What is it about?
We developed a distance based learning approach to analyze the underlying heterogeneity structure of shot selection among professional basketball players in the National Basketball Association (NBA). We use a log-Gaussian Cox process (LGCP) to model the spatial pattern of the shot attempts and compute the players' similarity matrix. Based on similarity matrices of fitted intensity among different players, a mixture of finite mixtures (MFM) model is incorporated for group learning. Our proposed method can simultaneously estimate the number of groups and group configurations. An efficient algorithm is also developed for our proposed model.
Photo by Markus Spiske on Unsplash
Why is it important?
The real data application give us the information about player's shooting habit location. Players can understand their own shooting habits, and they can also strengthen their weaker shooting locations. On the other hand, the professional coach can formulate a defensive strategy to reduce the opponent’s score with these information. Our grouping results will provide a good guidance for team managers trading the players with similar shot pattern.
Read the Original
This page is a summary of: Bayesian group learning for shot selection of professional basketball players, Stat, April 2021, Wiley, DOI: 10.1002/sta4.324.
You can read the full text:
The following have contributed to this page