Validity of a Wearable Accelerometer Device to Measure Average Acceleration Values During High-Speed Running

<jats:title>Abstract</jats:title> <jats:p>Alexander, JP, Hopkinson, TL, Wundersitz, DWT, Serpell, BG, Mara, JK, and Ball, NB. Validity of a wearable accelerometer device to measure average acceleration values during high-speed running. <jats:italic toggle="yes">J Strength Cond Res</jats:italic> 30(11): 3007–3013, 2016—The aim of this study was to determine the validity of an accelerometer to measure average acceleration values during high-speed running. Thirteen subjects performed three sprint efforts over a 40-m distance (<jats:italic toggle="yes">n</jats:italic> = 39). Acceleration was measured using a 100-Hz triaxial accelerometer integrated within a wearable tracking device (SPI-HPU; GPSports). To provide a concurrent measure of acceleration, timing gates were positioned at 10-m intervals (0–40 m). Accelerometer data collected during 0–10 m and 10–20 m provided a measure of average acceleration values. Accelerometer data was recorded as the raw output and filtered by applying a 3-point moving average and a 10-point moving average. The accelerometer could not measure average acceleration values during high-speed running. The accelerometer significantly overestimated average acceleration values during both 0–10 m and 10–20 m, regardless of the data filtering technique (<jats:italic toggle="yes">p</jats:italic> < 0.001). Body mass significantly affected all accelerometer variables (<jats:italic toggle="yes">p</jats:italic> < 0.10, partial η<jats:sup>2</jats:sup> = 0.091–0.219). Body mass and the absence of a gravity compensation formula affect the accuracy and practicality of accelerometers. Until GPSports-integrated accelerometers incorporate a gravity compensation formula, the usefulness of any accelerometer-derived algorithms is questionable.</jats:p>