Concurrent validity of novel smartphone-based apps monitoring barbell velocity in powerlifting exercises

The aim of this study was to determine the validity of three smartphone applications measuring barbell movement velocity in resistance training and comparing them to a commercially available linear transducer. Twenty competitive powerlifters (14 male and 6 female) completed a progressive loading protocol in the squat, bench press and deadlift (sumo or conventional) until reaching 90% of the highest load they had achieved in a recent competition. Mean velocity was concurrently recorded with three smartphone applications: Qwik VBT (QW), Metric VBT (MT), MyLift (ML), and one linear transducer: RepOne (RO). 3D motion capturing (Vicon) was used to calculate specific gold standard trajectory references for the different systems. A total of 589 repetitions were recorded with a mean velocity of (mean +/- standard deviation [min-max]) 0.44 +/- 0.17 [0.11-1.04] m<middle dot>s-1, of which MT and ML failed to identify 52 and 175 repetitions, respectively. When compared to Vicon, RO and QW consistently delivered valid measurements (standardized mean bias [SMB] = 0 to 0.21, root mean squared error [RMSE] = 0.01 to 0.04m<middle dot>s-1). MT and ML failed to deliver a level of validity comparable to RO (SMB = -0.28 to 0.14, RMSE = 0.04-0.14m<middle dot>s-1), except for MT in the bench press (SMB = 0.07, RMSE = 0.04m<middle dot>s-1). In conclusion, smartphone applications can be as valid as a linear transducer when assessing mean concentric barbell velocity. Out of the smartphone applications included in this investigation, QW delivered the best results.