Impact of penalties on score differentials and drive outcomes in the NFL

This project investigates the influence of offensive and defensive penalties on score differentials and drive outcomes in the NFL, incorporating various factors that could affect game dynamics. Employing linear regression, we initially look into the specific impacts of penalties on score differentials. Building upon this analysis, a linear regression model with random intercepts for teams and seasons was employed to further refine our understanding of these interactions. The study then delves into drive outcomes, utilizing logistic regression to examine the distinct effects of penalties and predict drive success. Additionally, a Random Forest Algorithm is implemented for the same purpose, allowing for a comparative assessment of the predictive capabilities of logistic regression and the Random Forest method. Through this comprehensive approach, the project identifies effective methods for predicting drive outcomes, shedding light on the intricate dynamics of penalties in the NFL games and providing insights for both fans and analysts.