GPS PRECISION IN MARATHON RACES

Once technology has entered sports practice, GPS watches have become one of the most common devices, especially in running.

GPS watches provide useful information. Pace, distance and height are usual parameters, but they are not always as precise as they should be. Among the factors affecting GPS accuracy we could find width of streets, buildings altitude, high voltage cables, trees or cloudy skies, that may affect reception from satellites.

A recent article evaluated the precision of different GPS models in some of the most relevant marathons.

 

The study

The data was obtained from data published voluntarily by popular athletes in two editions of the marathons of Berlin, Boston, Chicago, London, New York and Valencia. 73898 runners using 85 different GPS models were identified, and their devices classified in road, trail or cell phone types.

Regarding the measured distance Chicago gave the longest (44.2 and 44.6k), followed by New York (42.9k) and London (42.9 and 43.0k). Berlin, Boston and Valencia gave approximately similar results.

 

The results

  • GPS devices usually introduce an excess in measures. It was found that finishing time affected measurements: longer finishing times were usually related with bigger errors. Thus, slower runners usually get longer distances than faster runners in all races.
  • There were differences regarding the GPS model. In general, the road models were more accurate than the trail models and phone apps.
  • Newer devices are not better, and the brand is not relevant. Mobile phones are the devices with the highest deviation.
  • For road races the best choice (most accurate) was found to be the Garmin FR 235.
  • For trail races all models showed significant errors, with Garmin Fenix 5 showing the best results.

 

Do you use any of these models? Are them accurate?

We would to read your opinions.

 

Bibliography

Precision of Wearable GPS in Marathon Races. J. Lluch, M. Rebollo, Á. Calduch-Losa and R. Mollá. IEEE Consumer Electronics Magazine, vol. 10, no. 1, pp. 32-38, 1 Jan. 2021, doi: 10.1109/MCE.2020.2986820.

Photo by Cocoparisienne (Pixabay)

Leave a Reply

Your email address will not be published.