Is it a fallacy to believe in the hot hand in the NBA three-point contest?

Abstract

The NBA Three-Point Contest has been considered an ideal setting to study the hot hand, as it showcases the elite professional shooters that hot hand beliefs are typically directed towards, but in an environment that eliminates many of the confounds present in game action. We collect 29 years of NBA Three-Point Contest television broadcast data , apply a statistical approach that improves on those of previous studies, and find considerable evidence of hot hand shooting in and across individuals. Our results support fans' and experts' widely held belief in the hot hand among NBA shooters.

How rational is gambling?

“…Since Elo ratings have very good forecasting properties and are partially composed of recent results, it can be stated that there is a connection between future success and past performance. This is in line with the conclusions of Miller and Sanjurjo (2021, 2019, 2018). Research employing ELO ratings thus also suggests that hot hands really exist and believing in them is no mental fallacy.…”

The hot hand in the wild

“…The permutation test described Miller and Sanjurjo [6] is appropriate whenever examining a series of trials that are identically distributed (e.g., free throws or three point contest shots). Furthermore, in the studies by Miller and Sanjurjo on controlled shooting (from threepoint contests [20] and a shooting field experiment [21]) they also consider sequences of consecutive makes that might span different competition rounds, and, perform permutations across the different sessions/rounds for a player. This permutation eliminates any possible bias introduced by systematic variation between rounds, irrelevant to the hot hand, while it also eliminates any (partial) hot hand effect that is activated across sessions.…”

The Hot Hand in the NBA 3-Point Contest: The Importance of Location, Location, Location

“…They find consistent evidence for a hot hand in the NBA 3-Point Contest across their different measures of the hot hand. Our paper builds on the data collected by Miller & Sanjurjo (2021), and our methods are similar to some of their methods, so we summarize their results in more depth here and then quickly summarize the differences in our approaches. In the "Reconciliation with Miller and Sanjurjo (2021)" section, we describe a replication of main results from Miller & Sanjurjo (2021) and detail how our results differ.…”

“…We supplement the data collected by Miller & Sanjurjo (2021) with two additional data sources. First, we merge data on player characteristics collected from http:// www.basketball-reference.com using the bbr package in R. 5 We collect the season 3-point percentage as well as the total number of 3-point attempts and the player's age.…”

“…The combination of repeated shots within the same location and shots spanning locations in a relatively controlled setting allows a close examination of whether a hot hand exists and, if so, whether it is affected by a player moving shot locations. Miller & Sanjurjo (2021) collect data spanning nearly all of the contests from the first NBA 3-Point Contest in 1986 to the 2020 contest and find that a hot hand exists in the NBA 3-Point Contest. Furthermore, when examining shots between and across locations, they find that the hot hand effect tends to be the largest for the first few shots from each location, when players have just moved shot locations.…”