The Hot Hand Rises Again
May marks the 100th anniversary of Babe Ruth’s first home run.
This year, opening day in baseball signals the “closing day” for one of the classic truisms among sports statisticians: the belief that there is no such thing as “hot hitters” or “batter’s slump” in baseball, or the “hot hand” in basketball.
This belief is based on famous research by Gilovich and Tversky in 1985 — “The Hot Hand in Basketball: On the Misperception of Random Sequences.” They studied successions of foul shots in basketball and concluded that players’ chances of making a foul shot were unaffected by whether they had made or missed recent foul shots.
Wikipedia still has a whole article devoted to the “hot hand fallacy” based on this research.
More recent work, though, by Green and Zweibel asserts that, in baseball, a batter’s prior 25 at-bats are a significant predictor of performance on the next at-bat. For a hot hitter, his chances of reaching base rise 25%-30%. Similar effects were observed for home runs, and for pitching performance. And research by Bocskocsky, Ezekowitz and Stein demonstrated a similar, though smaller, effect in basketball.
The original Gilovich and Tversky conclusion may not die so quickly, as statisticians continue to test the new conventional wisdom. One point to note is that they were testing only foul shots — a relatively small component of the flow of action in basketball game, and so was not the most powerful test of the hot hand hypothesis.
And it is worth bearing in mind that just the perception that a hitter in baseball is on a roll (regardless of whether this is accurate) may, in fact, affect his chances of success. For example, pitchers will “pitch around” a hot hitter (i.e. throw pitches that are not strikes), increasing his chances of reaching base by a walk.