Model Predictions
| Model | Rider | Canisius | Spread | Result |
|---|---|---|---|---|
| Elo | 46.3% | 53.7% | - | โ |
| Bradley-Terry | 53.9% | 46.1% | - | โ |
| Bradley-Terry Recency | 55.4% | 44.6% | - | โ |
| Margin | 39.8% | 60.2% | - | โ |
| Margin Recency | 40.8% | 59.2% | - | โ |
| Pythagorean | 53.9% | 46.1% | - | โ |
| Adjusted Efficiency | 55.6% | 44.4% | - | โ |
| Log Adjusted | 55.6% | 44.4% | - | โ |
| Points Off/Def | 39.1% | 60.9% | - | โ |
| Points Off/Def Recency | 40.0% | 60.0% | - | โ |
| Core Ensemble | 50.2% | 49.8% | - | โ |
| Recency Ensemble | 50.4% | 49.6% | - | โ |
| Dynamic Bradley-Terry | 57.1% | 42.9% | - | โ |
| Adjusted Context Blend | 33.0% | 67.0% | - | โ |
Actual: Canisius +2 (85-83). Win probabilities: out-of-sample forecasts from the weekly fit before the game, the same forecasts as the model pages.