Introduction: Methods currently available to analyse speed variability in swimming rely on a few strokes sampled from the middle of each lap or on an average score of the laps (1). These methods only provide a performance snapshot, and offer limited information as to the evolution of instantaneous speed or other technical information during a swim. Here, we used an accelerometer to calculate swimmer`s speed, and validated this speed in comparison with 5-meters splits during an entire 100-m freestyle swimming performance. Methods: The acceleration signal was recorded at 100Hz (Nanotrack, Catapult) from an elite sprint swimmer during an all-out, 100-m swim. The signal was band-pass filtered to remove signal noise and gravity. Acceleration signal was then integrated twice to obtain time series positions, and positions were down-sampled to obtain 5-meters split. To validate this method, time splits were obtained every 5 meters from an overhang and two underwater cameras (60Hz). For these two methods, speed was obtained by a derivative of the time splits, and the error was quantified by calculation of the root mean square error (RMSe), coefficient of determination (r2), maximum error and ratio (maximum speed of the accelerometer divided by maximum speed derived from time splits). Results: The error between the two methods during the entire 100-m swim was as follows: RMSe was 0.1 m/s, r2 was 0.99, maximum error was 0.3 m/s and ratio 0.88. Mean speed for the entire race with the accelerometer (1.75 m/s) was very close to that estimated from video-based time splits (1,77 m/s). Discussion: The error in instantaneous velocity between the single accelerometer and the gold standard video-based time splits method was minor. The speed integrated from an accelerometer is therefore considered accurate during an entire race including dive, dolphin kicks, swim and turn. While the gold standard, video-based method is time consuming, the accelerometer-derived speed calculations can be made instantly, which greatly enhances coach`s feedback to athletes. Further, the easy-to-use accelerometer is very small and allows the swimmer to swim freely compared with other methods (2)
© Copyright 2014 19th Annual Congress of the European College of Sport Science (ECSS), Amsterdam, 2. - 5. July 2014. Veröffentlicht von VU University Amsterdam. Alle Rechte vorbehalten.