Chapter 8 Further Reading: Regression to the Mean — Why Hot Streaks Cool Down

Foundational Texts

1. Kahneman, Daniel. Thinking, Fast and Slow (2011). Farrar, Straus and Giroux.

Chapter 17 ("Regression to the Mean") is the single best introduction to this concept for a general audience. Kahneman tells the Israeli Air Force flight instructor story and draws out the full implications for how we understand feedback, causation, and performance evaluation. His Nobel Lecture is also available online and covers similar ground in a condensed format. The broader book is essential for understanding the cognitive biases that make regression to the mean so hard to detect intuitively.

2. Galton, Francis. "Regression Towards Mediocrity in Hereditary Stature." (1886). Journal of the Anthropological Institute of Great Britain and Ireland, 15, 246–263.

The original paper in which Galton documented regression to the mean in height data. Available through JSTOR and various historical archives. The paper is surprisingly readable for a Victorian scientific document and gives you a sense of how Galton reasoned his way toward the discovery — and toward the wrong causal explanation. Reading the original is a valuable exercise in seeing how a brilliant observer can detect a real pattern while misidentifying its mechanism.

3. Mlodinow, Leonard. The Drunkard's Walk: How Randomness Rules Our Lives (2008). Pantheon Books.

Mlodinow's treatment of regression to the mean is especially good on sports examples — batting average fluctuations, coaching effects, and the career arc of athletes. He writes with unusual clarity about why human beings find regression counterintuitive and why the intervention illusion is so powerful. Several chapters directly address the patterns discussed in this chapter.

Key Research Papers

4. Barnett, A.G., van der Pols, J.C., & Dobson, A.J. (2005). "Regression to the Mean: What It Is and How to Deal with It." International Journal of Epidemiology, 34(1), 215–220.

A concise, highly readable introduction to regression to the mean in the medical and epidemiological research context. The paper explains the mathematical mechanism, provides examples from clinical research (where regression to the mean has caused serious errors in evaluating treatment effectiveness), and offers practical suggestions for identifying and correcting for regression effects. Freely available online. Ideal for readers who want a more technical but still accessible treatment.

5. McDonald, I., & Smith, J. (2002). "Regression to the Mean in a Simple Model of Talent." Working Paper.

A relatively accessible formal treatment of regression to the mean in talent-based contexts. The paper develops the two-component model (true ability + random luck) formally and shows how it predicts the patterns observed in sports, academic, and business performance data. Useful for readers who want to understand the mathematics without reading a full statistics textbook.

6. Schall, T., & Smith, G. (2000). "Do Baseball Players Regress Toward the Mean?" The American Statistician, 54(4), 231–235.

An empirical paper examining regression to the mean in Major League Baseball batting statistics. The authors document precisely how much regression occurs from one half-season to the next and from one year to the next, providing concrete numbers for the sport that many readers are most familiar with. A model of clear empirical analysis of a specific performance domain.

The Sophomore Slump Literature

7. Staw, B.M. (1995). "Why No One Really Wants Creativity." In Ford, C.M., & Gioia, D.A. (Eds.), Creative Action in Organizations. Sage Publications.

While not directly about regression to the mean, this essay on why exceptional creative output is followed by conservative second efforts provides the psychological layer that complements the statistical explanation of sophomore slumps. Staw's argument that organizational and social pressure toward the conventional kicks in after a creative breakthrough is a genuine human phenomenon layered on top of the regression mechanism.

8. Surowiecki, James. "The Sophomore Slump." The New Yorker, 2002.

A highly readable journalistic essay on the sophomore slump in business, sports, and entertainment. Surowiecki is one of the best explainers of statistical ideas for general audiences, and this essay does an excellent job of combining the statistical story (regression) with the human story (expectation, pressure, comparison to debut). A good companion to the more technical treatments.

Applied and Practical

9. Smith, Gary. Standard Deviations: Flawed Assumptions, Tortured Data, and Other Ways to Lie with Statistics (2014). Overlook Press.

Smith's chapter on regression to the mean is one of the most practically oriented treatments available. He covers not just sports and business (the standard examples) but also medical interventions, financial advice, and educational programs. His treatment of how regression creates the illusion of effective interventions is especially sharp. A good next step after Kahneman for readers who want more examples and more depth.

10. Abelson, Robert P. Statistics as Principled Argument (1995). Lawrence Erlbaum Associates.

A graduate-level statistics text that approaches statistical concepts through their argumentative and rhetorical functions rather than purely their mathematics. Abelson's treatment of regression to the mean is unusually sophisticated — he shows how it functions as an argument in scientific and practical debates, and how recognizing it changes the burden of proof for causal claims. For advanced readers who want a more philosophical treatment of why regression matters for how we reason.

Online Resources

• The Regression to the Mean Simulator (various R and Python implementations): Search "regression to the mean simulation Python" for multiple freely available implementations that let you watch the phenomenon in action with simulated data.
• Kahneman's Nobel Lecture (2002): Available on the Nobel Prize website. Contains some of the clearest articulations of how regression to the mean relates to judgment and decision-making.
• Baseball Reference (baseball-reference.com): Extensive historical baseball statistics that let you empirically explore regression to the mean in batting averages, ERA, and other statistics.