Estimating an NBA player's impact on his team's chances of winning

TL;DR

This paper introduces a Bayesian win probability framework to evaluate NBA players' impact on their team's chances of winning, addressing limitations of traditional metrics by considering game context.

Contribution

It proposes a Bayesian linear regression model that accounts for game context and controls for other players, providing more accurate impact assessments.

Findings

Identifies players with high impact not captured by traditional metrics

Reveals highly paid players with low impact

Provides rank-orderings of players within teams and league

Abstract

Traditional NBA player evaluation metrics are based on scoring differential or some pace-adjusted linear combination of box score statistics like points, rebounds, assists, etc. These measures treat performances with the outcome of the game still in question (e.g. tie score with five minutes left) in exactly the same way as they treat performances with the outcome virtually decided (e.g. when one team leads by 30 points with one minute left). Because they ignore the context in which players perform, these measures can result in misleading estimates of how players help their teams win. We instead use a win probability framework for evaluating the impact NBA players have on their teams' chances of winning. We propose a Bayesian linear regression model to estimate an individual player's impact, after controlling for the other players on the court. We introduce several posterior summaries to derive rank-orderings of players within their team and across the league. This allows us to identify highly paid players with low impact relative to their teammates, as well as players whose high impact is not captured by existing metrics.