MODELING WINNING TEAMS FOR 2019 NCAA DIVISION I FBS FOOTBALL SEASON
- ECSU Author/Contributor (non-ECSU co-authors, if there are any, appear on document)
- De'quante Mckoy, student (Creator)
- Julian A. D. Allagan , Associate Professor (Contributor)
- Institution
- Elizabeth City State University (ECSU )
- Web Site: https://www.ecsu.edu/academics/library/index.html
Abstract: Given the 2019 NCAA football data, this thesis explores the effect of several variablessuch as opponent 3rd down Conversion, Turnovers, and Point per Play Margin, etc., on the Win percent or Win margin percent for each team. We run both logistic and linear regression models and found Point Per Play Margin to be consistentlystatistically significant at 5% risk level. With a linear model on a continuous Win margin, Point Per Play Margin and Opponent 3rd Down Conversion were statistically significant at explaining 90% of the variations in the response. However, with alogistic regression model on a binary win margin, Point Per Play Margin was the only statistically significant variable at classifying the response Win margin ( above 50% or not) with an accuracy rate of 86%.
MODELING WINNING TEAMS FOR 2019 NCAA DIVISION I FBS FOOTBALL SEASON
PDF (Portable Document Format)
904 KB
Created on 1/26/2023
Views: 425
Additional Information
- Publication
- Dissertation
- Language: English
- Date: 2020
- Keywords
- NCAA football data, logistic regression model, win margin