In recent years, football has undergone substantial changes, requiring players to adapt both mentally and physically. The outcomes of matches depend heavily on the dynamic interplay among teammates and opponents, with players constantly adjusting their positions in response to evolving game situations. Simultaneously, the monitoring of athlete loads has gained considerable attention, driven by technological advancements. It is now standard practice for football clubs worldwide to employ cutting-edge global positioning systems to monitor and assess player performance during matches. This study aims to develop a machine-learning model for clustering football performance based on match load zones derived from these tracking systems. The research focuses on an elite team from the Malaysia Super League (MSL) during the 2022 season. Using Louvain clustering, the study identified three performance levels: low (LP), moderate (MP), and high (HP). An Artificial Neural Network (ANN) was then trained to classify these clusters, achieving an accuracy rate of 86.4%. These findings are of significant value to coaches and sports managers, offering a method to evaluate players based on match demands and load zones.