The hot hand is a myth

The hot hand is a myth
	DISPUTED
	"The belief that basketball players get a "hot hand," becoming more likely to make a shot after making several in a row, is a cognitive illusion, as proven by statistical research."
Domain	Sports & Competition
Tactic	Statistical & Evidentiary Distortion;
Era	Current
First seen	1985
Origin	A 1985 study by psychologists Thomas Gilovich, Robert Vallone and Amos Tversky, which became a textbook example of the "hot hand fallacy" in behavioral economics and psychology
Last reviewed	2026-10-27
	Further reading: books on Sports & Competition

The claim that the hot hand is a myth is an unusual entry because it was, for decades, the expert consensus that corrected a popular belief. In 1985, Thomas Gilovich, Robert Vallone and Amos Tversky analyzed shooting data from the Philadelphia 76ers and controlled shooting experiments with Cornell University players, and found no evidence that players were more likely to make a shot after making previous shots. The "hot hand fallacy" became a famous example of how people see patterns in random data. In 2018 economists Joshua Miller and Adam Sanjurjo showed that the original analysis contained a subtle statistical bias, and that when it is corrected, the data show evidence of a hot hand. The status here is Disputed because researchers still debate how large and common the effect is.

The claim

In popular science writing, the hot hand is often presented as a settled example of human irrationality: fans, players and coaches believe in streaks, but the numbers prove streaks are illusions.

What the evidence shows

Original study. Gilovich, Vallone and Tversky, writing in Cognitive Psychology, compared players' shooting percentages after hits and after misses and found no meaningful difference. They concluded that belief in the hot hand reflected a misperception of random sequences.
The bias. Miller and Sanjurjo, in a paper published in Econometrica in 2018, showed that in a finite sequence of coin flips, the expected proportion of heads following a streak of heads is less than 50 percent, because of a selection effect. The original study's method, therefore, expected to see a lower shooting percentage after streaks if there were no hot hand. Finding equal percentages actually suggests a hot hand.
Reanalysis. Applying the correction to the original Cornell experiment data, Miller and Sanjurjo found evidence of a meaningful hot hand effect.
Other studies. Research using detailed NBA tracking data, such as a 2014 analysis by Andrew Bocskocsky, John Ezekowitz and Carolyn Stein, found a small hot hand effect after accounting for the tendency of "hot" players to take harder shots. Studies in other settings, such as controlled shooting contests and darts, have found effects of varying size.
What remains true. People do tend to see streaks in random data, and fans may overestimate how strong the hot hand is. The debate is about size, not whether misperception exists.

Why it persists

The "hot hand fallacy" was a memorable and widely taught example. The statistical error was subtle and went unnoticed for three decades. Debunkings can themselves become dogma.

Lessons for readers

Expert debunkings can also be wrong and deserve scrutiny.
Subtle statistical biases can reverse a conclusion.
Distinguish between the existence of an effect and its size.
Real-world data are complicated by confounding factors, such as shot difficulty.
Science corrects itself, sometimes slowly.

Sources

Thomas Gilovich, Robert Vallone and Amos Tversky, "The hot hand in basketball: On the misperception of random sequences," Cognitive Psychology 17, no. 3 (1985).
Joshua B. Miller and Adam Sanjurjo, "Surprised by the Hot Hand Fallacy? A Truth in the Law of Small Numbers," Econometrica 86, no. 6 (2018).
Andrew Bocskocsky, John Ezekowitz and Carolyn Stein, "The Hot Hand: A New Approach to an Old 'Fallacy'," MIT Sloan Sports Analytics Conference (2014).