MLB Performance
Windowed live grades by market, plus persisted model evaluation evidence.
Winners
Open market| Model | Market | Sample | W-L-P | Hit rate | Flat -110 ROI |
|---|---|---|---|---|---|
| v3 | winner | 14 | 8-6-0 | 57.1% | n/a |
Home runs
Open marketHit rates are split by top-k bucket and model version.
| Model | Market | Sample | W-L-P | Hit rate | Flat -110 ROI |
|---|---|---|---|---|---|
| mlb-hr-v1 | home_run top_10 | 86 | 16-70-0 | 18.6% | 13.8% |
| mlb-hr-v1 | home_run top_25 | 132 | 20-112-0 | 15.2% | 9.1% |
| mlb-hr-v1 | home_run top_50 | 220 | 39-181-0 | 17.7% | 30.1% |
| mlb-hr-v1 | home_run field | 554 | 77-477-0 | 13.9% | 17.2% |
Persisted backtests
Evaluation runs stored in Supabase, separate from live grading.
Evaluation runs
| Evaluation | Model | Test range | Accuracy | AUC | Brier / log loss | ROI | Status |
|---|---|---|---|---|---|---|---|
| performance_history_mlb_2026_ytd_moneyline | v3 | n/a–n/a | n/a | n/a | n/a / n/a | n/a | candidate |
Accuracy over time