# Regression to the Mean

> After a wild high or low score, the next one tends to be less wild.

**Field:** General Knowledge  
**Category:** Math & Statistics  
**Also known as:** Regression Toward the Mean, Reversion to the Mean

## Think of it like

It is like your best day at bowling. Next time, you will most likely score close to your usual.

## Definition

Regression to the mean is when an extreme result is likely to be followed by one closer to the average. A very high or very low score is often part skill and part luck. The luck part does not repeat, so the next score drifts back toward normal. Nothing has to change for this to happen. People often credit or blame the wrong cause for the drift.

## Example

Jack scores 95 on a test when he usually gets 75. On the next test he gets 80. His teacher thinks he got lazy, but most of the drop is just luck not repeating.

## Why it matters

It fools people into thinking praise makes things worse and blame makes them better. Good studies use a control group to see past it.

## Common mistake

Giving credit to whatever came just before the drift. A team that fires its coach after a bad run often improves anyway.

## Three principles

- **Skill plus luck.** Extreme results have extra luck in them.
- **Luck fades.** The next result drifts back to normal.
- **Wrong causes.** Do not credit or blame what came just before.

## Related terms

- [Mean (Average)](https://learn.benibix.com/term/general/mean)
- [Regression Analysis](https://learn.benibix.com/term/general/regression-analysis)
- [Control Group](https://learn.benibix.com/term/general/control-group)
- [Post Hoc Fallacy](https://learn.benibix.com/term/general/post-hoc-fallacy)
- [Confounding Variable](https://learn.benibix.com/term/general/confounding-variable)

## Told apart from

- [Regression to the Mean vs Regression Analysis](https://learn.benibix.com/compare/regression-to-the-mean-vs-regression-analysis): People think regression to the mean is a step in regression analysis, and it is not. Regression analysis is a method. It fits a line to data to see how one thing moves with another. Regression to the mean is an effect. A very high or low score tends to be followed by one closer to normal, because the luck in it does not repeat. One is a tool. The other is a trap that a fair test helps you spot.

## Taught in

- [Law of Large Numbers](https://learn.benibix.com/learn/general-math/law-of-large-numbers): Statistics, 3/10

![Four panels. Law of large numbers: a line for the share of heads in coin flips jumps around at first, then settles close to 50 percent as the flips pile up. Gambler's fallacy: five heads in a row, and the next flip still shows a 50 percent chance of heads. Regression to the mean: a first test score far above the average, and a second score closer to the average. Elo rating system: a player rated 1600 meets one rated 1400, with an expected score of about 0.76 and 0.24; a win moves the stronger player up by about 5 points, and an upset moves the weaker one up by about 15.](https://learn.benibix.com/diagrams/gen6-chance-in-practice.svg)

---

Source: https://learn.benibix.com/term/general/regression-to-the-mean
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-20
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
