Hot-Hand Fallacy

Category: Probability & Belief

The belief that a person on a streak of successes is "hot" and more likely to succeed on the next attempt, when the outcomes are actually independent.

How it works

Your brain is a pattern detector with the sensitivity dial cranked too high, so a run of four reads as a signal even when it is noise. The root cause is what Tversky and Kahneman called belief in the law of small numbers: you expect a tiny sample to mirror the "random" texture of a large one, so genuine randomness (which is lumpy and streaky) looks suspiciously non-random to you. When independent events cluster, as they always eventually do, you retrofit a cause: he is locked in, she is on fire, the machine is due. The kicker is that the effect is partly a measurement illusion too. Miller and Sanjurjo showed in 2018 that even the study which "debunked" the hot hand had a hidden counting bias, so the human intuition was closer to right than the experts admitted for thirty years.

Where you'll see it

  • Basketball itself. Gilovich, Vallone, and Tversky (1985) analyzed the Philadelphia 76ers and found a make after three straight makes was no more likely than after three straight misses. Players, coaches, and 91% of fans surveyed were certain the hot hand was real. Feed the ball to the guy who is 'hot' and you are often just feeding a coin that landed heads a few times.
  • Slot machines and their 'due' payout. Casino players pump money into a machine that just paid out, or one that has gone cold for hours, convinced a streak is building or breaking. Every spin is independent by law and by design. The house profits directly from your pattern-seeing.
  • Mutual fund 'star' managers. Investors pile into last year's top-performing fund because the manager is on a run. Morningstar's own research and the SEC-mandated 'past performance does not guarantee future results' disclaimer exist because hot streaks in fund returns are mostly noise, and money that chases them reliably underperforms.
  • Hiring the 'closer' on a streak. A salesperson lands five big deals in a row, so leadership fast-tracks them for the hardest account. Some of that is skill, but a large chunk is a favorable run of leads and timing being misread as an unbreakable hot streak, right before regression to the mean shows up.

Where it comes from

Coined by Thomas Gilovich, Robert Vallone, and Amos Tversky in their 1985 paper "The Hot Hand in Basketball: On the Misperception of Random Sequences" (Cognitive Psychology). Analyzing Philadelphia 76ers field goals, Boston Celtics free throws, and a controlled Cornell shooting experiment, they found makes and misses were statistically indistinguishable from coin flips, and concluded the hot hand was a cognitive illusion. The intellectual root is Tversky and Kahneman's 1971 "Belief in the Law of Small Numbers." In a major 2018 twist, Joshua Miller and Adam Sanjurjo (Econometrica) exposed a selection bias in the original counting method, and correcting it reversed the canonical result: a real, modest hot hand does exist in shooting data. So the "fallacy" is subtler than the textbook version. People overestimate streakiness, but streaks are not pure myth either.

How to counter it

Ask the independence question first. Before you bet on a streak, ask literally: does the previous outcome change the odds of the next one? For a roulette wheel or a slot machine, the answer is a flat no, and the streak is meaningless. For a basketball shooter or a salesperson, there may be a small real effect, but it is smaller than it feels.

Anchor on the base rate, not the last five results. A 45% shooter is a 45% shooter whether he just hit four or missed four. Write down the long-run rate before the streak seduces you, and weight it far more heavily than the recent run.

Assume regression to the mean is coming. Whoever is on the hottest streak is the person most likely to cool off next, purely by statistics. When you catch yourself doubling down on the current leader, that is precisely the moment to size the bet down.

The tell

You catch yourself saying "he's due" or "she's on fire" and treating it as a reason to act, when nothing about the underlying odds has actually changed. If your confidence spiked because of the last three results rather than the long-run rate, you are riding the hot hand.

Related biases

References

  1. Thomas Gilovich, Robert Vallone, Amos Tversky (1985). The Hot Hand in Basketball: On the Misperception of Random Sequences. Cognitive Psychology, 17(3), 295-314
  2. Amos Tversky, Daniel Kahneman (1971). Belief in the Law of Small Numbers. Psychological Bulletin, 76(2), 105-110
  3. Joshua B. Miller, Adam Sanjurjo (2018). Surprised by the Hot Hand Fallacy? A Truth in the Law of Small Numbers. Econometrica, 86(6), 2019-2047
  4. Thomas Gilovich (1991). How We Know What Isn't So: The Fallibility of Human Reason in Everyday Life. The Free Press
  5. Alan D. Castel, Aimee Drolet Rossi, Shannon McGillivray (2012). Beliefs About the 'Hot Hand' in Basketball Across the Adult Life Span. Psychology and Aging, 27(3), 601-605