Validity and reliability of GPS for measuring distance travelled in field-based team sports

The aim of the present study was to examine the effects of movement intensity and path linearity on global positioning system (GPS) distance validity and reliability. One participant wore eight 1-Hz GPS receivers while walking, jogging, running, and sprinting over linear and non-linear 200-m courses. Five trials were performed at each intensity of movement on each 200-m course. One receiver was excluded from analysis due to errors during data collection. The results from seven GPS receivers showed the mean (±s) and percent bias of the GPS distance values on the 200-m linear course were 205.8 ± 2.4 m (2.8%), 201.8 ± 2.8 m (0.8%), 203.1 ± 2.2 m (1.5%), and 205.2 ± 4 m (2.5%) for the walk, jog, run, and sprint trial respectively. Walk and sprint distances were significantly different from jogging and running distances (P < 0.05). The GPS distance values on the 200-m non-linear course were 198.9 ± 3.5 m (-0.5%), 188.3 ± 2 m (-5.8%), 184.6 ± 2.9 m (-7.7%), and 180.4 ± 5.7 m (-9.8%) for the walk, jog, run, and sprint trial respectively; these were significantly lower than those for the corresponding values on the linear course (P < 0.05). Differences between all non-linear movement intensities were significant (P < 0.05). The overall coefficient of variation within and between receivers was 2.6% and 2.8% respectively. Path linearity and movement intensity appear to affect GPS distance accuracy via inherent positioning errors, update rate, and conditions of use; reliability decreases with movement intensity.