What’s the Deal with GPS?
Sorry for the delay on this post, I’ve been sidelined with a stomach bug of sorts, so lately I’ve been curled up in a ball trying not to poop myself…..regardless, on with the show.
When I ran the Ottawa Marathon in the spring of 2010, the friend I went with was using a GPS watch for the first time….well not exactly the first time, it was his wife’s watch and while she was sidelined with plantar fasciitis he snuck out the door with it and ran(yes, pun intended). He was somewhat familiar with it, but definitely not a seasoned user. Needless to say, after he missed a PR by 3-4 seconds, he reviewed the GPS info and was pissed to see that he had run 42.5ish KM. In reality, he ran a fantastic race on a hard course, and once we had a chance to thrash out the who/what/where/whys of using GPS data he accepted his result and moved on. Deep down though, I think he’s still got a bad taste in his mouth about the whole thing.
After my most recent race there was a lot of discussion regarding distance measurements, and the pros and cons of using of GPS measurement/devices. There are vehement parties on both sides, insisting the other parties are wrong. The Pro-GPSers are fad-following technocrats who are stealing the soul of true running, whereas the Anti-GPSers are troglodytic luddites who opposed anything more complicated than a hand wound watch.
So who’s right? Neither…both…errrr…it’s complicated….
What people seem to forget about using GPS technology is that it is simply a tool to advise, or provide reference material to the user. It’s not a decision making tool, and they will never replace the human brain in deciding when to push and when to hang back etc. So, to blame your Garmin for wrecking your race is kind of like blaming your hammer for screwing up your renovation project. What it comes down to is expectations…
Considerations about GPS
GPS running watches generally take a point at 1meter intervals, and your speed/pace between these points is interpolated based on how long it took you to get to the next point. If you have ever had the chance to observe a GPS watch during a run, you will notice the pace calculation varying wildly with very little change in your perceived speed or effort. I often use the stock market analogy when people mention this; if you watch stock values constantly throughout the day (or even daily) you miss the overall trends of the data and get lost in the minute ups and downs of daily trading. If a user constantly obsesses over pace values, they tend to miss the overall trend of how fast they are going, whereas if they check it periodically and average the results they will have a pretty good idea of what is going on.
Another item to consider is the disruption of signal (loss or multipath) to the unit. The watch needs a clear path to the sky to obtain best results, but in reality there are many things that can cause things to go off the rails. Tall buildings or tree cover/branches will disrupt the signal by blocking it outright, or bouncing the signal causing it to travel further than it should thus throwing off the accuracy. This will force the GPS software to interpolate and do it’s best to calculate metrics….albeit with skewed data. In my last race, there is a portion where you pass under a highway overpass, and looking at the map of the running path I took after the race revealed a straight line from the last point the watch took before the bridge to the first point it took when I emerged from cover. Did it make a huge difference? Not likely, but certainly a few meters.
Speaking of accuracy, one thing that is overlooked when people throw that mini-laptop on their wrist and head out for a run is: no matter what kind/amount of cover you are dealing with, consumer grade GPS units (handheld, automotive, or watch format) are not nearly as accurate as we’d like them to be. My Garmin 305 has served me well, but I have idea of the point measurement accuracy and taken that into account during my training:
http://static.garmincdn.com/pumac/984_OwnersManual.pdf (pg. 64)
GPS Position Accuracy: less than 10 meters 50% typical
GPS Velocity Accuracy: less than 0.05 m/s
Points taken by my Garmin 305 are accurate to less than 10meters 50% of the time (again dependant on cover etc.). The software does a wonderful job of smoothing out the occasional oddball point, but it hardly makes my Garmin “accurate”. Is this a BAD thing? Not really. It’s a nature of the technology and things are only getting better, however without a huge antenna and/or a total-station you simply aren’t going to get cm accuracy out of a consumer grade product. Talk to those military dudes, I’m sure they have some GREAT stuff. Additionally, the number of satellites your device can connect to will directly influence the precision at which it can measure. The more satellites the better. By having multiple connections extremity points can be removed giving you something closer to reality.
Lastly, race routes are measured point to point, corner to corner (tangentially), which is NOT the way runners race. Race officials use straight lines, and often the inside point of corners to set their routes and route distances Runners meander within the course boundaries for a variety of reasons: having to contend with faster/slower people (passing/avoiding), moving over to one side or another for water stations/rest stops, adjusting which side of the road they are on to mix up the surface camber (angle) they are running on, object avoidance (fallen runners/debris) etc. Really, everyone’s race distance will vary depending on how closely they followed the point to point measuring of the course setter. I generally try and maintain my route as close to these lines as possible, but even my Garmin read 42.35ish KM after my last race. 150m+/- extra, but over 42.2KM it doesn’t look all that bad, that’s an average of 2.6m-ish for each KM (less than 10ft for you Americans out there). When you consider the road right-of-way (shoulder to shoulder distance) is 10m (~30ft) for the rural roads at the start, and 15m (~45ft) for the Red Hill Parkway you can see it’d be quite feasible to have twice that amount (300m, 1000ft) of meandering over the course of a race….not ideal mind you, but realistic, especially if you are having GI issues.
Overall, I hope that folks will remember that GPS isn’t perfect, and you still need to make decisions for yourself, however if you can use the metrics that a GPS watch provides to your benefit then kudos. Personally I’m a big fan, the instant feedback helps me maintain my focus. At the very least, it’ll open your eyes to how elevation or other factors affect your speed, not to mention showing you your improvement over time, which is the ultimate goal in my opinion.