Can the accuracy of your running GPS be trusted?
It's happened to everyone: you finish a race, look down at your GPS and find that the total distance is off by several tenths of a mile. Inaccurate course distances can be frustrating on either end. If the course is measured long, then an unrealistically slow time is logged next to your name for all time and for everyone to see. If the course is too short, then you may never live up to your supposed PR and always be holding yourself against an unrealistic standard. So who can be trusted? The GPS on your wrist that makes use of billion-dollar satellites, or the course measurement that makes use of very long tape measures?
How a GPS works
Global positioning devices (GPS) work by bouncing a signal off of satellites in a process called "triangulation," meaning it takes 3 satellites to figure out exactly where you are. Your GPS sends out a radio signal, which is then picked up by three or more satellites, and based on the amount of time it takes for the signal to bounce back to your device, it calculates the distance from each satellite. Data from two satellites allows the GPS to place you in one dimension (more or less a line); data from three satellites locates you on a plane (two dimensions); and data from four satellites gives a three-dimensional picture of where you are in space.
GPS errors
If a satellite and your device cannot "see" each other clearly (or if you have fewer than three satellites to work with), then your signal will be less accurate. Cloud cover, mountains, and even buildings can interfere with your GPS signal, producing an error of up to one hundred feet. Static "noise" or other electronics (such as radio towers, walkie talkies, etc.) working on a similar frequency can also interfere with your signal, producing an error of up to thirty-two feet. On average, Garmin GPS devices (such as Forerunner) are accurate to within 4.8 meters (about 15 feet), or within 20 to 32 feet (5.8-9.9 meters) 95% of the time. Still, even with the errors canceling each other out some of the time, this can lead to an error of several tenths of a mile depending on the length of the course and atmospheric conditions. According to Brian Cavanagh, USAT level 1 coach and certified USATF course measurer, running on a 400m track, a GPS could log your 10K split between 1:04 and 3:34 faster or slower than you actually ran. "A consumer-quality GPS will make more errors on any course which has turns compared to one which is one straight line from point to point with no turnaround," explains Cavanagh. "In the latter case, your GPS will measure between 6-10 meters longer or shorter than an accurate course."
Comments