The GPS Debate

My Charlotte Road Runners Club news letter arrived in the mail early this week and as usual, almost immediately I tore open the envelope to read it. Every month, Steve Staley writes an editorial on some topic. Sometimes, his topics are presented from a positive perspective other times he gives his in what I will call very "frank" opinion on a particular subject.

This month his column gave, a somewhat, terse commentary on the usage GPS devices with special focus on using them during races.

Before going sharing my response, I respect Steve and his right to his opinion and the right to share his opinion. I feel that I have to. Otherwise, it would be hypocritical of me to share only my opinion.

As I said before this all started with Steve's column in the monthly club news letter. To which, I followed with my thoughts in an email back to him. In my email, I provided counter points to many of his arguments.

Steve sent back a reply yesterday where he provided more of what I will term circumstantial evidences on the guilt (i.e. bad idea) of using GPS devices for measuring distance.

I will pull out a couple of points here from his new letter and from his email and provide my response.

Anyone interested in reading Steve column, I would be happy to scan and send it to you.

As I start, here are a couple of assumptions. Steve indicated that he either uses or has used a GPS device and is familiar with how GPS devices function. My experience comes from using both the Garmin 305 and 310 models for the last 3 years in nearly everyone of my runs and all of my races: running and cycling(minus one).

Example 1, Steve makes a comparison between the "clicks" that measure distance on a Jones Counter (if you are not familiar, the Jones Counter is a device that fits on a bike rim and is used by certification professionals to measure race courses) and the points captured by a GPS (for purpose of my discussion, I will assume a Garmin). In his comparison, he describes a runner doing a ½ marathon. By a Jones Counter, there would be 211,290 samples taken vs. a Garmin which assuming a 5 second gap between points or samples would yield 1,248 samples.

First, I don't think this is a truly fair comparison. The number Jones Counter "clicks" are determined by a distance moved by the wheel of a bike. Each interval is assumed to be exactly the same. Such things as speed bumps, pot holes, and curbs end up skewing the measurement. This is why course measurers have to be extremely careful when certifying a new course. On the other hand, the Garmin's measurement is based over an elapsed time period where the distance travel is the measurement between the two sampling points. Yes, both will measure a distance. But how the samples are taken is entirely different which is why I say this is not an apple to apple comparison. One other note on this particular example, Steve suggested a 5 second sampling, but my Garmin can be tweaked down to a 1 second interval. This makes an even more precise measurement.

Example 2, Steve describes using a GPS device on a track and seeing the points scattered. Fortunately, the TrySports guys held a 5k track race last year and I used my Garmin to measure it. When I went back to review my race, I found the points to be pretty much along the inside lane of the track and not scattered at all. Conversely, I will admit these points didn't exactly make a smooth track oval. I attributed this to weaving from side to side in the lane as I passed other runners during the race and my own inability to run right on the inside lane white line.

From Steve's email response, he shared that after the Santa Scramble; he found runners with GPS reading of 3.04, 3.05, 3.06, 3.09, and I had 3.1. From the South Park Turkey Trot 8k, he found samples from other runners with .03 to .07 over 5 miles. Personally, I had .07 or 315.03 ft or 105 yards which equated to 17 seconds.

In Steve's email, he wondered which one was actually correct. My response here is actually "all of them are correct". Without knowing the runners, their devices, or how they ran the tangents along the course, the only assumption that I can make is that they followed roughly race course from the start to the finish. They all covered a measured distance of just over 5 miles. Without more details from these runners, no prudent person could make assert a course being long .

After years of racing my running will almost instinctively take me along the tangent between 2 corners or cones depending on how a course is marked for the race. Yesterday, I probably ran 98% of the tangents along the Turkey Trot course. And, I admit I didn't run all of them due to weaving between the baby joggers. In addition, I noticed that we ran the course a little differently than the previous year. After reviewing the USAT&F certified map of the course, we or at least I and the runners around me ran 3 turns wide due to the place of the cones on the course. These turns were at the following locations: At the corner of Assembly and Carnegie, corner of Colony and Roxborough, and finally at the corner of Policy and Morrison. Based on the certified course map, we should have run corner to corner rather than corner to cone.

In conclusion, I admit the Garmin is not perfect. It measures everything based on a straight line and the sampling points even at a second can have an impact on the overall measure. Taking a "turn" means depending on where the last point was captured, the Garmin could have measured across a corner and not around it. Clearly running on trails is a prime example. Numerous switchbacks can leave the Garmin measuring well short of the actual distance run.

But with in this assertion is the basis for most every runner's objections about long courses. A Garmin never measures distance long. After 5k, if the Garmin reads 3.15, then the runner at the bare minimum ran 3.15 miles. In actuality, they will have most likely run more than 3.15 miles. If he or she ran the course as they should have and ran the course along the tangents, then they have an argument for a long course.


 

Just my $.02 for the on going GPS debate.


 

Thoughts from the Cool Down Runner


 


 


 

 

Comments

DRL said…
Thanks for your blog, I've learned much from reading it. Regarding GPS accuracy, I recently read the following, which was a thoughtful (and humorous) discussion of the topic. http://www.hamptonrockfest.com/hamptonhalf-GPS.html
Thanks for sharing. I followed the link you provided and read the column. Interesting, I looked a the maps down to 80 and I don't see the jaggedness of the points. But I am using the 310 XT vs. The 305 so there maybe some improvement in the technology.

Definitely appreciate the info and if you find any more, definitely let me know.
CL said…
when I did the santa scramble 2 years ago, I had 3.02 for the course and after the time I ran I was a bit skeptical honestly. As far as the GPS, when I race I figure we all run the same race and I am usually going for placing, so I run the tangents the best I can. GPS is all over the place so I don't take too much out of it, It's just to keep a record for me, my 310 is awesome...keep up the good work bill, your doing well as usual.
Thanks for the feedback, CL. Yeah, Santa Scramble is always a toss up. The starting line seems to vary from year to year and how you run the tangents along the course can make a huge difference.

Hey, keep up the Good work down under.
DRL said…
Also, here is a recent LetsRun thread on the GPS topic. As with all things from the message boards, its filled with vitriol and anecdotal evidence. But its an interesting read nonetheless. http://www.letsrun.com/forum/flat_read.php?thread=3811594

Popular Posts