Can we use GPS for assessing sprinting performance in rugby sevens? A concurrent validity and between-device reliability study

Abstract

The purpose of this study was to (1) provide data on maximal sprinting speed (MSS) and maximal acceleration (Amax) in elite rugby sevens players measured with GPS devices, (2) test the concurrent validity of the signal derived from a radar device and a commercially available 16 Hz GPS device, and (2) assess the between-device reliability of MSS and Amax of the same GPS. Fifteen elite rugby sevens players (90 ± 12 kg; 181 ± 8 cm; 26 ± 5 y) participated in the maximal sprinting test. A subset of five players participated in the concurrent validity and between-devices reliability study. A concurrent validity protocol compared the GPS units and a radar device (Stalker ATS II). The between-device reliability of the GPS signal during maximal sprint running was also assessed using 6 V2 GPS units (Sensorevery-where, Digital Simulation, Paris, France) attached to a custom-made steel sled and pushed by the five athletes who performed a combined total of 15 linear 40m sprints. CV ranged from 0.5, ±0.1 % for MSS and smoothed MSS to 6.4, ±1.1 % for Amax. TEM was trivial for MSS and smoothed MSS (0.09, ±0.01) and small for Amax and smoothed Amax (0.54, ±0.09 and 0.39, ±0.06 respectively). Mean bias ranged from -1.6, ±1.0 % to -3.0, ±1.1 % for smoothed MSS and MSS respectively. TEE were small (2.0, ±0.55 to 1.6, ±0.4 %, for MSS and smoothed MSS respectively. The main results indicate that the GPS units were highly reliable for assessing MSS and provided acceptable signal to noise ratio for measuring Amax, especially when a smoothing 0.5-s moving average is used. This 16 Hz GPS device provides sport scientists and coaches with an accurate and reliable means to monitor running performance in elite rugby sevens.

How far from the gold standard? Comparing the accuracy of a Local Position Measurement (LPM) system and a 15 Hz GPS to a laser for measuring acceleration and running speed during team sports

“…We found that the GPS systematically underestimated v max ; the MPBs of v max ranged from –3.3 to −1.0%. These results are comparable to those of Lacome et al [ 18 ], who found an overall bias of –3.0%. Buchheit et al [ 32 ] observed a positive mean bias of v max of 0.3% for 5 Hz GPS measurements.…”

“…In contrast, Buchheit et al [ 32 ] found that CV values of v max measured by GPS tend to be higher than those of the LPM system, and are considerably higher than the ones measured in this study. As with a max , inter-unit CV measures of v max for GPS are similar to those found by Lacome et al [ 18 ] (CV of 0.5%), but better than those reported in studies by Coutts and Duffield [ 6 ] (CV 2.3 to 5.8% for different types of GPS devices) and by Johnston et al [ 11 ] (CV of 8.1%; 15Hz GPS).…”

“…Akenhead et al [ 16 ] demonstrated that the validity and reliability of speeds measured with a 10 Hz GPS were inversely correlated to acceleration—i.e., the higher the acceleration, the lower the validity and reliability of speed measurements during the acceleration. Furthermore, Buchheit et al [ 17 ] compared three different GPS models and showed a between-unit variation in peak acceleration of 10%; and the CV for maximal speed and maximal acceleration in Lacome et al [ 18 ] was 0.5% and 6.4% respectively, as measured during a 40 m sprint test. However, to our knowledge, no studies have investigated accelerations during changes of direction.…”

Untitled

“…We found that the GPS systematically underestimated v max ; the MPBs of v max ranged from -3.3 to −1.0%. These results are comparable to those of Lacome et al [18], who found an overall bias of -3.0%. Buchheit et al [32] observed a positive mean bias of v max of 0.3% for 5 Hz GPS measurements.…”

“…Additionally, a vehicle offers the opportunity to wear several devices at the time. However, to date, only three GPS accuracy studies have analyzed acceleration using an object other than a person as a device carrier [16][17][18].…”