For every bucket of predicted probability, what did the model actually hit? A well-calibrated model has actual win rate ≈ predicted prob. A negative gap means the model is overconfident in that bucket (dangerous); positive gap means it's underpredicting (safe). Rows turn red when |gap| > 5pp AND N ≥ 10.
Combined "all sports" is rarely meaningful — sports differ in market efficiency, signal availability, and base rates. Use the chips to drill into a single sport. N<50 (red) means the calibration is brittle; N≥200 (green) is trustworthy.
Filter: window=90d · sport=tennis_atp
| Predicted-prob bucket | N | Mean predicted | Actual win rate | Gap (actual − predicted) | Brier |
|---|---|---|---|---|---|
| 50-55% | 98 | 51.9% | 59.2% | +7.3pp | 0.246 |
| 55-60% | 201 | 57.8% | 60.7% | +2.9pp | 0.240 |
| 60-65% | 84 | 61.8% | 65.5% | +3.7pp | 0.226 |
| 65-70% | 90 | 67.7% | 68.9% | +1.2pp | 0.212 |
| 70-75% | 45 | 71.3% | 55.6% | -15.8pp | 0.271 |
| 75-80% | 78 | 77.5% | 83.3% | +5.8pp | 0.144 |
| 80-90% | 43 | 84.0% | 88.4% | +4.4pp | 0.102 |
| 90%+ | 20 | 92.0% | 85.0% | -7.0pp | 0.134 |