Document Type
Article
Date
8-26-2025
Keywords
Pay, Clustering, NBA, Performance, Linear programming, Simulation
Language
English
Disciplines
Applied Statistics | Data Science | Econometrics | Finance | Labor Economics | Sports Management | Sports Studies | Statistical Models
Description/Abstract
The purpose of this study was to identify player types that exist in the modern National Basketball Association (NBA), test whether player types are paid differently controlling for performance and other factors and construct successful rosters with cheaper payrolls.
We collected performance statistics and salary data for players and teams across five seasons (2018-19 to 2022-23). Cluster analysis is leveraged to group together player-seasons to identify the player types that exist in the NBA. Linear regression models are run to test for differences in pay by cluster membership while controlling for performance, age, and contractual details. Linear programming simulation models construct optimized rosters under payroll and roster construction constraints.
We find nine distinct player types in the NBA. Using advanced metrics, we find that elite playmakers earn significantly greater salaries compared to other player types while floor spacers, traditional bigs, defensive specialists, and three-point specialists earn significantly less compared to the reference category (perimeter operators). Pay differences may be as large as 72% greater for elite playmakers compared to the lowest paid player type. Optimized rosters show traditional bigs and three-point specialists should appear on rosters more often than they currently do.
This paper extends the literature on NBA pay determinants, player types, and roster construction. It provides evidence of inefficiencies in the NBA labor market that can be exploited by NBA teams to reduce payroll expenditures and potentially increase team success.
Recommended Citation
Nick R. Riccardi, Rodney Paul; NBA player types and salaries: assessing the disparities in pay. Managerial Finance 2025; https://doi.org/10.1108/MF-04-2025-0251
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
Included in
Applied Statistics Commons, Data Science Commons, Econometrics Commons, Finance Commons, Labor Economics Commons, Sports Management Commons, Sports Studies Commons, Statistical Models Commons
Additional Information
Accepted: 13-Jul-2025