Predicting Professional Darts Match Outcomes Using Player Performance Statistics and Machine Learning

Darts has grown in popularity in recent years, accompanied by increasingly detailed statistical indicators describing player performance. While early analysis was based on simple metrics such as average scores or checkout efficiency, modern tournaments record a wide range of features, including scoring patterns, starting-position averages, and segment-specific accuracy. The aim of this paper is to predict the outcome of darts matches using machine learning methods applied to these diverse statistics. After collecting and analyzing tournament data, several algorithms were applied and compared to identify the most effective model. The best-performing approaches were tested on a dedicated tournament test set. The results highlight both the potential and the limitations of data-driven outcome prediction in professional darts.