USING DATA ANALYTICS TO PREDICT NBA PLAYER SALARIES

Abstract

This project uses advanced data analytics to develop a model to predict NBA player salaries using a combination of in season statistics and off the court characteristics of each player. The model aims to help aid general managers by providing them with an unbiased, data driven approach to evaluating NBA player value, to be used in important decisions like trades, free agent negotiations, and contract extensions. It helps by flagging players as "Underpaid" or "Overpaid" based on their performance and experience, helping general managers avoid players with hefty contracts that don't contribute to winning. Outside of the NBA front office, this model has lots of broad applications including helping player agents in contract negotiation and providing insight for sports betting companies and bettors to identify differences in mispriced player lines. Additionally, the model can be leveraged by sports companies to provide insight to audiences as to why a certain NBA player was able to sign a lucrative contract, and if it was an overpay or underpay by the team. Ultimately, this model offers a comprehensive framework for understanding the intersection of player performance, salary, and team economics, making it a valuable resource for general managers in and around the NBA.