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=mma_mixed_martial_arts
| Predicted-prob bucket | N | Mean predicted | Actual win rate | Gap (actual − predicted) | Brier |
|---|---|---|---|---|---|
| <50% | 135 | 40.3% | 35.6% | -4.7pp | 0.229 |
| 50-55% | 71 | 52.5% | 60.6% | +8.1pp | 0.244 |
| 55-60% | 58 | 57.7% | 55.2% | -2.6pp | 0.246 |
| 60-65% | 72 | 62.2% | 61.1% | -1.1pp | 0.241 |
| 65-70% | 60 | 67.0% | 71.7% | +4.7pp | 0.205 |
| 70-75% | 34 | 72.2% | 73.5% | +1.3pp | 0.200 |
| 75-80% | 39 | 77.1% | 74.4% | -2.8pp | 0.191 |
| 80-90% | 39 | 83.3% | 79.5% | -3.8pp | 0.164 |
| 90%+ | 10 | 93.9% | 90.0% | -3.9pp | 0.089 |