Regression to the mean in football: how AI exploits performance cycles

Regression to the mean is one of the most powerful — and most consistently overlooked — principles in football prediction. Teams that dramatically outperform their underlying xG metrics over a short period (winning more than their shots quality suggests) are statistically likely to 'correct' in subsequent matches. AI models that correctly identify regression candidates produce some of their most consistent edges: backing opponents of overperforming teams, or betting Under on teams whose goal tallies exceed their xG by the widest margin.

Today's football predictions · Leagues · Guides