← All accuracy

Match result accuracy

How often what we forecast on match result matches what actually happened, published unedited — the same numbers the overview shows, just this market alone.

In one sentence

Across 85,308 held-out matches, we are well calibrated on match result, an average gap of -0.0pp between what we said and what happened. The chart below shows each probability band against how often it actually came in.

85,308 matches · model 20260904-0011

What we said, against what happened

What we said against what happened, for Home win0%100%100%0%We said 8%; it happened 10% of the time, across 125 matches.We said 16%; it happened 16% of the time, across 1151 matches.We said 26%; it happened 27% of the time, across 2913 matches.We said 35%; it happened 37% of the time, across 6567 matches.We said 44%; it happened 45% of the time, across 9268 matches.We said 54%; it happened 54% of the time, across 4824 matches.We said 64%; it happened 64% of the time, across 2216 matches.We said 74%; it happened 75% of the time, across 990 matches.We said 84%; it happened 83% of the time, across 360 matches.We said 92%; it happened 100% of the time, across 22 matches.

Each dot is a tenth of the probability range: what we said, against how often it happened. On the dashed line we were exactly right; above it we were too cautious, below it too confident. The rule through each dot is one standard error, and the dot's area is how many matches it rests on — the sparse ones at the ends will always scatter.

Furthest out: when we said 92%, it happened 100% of the time, across 22 matches.

All 12 gate values for this market, unedited
GateValueResult
ece home0.0065pass
max reliability gap pp home7.8184fail
max reliability gap pp home measurable2.3105pass
reliability within sampling envelope home2.3105pass
ece draw0.0055pass
max reliability gap pp draw9.7752fail
max reliability gap pp draw measurable2.5280pass
reliability within sampling envelope draw2.5280pass
ece away0.0094pass
max reliability gap pp away6.2980fail
max reliability gap pp away measurable4.2311fail
reliability within sampling envelope away4.2311pass

Common questions

How accurate are your match-result predictions?

We publish how often the outcome we rated most likely actually happened, decile by decile. The chart shows what we said against what happened, and the summary above states in one sentence whether we were well calibrated, slightly over-confident, or slightly under-confident on this market across the held-out matches we checked.