Model Predictions
| Model | Toledo | Marshall | Spread | Result |
|---|---|---|---|---|
| Elo | 49.9% | 50.1% | - | โ |
| Bradley-Terry | 56.5% | 43.5% | - | โ |
| Bradley-Terry Recency | 55.4% | 44.6% | - | โ |
| Margin | 43.6% | 56.4% | - | โ |
| Margin Recency | 44.2% | 55.8% | - | โ |
| Pythagorean | 51.4% | 48.6% | - | โ |
| Adjusted Efficiency | 62.3% | 37.7% | - | โ |
| Log Adjusted | 62.4% | 37.6% | - | โ |
| Points Off/Def | 44.5% | 55.5% | - | โ |
| Points Off/Def Recency | 45.0% | 55.0% | - | โ |
| Core Ensemble | 51.1% | 48.9% | - | โ |
| Recency Ensemble | 51.2% | 48.8% | - | โ |
| Dynamic Bradley-Terry | 46.6% | 53.4% | - | โ |
| Adjusted Context Blend | 36.1% | 63.9% | - | โ |
Actual: Marshall +15 (72-57). Win probabilities: out-of-sample forecasts from the weekly fit before the game, the same forecasts as the model pages.