Gurashish Bagga
Title: Offensive and defensive penalties on score differentials and drive outcomes in the NFL
Date: Monday, December 18th, 2023
Time: 10:30am
Location: Zoom
Supervised by: Dr. X. Joan Hu
Abstract: This project studies the impact of offensive and defensive penalties on score differentials and drive outcomes in the NFL while taking into account the other variables that might impact the game. Using linear regression, we first isolated the effects of penalties on score differentials. Following this, a mixed-effect model with random effects for teams and seasons further refined our understanding of these dynamics. The project then explored drive outcomes through logistic regression, examining the different impacts of penalties and predicting drive success. Subsequently, a Random Forest model was employed for the same purpose. Comparing the predictive capabilities of logistic regression and Random Forest models, this study aims to identify the most effective method for predicting drive outcomes. This comprehensive approach not only elucidates the impact of penalties in NFL games for the fans but also offers insights into predicting drive outcomes.