An Analytical Framework for Modelling, Analysis and Prediction of Athletes’ Game Performance

Abstract

In this chapter, we present a Machine LearningMachine Learning (ML)(ML) and Computer VisionComputer Vision (CV) (CV)-based framework to build robust models that can analyse complex and high-dimensional in-game and non-game sports data to predict an athlete’s game performanceGame performance for designing strategies that maximize the team’s game performance. This framework (i) integrates data from various sources including game statistics daily training metrics, and cognitive assessments (ii) performs factor analysisFactor analysis on the integrated data to extract a small number of latent features for building models for further analysis and prediction of athletic game performanceAthletic game performance and readiness, and (iii) leverages on these extracted latent factors to build robust DT based non-linear models to provide predictions on an athlete’s game performanceGame performance and readiness with statistical guarantees. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.