Chapter 17, “Regression to the mean”, (originally referred to as, regression to mediocrity”) illustrates how people mistakenly think that someone (e.g., athlete, interviewee, student, or trainee),  or something (e.g., stocks, horses in a race, business or team performance ) that performed especially well will maintain that performance. And someone or something that performed especially poorly is going to repeat that performance the next time. Without considering the statistical rule of regression to the mean. Regression to the mean is the phenomenon of everything regressing towards the mean, or average. Someone who performs especially well one day, is likely to perform less well on the next day and someone who performs very badly one day, is likely to perform better the next. 

The chapter discusses people’s misinterpretation of this phenomenon as proof that their behavior (e.g., scolding, coaching, additional training) following a poor performance was responsible for the improved performance to follow. And that over confidence in a good performance or pressure to perform, resulted in poorer performance the following time. The truth being that regression to the mean is merely a statistical rule that nature tends to follow and is even evident in genetics and intelligence. Nevertheless, the unaware tend to attribute talent and luck to things that are undeserving. 

Kahneman goes on to explain some of the history and application of regression analyses through the explanation of correlation coefficients and weights of factors. A common error discussed is confusing correlation with causation and “errors of Intuitive prediction” (p.183).  Kahneman then reiterates, that “our mind is strongly biased toward causal explanations and does not deal well with “mere statistics” (p.182). And, “causal explanations will be evoked when regression is detected, but they will be wrong because the truth is that regression to the mean has an explanation but does not have a cause” (p.182).

Side note: When writing about “Talent and Luck”, Kahneman references “John Brockman, who edits the online magazine Edge“. You might want to check it out.

 

Reference

Kahneman, D. (2011). Regression to the Mean. In Thinking Fast and Slow (pp. 156-165). New York, NY: Farrar, Straus and Giroux.