Big Score For The Hot Hand

In the February 25 Issue:

Big score for the hot hand

Volleyball players’ scoring streaks get statistically fired up

By Bruce Bower

February 11th, 2012; Vol.181 #3 (p. 9)

Text Size

Volleyball players have taken a stand — make that a leap — for the existence of that statistically elusive feat known as the hot hand.

Not only do top volleyball strikers go on scoring runs that can’t be chalked up to chance, but players and coaches notice when a player is on a hot streak and funnel the ball his or her way, say psychologist Markus Raab of German Sport University Cologne and his colleagues, who studied the hot-hand phenomenon by analyzing playoff game data from a German volleyball league.

That strategy usually works, because players who on average score on a high percentage of shots tend to get hot hands. So getting them the ball during a scoring streak boosts a team’s score, the researchers will report in an upcoming Journal of Experimental Psychology: Applied. This tactic backfires if a player with a low scoring average develops a hot hand and draws shots away from better scorers, the scientists hold.

Debate about whether hot hands are real or illusory has raged since a 1985 report that professional basketball players’ shooting and free throw records contain no chance-defying streaks.

“Volleyball is a good test-bed for the hot hand, as the defense cannot directly block ball allocations to a hot player because a net separates opposing teams,” Raab says.

Finding that hot handedness not only exists but affects game decisions fits with evidence that people expect to encounter food and other desired items in patches, says psychologist Andreas Wilke of Clarkson University in Potsdam, N.Y. (SN: 2/12/11, p. 26). “Belief in the hot-hand phenomenon can be seen as part of humanity’s evolutionary heritage of trying to detect streaks in the locations of key resources,” he says.

Raab’s team thoroughly analyzed patterns of successful and failed shots in volleyball, but further work should examine players’ neutral shots kept alive by the defense for evidence of streaks that may alter game strategy, says psychologist Alan Reifman of Texas Tech University in Lubbock.

In the new study, 86 of 94 sport science students, all experienced athletes, believed in a volleyball hot hand — defined as a higher probability of scoring after two or more hits compared with two or more misses.

An analysis of playoff data from the 1999/2000 season for 26 top scorers in Germany’s first-division volleyball league identified 12 players as having had scoring runs that could not be chalked up to chance. Hot-handed players’ shots contained fewer sequences of consecutive scores than expected by chance, the result of a small number of especially long scoring runs.

When shown videos of volleyball matches edited so that one player on each side took all shots, 16 German volleyball coaches accurately monitored each man’s scoring rate as matches progressed. Coaches believed in the hot hand and said that their players created scoring opportunities for hot-handed teammates.

When questioned during similar volleyball videos, 21 volleyball-playing students accurately tracked two players’ scoring averages and said that they would give the ball more often to a player on a scoring streak who fell just short of having a statistical hot hand. Even if one player deemed to be hot had a much lower scoring average than the other, hot-handedness prevailed in deciding who got the ball.

Print
|
Comment
|
|
|
Tweet
|

Found in: Humans and Psychology

Share & Save

Comments 6

Please alert Science News to any inappropriate posts by clicking the REPORT SPAM link within the post. Comments will be reviewed before posting.

recently we found this phenomenon in basketball and baseball.
Since links are not allowed in comments here try to google for "blog gostorm plos one"

Gur Yaari

Jan. 13, 2012 at 1:11pm
Maybe this is already involved in the calculations, but it seems like the real question has to do with the length of fluctuations in skill compared to the length of the typical random "streak." A player with a given skill level will hit accurately a given percentage of time, but the hits and misses will be arranged randomly, so there will be some short and some long streaks, but for a given skill level there will be an average streak length beyond which a streak will be exceptional. At the same time, it is clear that all of us have varying levels of performance -- whether a product of fatigue, attitude, or other factors. If periods of higher skill are shorter than the average random streak, then players would notice random streaks as much as "real" skill-based streaks, and observers would say "hot hands" are illusory. On the other hand, if a player's period of enhanced skill is longer than an average random streak, then noticing better performance will really identify a hot hand. So basically, I think it goes without saying that players have hot streaks, but what this study shows is that the streaks are long enough to be identified beyond the "noise" of random streaks.

Fig and Flan

Jan. 17, 2012 at 10:15am
Belief in a "hot hand" is a positive survival characteristic in the natural world because scarce resources have a tendency to cluster. The same belief is fatal in games of chance. This has something to do with why so many people persist in playing these games despite their mounting losses.

Conrad Seitz

Jan. 17, 2012 at 10:16am
Conrad Seitz makes an excellent point. It seems that our brains try to find patterns even when there are none.

Ralph Dratman

Jan. 23, 2012 at 11:05am
Fig and Flan have mostly expressed what has been bothering me for some time about the "hot hands" issue. It would seem natural for the performance of an athelete to vary with time based on current health, fatigue, confidence and morale level, recent training and practice, the same factors about opposing players and teammates, and many other factors. I do not know how the hypothesis testing was set up, but I wonder whether the hypothesis was tested in the correct direction. The assertion that the variation in performance of an athelete with time about the athlete's average is random should be established as valid before the claim can be made that "hot hands" does not exist. This puts the demands for exceeding "confidence limits" where it belongs. The athlete at "time A" is not the same athlete at "time B" because of the factors noted, so unless it is demonstrated that performance variation with time is very limited, a hypothesis testing that ignores time is not valid.

I have not commented previously because of not being familiar with the details of these analyses, but have done so now because the claimed result that a statisticial claim that "hot hands" does not exist is so extraordinary. To accept this, I would need clarification of exactly what is meant by the term "hot hands" and how the hypothesis was tested.

Tony Cooley

Jan. 23, 2012 at 11:05am
I would believe the perceived effect, as part of Tony Cooley's comment, is tied with confidence and morale level. A good player tends to scout consciously or unconsciously for an advantage. When, that advantage is found and exploited, it is up to the opposition to take notice, develop a counter. The ability of the opponents to successfully adapt (substitute the player that is a weak link or shift tactics) and the morale deficit created by repeated exploitation is possibly more important than the player with the "hot hands". If the opposition’s confidence is eroded quickly enough without a counter, the morale impact feeds into the “hot hands” dominance. (“This guy is just unstoppable”- type of mentality)

Shiayae

Feb. 16, 2012 at 1:04pm

Registered readers are invited to post a comment. To encourage fruitful discussion, please keep your comments relevant, brief and courteous. Offensive, irrelevant, nonsensical and commercial posts will not be published. (All links will be removed from comments.)

You must register with Science News to add a comment. To log-in click here. To register as a new user, follow this link.