Division of Credit Modeling for Team Sports with an Emphasis on NCAA Volleyball

Zachary J Hass, Purdue University

Abstract

Assessing player contribution in team sports has direct application to setting lineups and constructing team rosters. It also plays a big role in fan engagement, providing media content for talk show debates and opinion articles. Traditionally collected player contribution metrics have focused on a single aspect of game play and, as a result, haven’t captured the full contribution of a player. More recent metrics have been developed to more fully capture total contribution, placing players on a common basis of comparison. This dissertation proposes a general framework known as Division of Credit modeling for team sports. The purpose of which is to develop data driven approaches to value player contribution based on the apportioning of value from play outcomes. Many of its subcomponents can be found in the sports literature, but are presented here as a cohesive framework and applied directly to National Collegiate Athletic Association Women’s Volleyball. Volleyball is a generally underexplored sport in the literature for player evaluation, but has the necessary elements for a Division of Credit metric to make it a prime example. Models are presented to value contribution based on player presence, similar to the adjusted plus/minus, and to value contribution based on player action grades, a more thorough approach. The work concludes by describing extensions of these models to football.

Degree

Ph.D.

Advisors

Craig, Purdue University.

Subject Area

Statistics

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS