A Descriptive Analysis of Football Matches using Logistic Regression

Sammanfattning: The aim of this study was to explore how match-related statistics contribute to winning association football matches. This is relevant for stakeholders in the football industry to facilitate the understanding of what factors contribute to winning matches and can thus be of use when formulating match tactics. A model was constructed through the use of binary logistic regression, where winning/not winning was used as the response variable and standardized match-related statistics were used as predictor variables. Using the acquired coefficients, it was concluded that, among other variables, the home advantage and the ability of a team to finish on target has a strong correlation with winninggames. Further, the study explores the impact of a team’s ability to win football games on the financial landscape of the modern football world. The results show that some of the examined statistics are well correlated to winning a match, but that the tactical useability of these insights is low.