A basketball player hits four shots in a row, and the crowd, the commentators and the teammates all agree: she is on fire. The next pass goes to her. This is the hot hand, the belief that success makes further success more likely. It is one of the most discussed ideas in the psychology of chance, and it is also one of the most misunderstood, because the honest answer is more nuanced than either “it is a myth” or “it is real”.
The idea, and the original research
The hot hand is a claim about sequences: after a success, the chance of another success is higher than usual. A widely cited study of basketball shooting in the 1980s looked at sequences of made and missed shots and found little evidence that a made shot raised the chance of the next one. The authors concluded that the belief was a cognitive illusion, and for many years “the hot hand fallacy” became the standard label.
Why the debate did not end there
Later researchers re-examined the question and raised several points, which are worth stating in general terms because they show how hard the problem is.
- The statistical method matters. Some later work argued that a common way of analysing short sequences contains a subtle bias that can hide a modest real effect. Corrected analyses suggested that small hot-hand effects may exist in some settings.
- Real conditions change. Shooters are not coins. Their fatigue, confidence and the defence’s response all vary from shot to shot, so the chance of success is not identical every time.
- Effects are small if they exist. Even researchers who find evidence of streakiness generally describe modest effects, much smaller than what fans and players perceive.
A fair summary is that the strong version of the belief, that streaks are highly predictive, is not supported, while the question of whether a small real effect exists is still discussed. If you read a confident claim in either direction, treat it with caution.
Why perceived streaks mislead
Whatever the true size of the effect, there are solid reasons why our sense of streaks is unreliable.
Random sequences contain streaks. A player who makes half of his shots, with every shot independent, will still produce runs of four or five made shots at times. Streaks and clusters are a normal feature of random sequences, so seeing one is not evidence of anything special. The chance of four made shots in a row for a 50 per cent shooter is 1/2 × 1/2 × 1/2 × 1/2 = 1/16, which is not rare when many shots are taken.
Small samples. A handful of shots is far too few to separate a hot player from an ordinary one. The problem of small sample sizes means a short run of results can look dramatic and still be consistent with plain luck.
Memory is selective. We remember the sizzling stretches and forget the many ordinary ones, and we notice runs of success more than other patterns.
Regression to the mean. A player who has just played unusually well will usually play closer to their normal level next. That is not a “cold” streak following a hot one; it is what happens when an exceptional result is partly luck.
Hot hand versus gambler’s fallacy
These are often described as mirror images. The gambler’s fallacy expects a streak to end (“tails is due”), while the hot hand expects a streak to continue. In a game of chance with independent outcomes, such as a fair coin, a die or a roulette wheel, both are mistaken, since the next outcome does not depend on the last. In a skilled activity the situation differs, because a player’s ability and form can genuinely vary, so the past is at least potentially informative.
That difference is why the sports question remains open while the casino one does not. For a spin of a wheel, a run of reds says nothing about the next spin. For a shooter or a gamer, a run of successes may reflect real changes in form, but it is far weaker evidence than it feels like, and often mostly noise.
Video games and esports
The same reasoning applies in competitive video games. A player who wins several rounds may be in good form, may have found a strong approach, or may simply have had favourable matchups. Telling these apart takes many games, not a handful. Where a game uses hidden randomness, such as critical hits, a run of crits is simply a run, and does not mean the next attack is more likely to crit.
A careful way to think about streaks
- Ask how many attempts the streak is based on. Very small numbers prove little.
- Ask whether the outcomes are independent (dice, coins) or skill-dependent (shooting, aiming). The reasoning differs.
- Compare the streak to what chance alone would produce, using the basic multiplication of probabilities.
- Be wary of anyone who sells a streak as a prediction, especially when money is involved.
Betting on sports or any other contest carries financial risk, and outcomes remain uncertain whatever the form guide says. If gambling is causing harm, please contact a local support service.