This is the public repository for the paper by Williams, Schliep, Fosdick, and Elmore (2023+) on shooting in the NBA.
Team and player evaluation in professional sport is extremely important given the financial implications of success/failure. It is especially critical to identify and retain elite shooters in the National Basketball Association (NBA), the premier basketball league worldwide. We propose a Bayesian hierarchical modeling framework to cluster teams and players to investigate shooting ability in the NBA. The model can be used to evaluate teams and players on their shot taking tendencies, as well as their ability to make shots. Using data collected from the 2008-09 through the 2020-21 seasons, we identify a clear evolution in shot selection over time, including the widely-recognized trend towards a greater percentage of three point shot attempts. In addition, we highlight the utility of our proposed methodology by introducing an "expected points"" metric to compare teams across the league as well as across seasons. Finally, we propose a novel player-evaluation metric, "expected points above average" (EPAA), as a means for comparing individual players with respect to their shooting abilities. We compute EPAA for the top 100 shot takers over the last decade and find Stephen Curry and Russell Westbrook as having the greatest (2018-19 season) and lowest EPAA (2020 - 21 season), respectively.
The EPAA Shiny application supports additional analysis.