A Data-Driven Analysis of Performance Metrics and Match Outcomes in High School Boys' Volleyball

This study examined the relationship between volleyball performance metrics and match outcomes in high school boys' volleyball. A dataset of 71 matches collected across two competitive seasons was analyzed using descriptive statistics, Pearson correlation analysis, and logistic regression modeling. Correlation analysis identified Side-Out Percentage, Hitting Percentage, and Serve W/L Percentage as the variables most strongly associated with match success. Two logistic regression models were developed to examine the relationship between these performance metrics and match outcomes. A reduced three-variable model consisting of Side-Out Percentage, Hitting Percentage, and Serve W/L Percentage achieved an average five-fold cross-validation accuracy of 88.8%, outperforming an expanded eight-variable model, which achieved an accuracy of 87.3%. Side-Out Percentage demonstrated the strongest association with match outcomes in both models, followed by Serve W/L Percentage and Hitting Percentage. Additional evaluation of the reduced model produced a classification accuracy of 88.9% and an ROC-AUC of 0.977, indicating a strong association between these metrics and match outcomes within this dataset. These findings suggest that a small number of key performance metrics are closely associated with match success and provide practical insight for data-driven volleyball performance evaluation, although validation across additional teams and seasons is needed.