The statistical principle that extreme performances tend to be followed by more average performances, not because of any causal mechanism but simply due to random variation.
When performance has a random component, extreme outcomes are likely to be followed by less extreme outcomes—not because anything changed, but because extreme results require both skill and luck, and luck doesn't persist. People systematically misunderstand this, attributing regression to causal explanations. Coaches think punishment works because poor performance (which was partly bad luck) is followed by improvement (regression to mean), while praise seems ineffective because great performance (partly good luck) is followed by decline. This leads to spurious causal theories and ineffective interventions. Understanding regression to the mean is essential for interpreting performance data in any domain with randomness.
A mutual fund with exceptional 5-year returns is likely to have more average returns in the next 5 years, not because the manager lost skill, but because the exceptional performance required both skill and luck, and the luck component won't persist. Investors who chase past performance are ignoring regression to the mean.
If performance declines after an extreme high, something must have caused the decline—actually, regression to the mean is a statistical necessity, not a causal phenomenon requiring explanation.
The slow, deliberate, effortful mode of thinking that allocates attention to complex computations, self-control, and conscious reasoning.
Mental ModelThe fast, automatic, intuitive mode of thinking that operates effortlessly and generates impressions, intuitions, and feelings without conscious control.
Mental ModelJudging the frequency or probability of events by how easily examples come to mind, leading to overestimation of vivid or recent events.
Mental ModelJudging probability by how much something resembles a typical case while ignoring base rates, sample size, and statistical principles.
Mental ModelThe tendency to rely too heavily on an initial piece of information (the anchor) when making subsequent judgments, even when the anchor is arbitrary or irrelevant.
Mental ModelThe principle that losses loom psychologically larger than equivalent gains, with losing something feeling roughly twice as bad as gaining the same thing feels good.
PrincipleA descriptive model of decision-making under risk showing that people evaluate outcomes relative to a reference point, are loss-averse, and weight probabilities non-linearly.
FrameworkSystem 1's tendency to construct the most coherent story possible from currently available information without considering what's missing or questions not asked.
PrincipleThe statistical principle that extreme performances tend to be followed by more average performances, not because of any causal mechanism but simply due to random variation.
A mutual fund with exceptional 5-year returns is likely to have more average returns in the next 5 years, not because the manager lost skill, but because the exceptional performance required both skill and luck, and the luck component won't persist. Investors who chase past performance are ignoring regression to the mean.
If performance declines after an extreme high, something must have caused the decline—actually, regression to the mean is a statistical necessity, not a causal phenomenon requiring explanation.
Regression to the Mean is explored in depth in "Thinking, Fast and Slow" by Daniel Kahneman. Distilo provides a deep AI-powered analysis with key insights, audio narration, and practical frameworks.