Machine learning, although widely applied across various fields for many years, has only recently gained traction in sports science. It shows great promise in areas such as injury prediction and athlete performance forecasting. This study explores the potential of using machine learning to predict archers' performance based solely on specific physiological parameters. Initially, the total success score of each archer was recorded after three sets of six shots, alongside measurements of their quiet eye parameters. The quiet eye represents the final fixation of the eye on a target, critical for task accuracy. Based on their scores, archers were categorized as low-potential athletes or high-potential athletes using an machine learning clustering method. Subsequently, machine learning classification algorithms were applied to predict the performance class of archers using only quiet eye parameters. The study involved 18 archers (equally split by sex) who had consistently trained in archery for at least two years. The results indicated that the performance classification of archers could be predicted with approximately 90% reliability using only quiet eye data. The findings suggest that machine learning based on quiet eye parameters offers a time-efficient alternative to traditional methods for identifying high-potential archers.
© Copyright 2025 Sports Engineering. The Faculty of Health & Wellbeing, Sheffield Hallam University. Alle Rechte vorbehalten.