<p>We present an innovative method for evaluating the performance level of ping pong trainees through the integration of virtual reality (VR) technology and multi-camera data fusion. Traditional ping pong training evaluations often rely on subjective assessments, which can be inaccurate. To address these limitations, we propose a comprehensive evaluation framework that utilizes the time and spatial alignment of VR handles, combined with synchronized data from two cameras capturing the trainee’s movements and the ball’s trajectory within the VR 3D space. This data fusion approach enables a more accurate and objective assessment of trainee performance. Our evaluation metrics are organized into five main categories: basic actions, tactics, physical fitness, scoring, and spin and net clearance trajectory. A key novelty of this study is the expansion of these five categories into 55 detailed evaluation metrics, providing a granular and multifaceted analysis of trainee skills. The methodology involves precise synchronization and alignment of data from the VR handles and cameras, allowing for real-time tracking and analysis of both the trainee’s actions and the ball’s dynamics. The results demonstrate that our system can effectively differentiate between various skill levels, offering detailed insights for personalized training programs. The detailed metrics enable coaches and trainees to identify specific strengths and areas for improvement, such as stroke mechanics, tactical decision-making, and physical conditioning. This framework maintains good consistency with the current popular standard, “China Table Tennis Association’s Level Evaluation Standard and Guidelines for Table Tennis,” while significantly enhancing the objectivity and detail of the evaluation process.</p>