Hit rate and base rate
"Right 50% of the time" sounds like something. Whether it is depends on one other number: how often it would have happened anyway.
The hit rate
A signal fires; afterwards, the thing it points to happens or not. The share of times it happens is the hit rate. Our alerts point to a team scoring again, so their hit rate is how often the alerted team scored later in the match.
The base rate
Teams score in most matches even without any signal. From the 30th minute, about two teams in three score again before the end; from the 75th, about one in four. That is the base rate: what happens anyway, at the same minute and with the same score.
The edge
The edge is the hit rate minus the base rate. In Xgoals' record for the last 30 days (29 October 2026): the alerted team scored later in 50.5% of alerts, and a team in the same minute and score would have scored anyway in 45.3%: an edge of about five points. From the 76th minute on the edge was nil: 19.7% against 21.3%.
Sample size
Over a thousand alerts, an edge of five points is about three times what chance alone moves. Over sixty, it can be chance from start to finish: a group of sixty can swing by ten points without meaning anything. That is why our record shows how many alerts stand behind every figure, and why we do not pick the best-looking corner of the data after the fact.
Three questions for any signal
- How often would it have happened anyway?
- How many cases stand behind the number?
- Was the rule fixed before the results, or chosen after looking at them?
Questions
Is a 50% hit rate good?
Only if the base rate is lower. If teams score again half the time anyway, a 50% signal adds nothing.
Why not show only the alerts that worked?
Because then any signal looks good. A record is only worth something if it keeps every case.
Sources
Checked 2026-10-28. Methods change: every figure here is dated to its source.
xG explains football; it is not a betting tip. 18+ · Please gamble responsibly.