CORTENA historical percentage cannot be read without knowing how many matches it came from: 3 out of 5 and 300 out of 500 both give 60%, with very different uncertainty. The 95% Wilson interval shows that uncertainty; for 24 out of 40 it is about 44.60%–73.65%. The interval does not tell you the result of a single future match.
The sentence “this result happened in 60% of similar matches” is incomplete without a sample size. 3 out of 5 and 300 out of 500 give the same percentage; the uncertainty is not the same.
Show the numerator and the denominator
First define which matches were selected: date range, league, odds range, missing data and the condition that the match has finished. Then write the figure as occurrences / matches examined. Quietly adding unfinished results to the denominator, or picking only convenient examples, distorts the rate.
What is a 95% Wilson interval?
The Wilson interval is a statistical range that shows the sampling uncertainty of an observed rate. For example, if a feature appears in 24 of 40 observations, the observed rate is 60% and the 95% Wilson interval is roughly 44.60%–73.65%. That does not mean a single future match will land inside this range.
Why be careful even when the percentage is high?
Narrow filters, mismatched matches, missing historical data and leaking information from after the analysed match into the past can all mislead. If the same rule does not give a similar result in other periods, do not draw a conclusion from a high historical percentage alone.
Further reading: Odds analysis guide · Data limits.
Last updated: 28 October 2026 · Responsible: Cortenscore editorial team · Spotted an error? Report a correction. This page is for statistics education; it is not a prediction.