Hot Hands, Runs of Form and Perceptions of Randomness

World Cup post 2: It is hard to watch sport without screaming at the commentators: “what are you talking about”. I don’t say this to criticize commentators. Almost anyone forced to speak for 90 minutes straight will say a bunch of things that, to put it charitably, don’t make a lot of sense.

A favourite commentator obsession is discussing the players’ form — that small sequences of events indicate something. A striker who doesn’t score for a few games is seen as suddenly a terrible player. Often it is nothing more than chance. A player who scores on average one goal every three games in their career can easily go long streaks of games without scoring for reasons attributable purely to randomness. Any random sequence is going to have streaks when things go well and when they don’t. This is just plain luck — it doesn’t indicate that the striker has got any better, or worse.

Gilovich, Vallone and Tversky looked at this phenomenon in basketball — the “Hot Hand”. This is the belief that because a player has just scored they are suddenly more likely to score again. The researchers show that scoring is random it is only our brains that like to see patterns in random data. As they say…”the “detection” of  steaks in random sequences is attributed to a general misconception of chance according to which even short random sequences are thought to be highly representative of their generating process” (Gilovich, Vallone and Tversky, 1985, page 295). People want random events to look random and so wrongly think streaks, which happen all the time in random sequences, mean something.

When you hear a commentator talking about the form of a player it is quite likely that what they are saying doesn’t make any sense at all but to be fair they have to say something to fill the air time.

Read: Thomas Gilovich, Robert Vallone and Amos Tversky (1985) The Hot Hand In Basketball: On the Misperception of Random Sequences, Cognitive Psychology, 17, pages 295-314.