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% | 82 | 51.8% | 61.0% | +9.2pp | 0.246 |
| 55-60% | 186 | 57.8% | 59.7% | +1.9pp | 0.241 |
| 60-65% | 60 | 61.0% | 66.7% | +5.7pp | 0.222 |
| 65-70% | 68 | 67.9% | 67.6% | -0.3pp | 0.216 |
| 70-75% | 47 | 71.2% | 59.6% | -11.6pp | 0.254 |
| 75-80% | 78 | 77.5% | 82.1% | +4.6pp | 0.150 |
| 80-90% | 27 | 83.6% | 85.2% | +1.6pp | 0.128 |
| 90%+ | 19 | 92.1% | 89.5% | -2.6pp | 0.096 |