Validating two systems for estimating force and power

This study examined the validity of 2 kinematic systems for estimating force and power during squat jumps. 12 weight-trained males each performed single repetition squat jumps with a 20-kg, 40-kg, 60-kg and 80-kg load on a Kistler portable force plate. A commercial linear position transducer (Gymaware [GYM]) and accelerometer (Myotest® [MYO]) were attached to the bar to assess concentric peak force (PF) and peak power (PP). Across all loads tested, the GYM and MYO estimates of PF and PP were moderately to strongly correlated ( P≤0.05-0.001) with the force plate measurements ( R=0.59-0.87 and R=0.66-0.97), respectively. The mean PF and PP values were not significantly different between the 2 kinematic systems and the force plate, but the estimates did produce some systematic bias and relatively large random errors, especially with the 20-kg load (PF bias >170 N, PF error >335 N, PP bias >400 W, PP error >878 W). Some proportional bias was also identified. In summary, the estimation of PF and PP by a linear position transducer and accelerometer showed moderate to strong relative validity and equivalent absolute validity, but these estimates are limited by the presence of bias and large random errors.