Is it a Fallacy to Believe in the Hot Hand in the NBA Three-Point Contest?

The hot hand fallacy refers to a belief in the atypical clustering of successes in sequential outcomes when there is none. It has long been considered a massive and widespread cognitive illusion with important implications in economics and finance. The strongest evidence in support of the fallacy remains that from the canonical domain of basketball, where the widespread belief in the existence of hot hand shooting, among expert players and coaches, has been found to have no evidential basis (Gilovich, Vallone, and Tversky 1985). A prominent exhibit of the fallacy is Koehler and Conley (2003)'s study of the NBA Three-Point Contest (1994-1997), a setting which is viewed as ideal for a test of the hot hand (Thaler and Sunstein 2008). In this setting, despite the well-known beliefs of players, coaches, and fans alike, Koehler and Conley find no evidence of hot hand shooting. In the present study, we collect 29 years of shooting data from television broadcasts of the NBA Three-Point Contest (1986-2015), and apply a statistical approach developed in Miller and Sanjurjo (2014), which is more powered, contains an improved set of statistical measures, and corrects for a substantial downward bias in previous estimates of the hot hand effect. In contrast with previous studies, but consistent with Miller and Sanjurjo (2014)'s recent finding of substantial hot hand shooting in all previous controlled shooting studies (including that from the original study of Gilovich, Vallone, and Tversky), we find substantial evidence of hot hand shooting in the NBA Three-Point Contest. This leaves little doubt that the hot hand not only exists, but actually occurs regularly. Thus, belief in the hot hand, in principle, is not a fallacy.