Generalized Linear Model approach to the Prediction of the outcome of Mixed Martial Arts Fights

Abstract

Mixed martial arts is a complex combat sport that encompasses striking, grappling, and submissions. In a sport where fights can be won by finishing a fight or go to decision there is a multitude of factors that can influence the outcome of a fight. In an effort to determine which factors are statistically significant to a fight a generalized linear model approach was selected. Since mixed martial arts is a sport in which two competitors fight, and one is declared a winner, the result of a fight can be thought of a binary classification problem.