Compare forecasts with outcomes
Divide the number of recorded occurrences by the number of cases. Express that observed frequency as a percentage, then find each forecast’s absolute distance from it. The smaller gap fits this batch better; equally distant forecasts tie.
Forecasts were recorded before the outcomes.
Group: All
Events observed: 42 of 60 · 70%
Forecast A: 60%
Forecast B: 80%
A 60% forecast means a stated chance of 0.60 for each event, not that every event must occur.
These are fictional, completed events. The comparison describes this batch; it does not establish calibration or accuracy on future events.