Calibration
When the model says 60%, it should happen about 60% of the time. Each dot is a bucket of predictions; the diagonal is perfect calibration.
Rolling RPS
Ranked Probability Score over a trailing 30-day window. Lower is better; the dashed line is the baseline window.
Season by season
Long-run advantage
A season-path Monte Carlo. Whole seasons are resampled 2,000 times to check whether the edge over the baseline would have held in any season, not just this one.
Every figure here is produced by a scheduled job from the archived prediction log and the raw match results — the same numbers the model itself is judged on.