Reliability and validity of the “XingXun” system for measuring punch acceleration and velocity in elite boxers

Introduction: Punching velocity-related indicators are crucial in boxing, and accurately quantifying these metrics is significant for evaluating athletes' performance. This study aims to assess the reliability and validity of a commercial inertial sensor-based boxing training monitoring system ("XingXun") for measuring punch acceleration and velocity. Methods: Nine male boxers from the Shanghai University of Sport (age: 24.8 ± 3.1 years) with national-level competitive experience participated in the study. Participants wore "XingXun" sensors on both hands and performed maximum-effort tests involving six single punches (jabs, hooks, and uppercuts) and a 3-minute combination punch test. Concurrent validity was established by comparing the "XingXun" measurements against a 3D motion capture system (Qualisys) as the gold standard. Reliability was assessed using the Intraclass Correlation Coefficient (ICC) and Coefficient of Variation (CV). Validity was determined through Pearson correlation analysis, paired t-tests, Mean Deviation (MD), and Mean Absolute Error (MAE). Bland-Altman plots were used to visualize the agreement between the two systems. Results: The "XingXun" system demonstrated acceptable to excellent reliability for all punch types, with ICC values ranging from 0.883 to 0.950 and CV from 2.27% to 7.67% for acceleration and velocity. Validity analysis showed a strong correlation between the "XingXun" and Qualisys systems for both punch acceleration (r = 0.836-0.911) and velocity (r = 0.785-0.854). The measurement errors were small and not statistically significant (P > 0.05 or ES < 0.2), and Bland-Altman plots confirmed a high level of agreement between the two devices. Conclusion: The "XingXun" boxing training monitoring system is a reliable and valid tool for quantifying punch acceleration and velocity, offering a practical alternative to laboratory-based 3D motion capture for athlete performance monitoring.