Model Predictions
| Model | Cornell | Army | Spread | Result |
|---|---|---|---|---|
| Elo | 34.3% | 65.7% | - | โ |
| Bradley-Terry | 47.1% | 52.9% | - | โ |
| Bradley-Terry Recency | 46.7% | 53.3% | - | โ |
| Margin | 52.2% | 47.8% | - | โ |
| Margin Recency | 51.9% | 48.1% | - | โ |
| Pythagorean | 59.3% | 40.7% | - | โ |
| Adjusted Efficiency | 69.1% | 30.9% | - | โ |
| Log Adjusted | 69.2% | 30.8% | - | โ |
| Points Off/Def | 51.8% | 48.2% | - | โ |
| Points Off/Def Recency | 51.2% | 48.8% | - | โ |
| Core Ensemble | 54.0% | 46.0% | - | โ |
| Recency Ensemble | 53.9% | 46.1% | - | โ |
| Dynamic Bradley-Terry | 43.4% | 56.6% | - | โ |
| Adjusted Context Blend | 26.3% | 73.7% | - | โ |
Actual: Army -19 (50-69). Win probabilities: out-of-sample forecasts from the weekly fit before the game, the same forecasts as the model pages.