Regression to the Mean
Where a measurement combines a stable component with a random one, extreme readings are extreme partly by luck, and that part does not repeat, so the measurement moves back.
What it means
Nothing has improved or declined. The measurement has shed the component that was never going to repeat, which in football is usually a conversion spike.
Why it matters
It is the reason a record season is unlikely to be repeated exactly and why a side that finished far above its underlying numbers tends to fall back.
Where it is used
Reading any record season, run of form or exceptional individual campaign.
Where it misleads
It is not a prediction about a specific case. It describes what happens on average and says nothing about which team you have, so converting it into a forecast about one club is a misuse.
Key points
- Rates regress; volumes generally do not.
- The further from the mean, the further a reading tends to travel back.
- Use it to decide which numbers to trust, not to forecast.
- If chance volume rose with the output, the improvement is real.
- League position regresses because tables rest on one-goal margins.
- It describes an average case, never a specific one.
The metric is a measurement with a denominator, and the denominator is what makes the figure comparable rather than merely large.