Model Predictions
| Model | Siena | Manhattan | Spread | Result |
|---|---|---|---|---|
| Elo | 49.7% | 50.3% | - | โ |
| Bradley-Terry | 67.1% | 32.9% | - | โ |
| Bradley-Terry Recency | 65.8% | 34.2% | - | โ |
| Margin | 44.8% | 55.2% | - | โ |
| Margin Recency | 44.5% | 55.5% | - | โ |
| Pythagorean | 53.6% | 46.4% | - | โ |
| Adjusted Efficiency | 67.9% | 32.1% | - | โ |
| Log Adjusted | 68.0% | 32.0% | - | โ |
| Points Off/Def | 41.1% | 58.9% | - | โ |
| Points Off/Def Recency | 40.8% | 59.2% | - | โ |
| Core Ensemble | 56.2% | 43.8% | - | โ |
| Recency Ensemble | 56.1% | 43.9% | - | โ |
| Dynamic Bradley-Terry | 63.0% | 37.0% | - | โ |
| Adjusted Context Blend | 53.7% | 46.3% | - | โ |
Actual: Manhattan +5 (56-51). Win probabilities: out-of-sample forecasts from the weekly fit before the game, the same forecasts as the model pages.