A predictive analytics model for forecasting outcomes in the National Football League games using decision tree and logistic regression

Abstract

Sports analytics has gained rapid popularity in recent years and will likely continue to evolve. In this study, we construct predictive analytics models to forecast the NFL games outcomes in a season using decision trees and logistics regression. Several variables are used as predictors (independent variables). The binary win–loss outcome measure is used as a target (dependent) variable. Decision tree and binary logistic regression models are constructed to describe the relationships between the predictors and football game outcomes in the NFL.