News

The pitfalls of measuring with GPS watches

Last changed on

"I'm not convinced that the course length is right!" We are confronted with this and similar statements every year in the runner survey or in emails.

Runner 1: "The marathon distance was wrong: My Garmin GPS watch measured the marathon distance as 42.56 km! It should have been 42.2 km. My GPS watch has always given the exact distance for about 30 races so far."

Runner 2: "I'm not convinced that the distance is correct! My Garmin Fenix 3 and friends of mine with Suunto running watches have all measured 21.50 km. I know that the GPS and GLONASS have deviations."

We don't want to give you a recommendation as to which GPS watch measures the most accurately, because we can't do that. However, we can reassure you first and foremost about the distance of the SwissCityMarathon - Lucerne. As our course has been measured and certified several times by an international IAAF measurer with the appropriate measuring technology (so-called Jones counter), we can confirm that the courses comply with the specifications and therefore have the correct distances.

Nevertheless, the recurring feedback regarding the length of the courses has given us pause for thought. We therefore went in search of possible reasons for the reported difference between the official measurement procedure and the local GPS positioning. We found our first clue in an article in Fit for Life magazine, which interviewed international surveyor John Kunkeler: "GPS watches are simply too inaccurate. They interpolate, have dropouts under trees and in urban canyons. We once invited 80 authors of these complaints to an information event, and 25 of them actually turned up. Together we walked the 1000 metre reference distance measured by the civil engineering office. The best came within seven metres. Discussion finished." (Source: https://www.fitforlife.ch/artikel/marathon-streckenvermessung).

According to this practical test, it is already clear that the "best" watch will show a difference of 294 metres (42x7m) over the course of a marathon. With this deviation, we are pretty much exactly in the range of the statement made by our runner 1.