the-spread win rate for scored spread picks.
- CLV
- Closing-line value: average edge captured against the final market price.
- Max drawdown
- Worst peak-to-trough loss in units across the scored run.
What gets audited next
- Replayable runs: every benchmark row links back to immutable model versions and scored prediction windows.
- Leakage checks: validation kits runs that train on the same rows they claim to evaluate.
- Verified badges: signed benchmark runs can be surfaced on model cards and leaderboard rows.