Concurrent validity of VmaxPro, Kinovea, and Speedograph for the assessment of peak barbell velocity during the bench press: A comparison of technological approaches and historical evolutions

Abstract

Measurement of barbell velocity is a simple and effective way to control strength training. To assess the concurrent validity of different technological approaches measuring barbell velocity, video-analysis (Kinovea), linear velocity transducer (Speedograph), and an inertial measurement unit (V max Pro) were compared. Sixty-eight female and male sport science students lifted two repetitions in the bench press exercise at self-selected barbell loads. Peak vertical barbell velocity (V max ) was parallel measured during the concentric phase of the lift using the aforementioned devices. Concordance correlation coefficient (CCC), Deming regression (DR) and Bland-Altman analysis (BA) were used to assess relative and absolute concurrent validity of V max measured with Kinovea, Speedograph, and V max Pro. Results confirmed high concurrent validity of Speedograph and V max Pro (CCC = 0.99, standard deviation of differences [SDD] = 0.04 m∙s -1 ) without detecting proportional or constant bias. In contrast, V max measured with Kinovea showed poor concurrent validity to Speedograph (CCC = 0.83) and V max Pro (CCC = 0.81) with significant proportional and constant bias. Regression based re-calibration of V max from Kinovea resulted in an SDD = 0.09 m∙s -1 compared to Speedograph and an SDD = 0.08 m∙s -1 compared to V max Pro. Among the three tested devices, V max assessed using Kinovea showed poor concurrent validity. Furthermore, as Kinovea showed proportional bias compared to Speedograph and V max Pro, application-specific re-calibration of Kinovea should be applied when barbell velocity data is compared to Speedograph and V max Pro.