Comparative Effects Of E-Content, Game-Specific Drills, And Combined Training On Offensive Skills Development In Kabaddi Players Using Machine Learning Models

Abstract

This study investigates the effectiveness of e-content-based training, game-specific drills, and their combination in enhancing offensive skills among collegiate Kabaddi players through the integration of machine learning models. A pre-test and post-test randomized group design was adopted, involving four groups: RTWECG (Regular Training with E-Content Group) representing APSA College Kabaddi players; RTWGSSG (Regular Training with Game-Specific Skills Group) representing Alagappa University College of Physical Education players; COMBTG (Combined Training Group) representing Rajarajan College of Engineering players; and CG (Control Group) representing Alagappa Arts College players. A total of 80 male players from four different colleges in Sivagangai District, Tamil Nadu were randomly assigned into these four groups. Each group underwent a 12-week intervention targeting seven key offensive skills: toe touch, hand touch, scorpion kick, side kick, back kick, dubki, and lion jump. Pre- and post-test data were analyzed using statistical methods (t-tests, ANCOVA) and predictive machine learning models—Support Vector Machine (SVM) and Random Forest (RF). The results revealed significant improvements in all experimental groups, with the COMBTG showing the highest performance gains. Machine learning models effectively classified skill improvement levels and validated the superiority of the combined training approach. These findings highlight the potential of integrating traditional sports training with digital content and data-driven analytics to improve offensive skills in Kabaddi.