Ranking Colleges by Categories

• Activity Number: 43
• Type: Contributed
• Date/Time: Sunday, July 29, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Sports
• Author(s): Jerome Keating*+ and Robert L Mason and Ryan S Gill
• Companies: The University of Texas at San Antonio and Southwest Research Institute and University of Louisville
• Keywords: Regression ; Logistic Regression ; BCS

Abstract

Football is one of the most popular sports in the United States, particularly at the college level. One result of this interest is the desire to be able to predict the outcome of a game using the available statistics on the two competing teams. These statistics typically include measurements of both the offensive and defensive capability of a team, such as the totals of the passing or rushing yards for and against each of the teams. In this presentation, we construct various types of regression models, using both logistic and least squares regression methods, to construct such models. The methodology can also be modified to rank team performances by category such rushing yards, points allowed, etc. These rankings provide adjusted rankings based on the level of a team's competition or its strength of schedule in a particular category. The data source is the list of the statistics for all of the games and teams competing in 2011 at the NCAA Bowl Championship Series level. The results indicate that good models can be obtained, and provide some interesting outcomes. Outliers are of particular interest.