The idea of the ‘hot hand’, where a player who makes several successful shots has a higher chance of making some more, is popular with sports fans and team coaches, but has long been considered a classic example of a cognitive fallacy – an illusion of a ‘streak’ caused by our misinterpretation of naturally varying scoring patterns.
But a new study has hard data to show the hot hand really exists and may turn one of the most widely cited ‘cognitive illusions’ on its head.
A famous 1985 study by psychologist Thomas Gilovich and his colleagues looked at the ‘hot hand’ belief in basketball, finding that there was no evidence of any ‘scoring streak’ in thousands of basketball games beyond what you would expect from natural variation in play.
Think of it like tossing a weighted coin. Although the weighting, equivalent to the players skill, makes landing a ‘head’ more likely overall, every toss of the coin is independent. The last result doesn’t effect the next one.
Despite this, sometimes heads or tails will bunch together and this is what people erroneously interpret as the ‘hot hand’ or being on a roll, at least according to the theory. Due to the basketball research, that seemed to show the same effect, the ‘hot hand fallacy’ was born and the idea of ‘scoring streaks’ thought to be sports myth.
Some have suggested that while the ‘hot hand’ may be an illusion, in practical terms, in might be useful on the field.
Better players are more likely to have a higher overall scoring rate and so are more likely to have what seem like streaks. Passing to that guy works out, because the better players have the ball for longer.
But a new study led by Markus Raab suggests that the hot hand does indeed exist. Each shot is not independent and players that hit the mark may raise their chances of scoring the next time. They seem to draw inspiration from their successes.
Crucially, the researchers chose their sport carefully because one of the difficulties with basketball – from a numbers point of view – is that players on the opposing team react to success.
If someone scores, they may find themselves the subject of more defensive attention on the court, damping down any ‘hot hand’ effect if it did exist.
Because of this, the new study looked at volleyball where the players are separated by a net and play from different sides of the court. Additionally, players rotate position after every rally, meaning its more difficult to ‘clamp down’ on players from the opposing team if they seem to be doing well.
The research first established the belief in the ‘hot hand’ was common in volleyball players, coaches and fans, and then looked to see if scoring patterns support it – to see if scoring a point made a player more likely to score another.
It turns out that over half the players in Germany’s first-division volleyball league show the ‘hot hand’ effect – streaks of inspiration were common and points were not scored in an independent ‘coin toss’ manner.
What’s more, players were sensitive to who was on a roll and used the effect to the team’s advantage – more commonly passing to those on a scoring streak.
So it seems the ‘hot hand’ effect exists. But this opens up another, perhaps more interesting, question.
How does it work? Because if teams can understand the essence of on court inspiration, they’ve got a recipe for success.
Link to blocked study. Clearing a losing strategy.
Link to full text which has mysteriously appeared online.
16 thoughts on “The hot hand smacks back”
Will somebody pirate this paper so that a mere plebian non-academic can read it?
To hell with scientific publishers.
Interestingly, an inverse hot-hand effect (or rather, inverse hot-foot effect) is part of the folk psychology of football (soccer for our American friends).
The idea is that if a player fails to score goals for a while, they can suffer a goal drought, and not score again for a long time. The proposed explanation is that when players fail to score, they lose confidence, and this impairs their performance and they makes less attempts to score goals (in strikers at least).
Volleyball, like every other competitive ball sport, has established tactics for defending against the opposition’s best attacking players. Also, the rotation of players only lasts until the ball is served, when players are free to move around the court. Teams can, and do, rearrange themselves into their preferred strategic positions once the ball is served.
This study seems to be based on some badly flawed premises.
I’ve acquired the paper through the underground academic railroad and made it available here: https://docs.google.com/viewer?a=v&pid=explorer&chrome=true&srcid=0B9hXQWuTea4MMGVlMDk2MzktZjljNy00MmQyLWJmN2EtNTkzOWEwYzU4Zjc5&hl=en_US
Interesting finding. I wonder if Snyder’s work on Hope Theory might be relevant to this. His model of Hope proposes that each inicident of goal attainment increases a person’s agency (sense of self as someone who is able to achieve goals) and pathways thoughts (ability to plan how to achieve future goals), increasing the chances of future goal attainment. Conversely, incidents of goal non-attainment reduces the person’s overall level of hope and makes future goal attainment less likely. Is there a ‘cold hand’ effect for players on a losing streak?
The problem with this study is that it examines what is essentially a solitary activity within a team sport.
I believe the “hot hand” may be attributable to “flow” psychology ala Mihaly Csikszentmihalyi. As ones personal competence in the skills needed to perform any activity increase and they meet increasing success a decrease in conscious awareness (critical self commentary) of ones actions decreases allowing for more spontaneous, and efficient reactions. Especially in the face of targeted defensive obstacles (two defenders at the net, the nasty slider hitting the outside corner) an “automatic” millisecond reaction is what works.
I think you’re spot on. I remember looking at the original study in a psychology class in college and finding it hard to digest. I didn’t like the data they used to make the conclusion. If I remember correctly, the study looked at whether or not making a basket, makes it more likely for a player to make the next basket, but that’s missing the true meaning of the ‘hot hand’. Shooting the ball is partly mechanical, but a big part of it is mental, so being in ‘flow’, i.e. not over-thinking your shot, not having a bad mental attitude, makes a huge difference in how many shots you will make. there’s no make-miss pattern to look for. That being said, I’d imagine it’s pretty hard to come up with a mathematical formula that describes ‘flow’.
There might be some use in considering this phenomenon in relation to cerebellar forward and inverse models. We provided some summaries in these two papers: http://www.ncbi.nlm.nih.gov/pubmed/21630084 and http://www.ncbi.nlm.nih.gov/pubmed/20680539
The hot hand idea seems to me almost certainly true. Imagine, for a moment, the ‘cold hand’. If a player is upset, stressed, distracted, for a period of 20 minutes, we wouldn’t be surprised if he/she performed missed the shot a large number of times (beyond random chance). That, in itself, makes it understandable that during periods devoid of such interferences the player performs above his mean performance level (since the mean performance included performance during the periods of distraction, stress… I think there’s more to it than that, but this may explain a lot. I.e., good performance occurs during periods when there are no factors leading to bad performance.
I wonder if the Gilovich data doesn’t itself support the hot hand hypothesis. Suppose that there is no such thing as a hot hand – each player has an unchanging probability of making a shot as a function of his skill level – but that there is a belief in the hot hand. In that case, the opposition team will take steps to prevent someone who has a hot hand from getting shooting opportunities. So if there is no hot hand, then we ought to expect that players who are thought to have hot hands will *under*perform: do worse than their skill level would suggest, due to the effects of the opposition on their success. If we do not see underperformance, but rather constant performance, that is evidence that there is a hot hand: a player is doing better than expected, given the attention paid to them by the opposition.