A Second Crack at the NYC Marathon (Part 2: Lies, Damn Lies, and Your Phone’s GPS)

In part 1, I took a look at my 2014 approach to the NYC Marathon. Here in part 2 I’m going to take a look at something pretty significant I didn’t talk about last time, which is GPS accuracy.

During the 2014 training, although I had a sports watch, I actually used an iPhone app called MapMyRun and the iPhone’s GPS to track all of my runs. I used my watch at the track and during races, but otherwise this app was king. The reason is that I’d been using it since I started running, and wanted the continuous record to track my total miles, improvement and share with the community on Facebook. All of this is possible simply by syncing MMR with your watch or other apps, but at the time I didn’t know that. So, why is this important.

Your iPhone’s GPS is Lying to You

OK, it’s not just the iPhone… pretty much any phone’s GPS is a bare-faced liar, and certainly not as accurate as a dedicated GPS watch; even a fairly low-cost one. Let’s take a look at two examples.

Here is what my GPS thought I ran
iPhone GPS Measures This as One Mile

This is typically what a phone GPS thinks you’re doing in an urban area. Buildings, bridges, cloud cover and other things interfere with the GPS signal. Not only that, but iPhones and some other phones use wifi signals, and even your mobile signal to estimate location, neither of which are as accurate as a GPS signal on a proper GPS device, despite what the marketing tells you.

Mile 4 (Real)
Measured Distance I ran

Above you can see the true distance I ran, which takes out the curves and wobbles, and is actually 0.94 of a mile. Usually in my street running, I found my phone to be anywhere between 0.02 and 0.07 per mile off. Over a typical training session of 6 miles, that’s about 0.35 of a mile, or about 5 – 6%. So how significant is this? It’s huge!

Let’s say you record a time of 57:10 for your run (like I did in the above example). If you’re following a training plan, or your coach (especially a virtual coach or one you correspond with over email/Skype etc), it’s the difference between running 6.35 @ 9 mins/mile, which is what your GPS is telling you, Vs. actually running 6 miles at 9:32/mile. So not only do you think you’ve run further, you think you’ve run faster.

OK, so what if you’re running laps in a park, or even a track? Is it more accurate then?

Mile 10 Central Park (GPS)
Last mile of a 10 miles run. My GPS happily tells me I’ve run it in 9:34, which was at the faster end of my target that day.
Mile 10 Central Park (Measured)
Really I had run 0.94 miles, so my 9:34 min/mile was closer to 10:11 min/mile.

As you can see from above, the compounding effect of time and distance is quite significant, and if you’re giving feedback to your coach, the times are going to be misleading. During marathon training, the combination of your regular run (5 – 8 miles) and long runs (10 – 22 miles) are usually done at or just below your predicted marathon pace. Its a self feeding cycle designed to get you to the time you want, but also to see if your body is really up to it along the way. Inaccurate timing is going to mess this up.

Given the stage of my training I was at, and that it’s the 10th mile of a long run, this pace was supposed to be an indicator of what my pace should have been. At 9:34/mile, I would complete a marathon in 4:10:39 (which is what my predicted time was, within a few mins). At 10:11/mile, I would complete a marathon in 4:26:48 (much closer to my actual finish time).

Honest Garmin

OK, so is a dedicated sports watch any better? A decent GPS watch is going to set you back anything from $250 to north of $500, and given a top end smart phone can be had for lower end of that scale, it’s tempting to not bother. But you should reconsider. The GPS tracking on your watch, even if you have a ‘budget’ GPS watch, is likely far more accurate than your phone. First of all, your watch will actually wait until it has a GPS lock until it tells you to start running. Sometimes this can take up to a minute. You probably thought this was your watch being a bit crap, because your phone locks on immediately, right? Wrong. Your phone only tells you when it has a GPS signal, which might be from only one satellite, and just because it has a signal, it can’t necessarily tell you where you are located exactly. Your watch on the other hand will know where you are to more or less within 30 feet, or about 10 meters. Don’t quite believe me?

Here is a workout from today… a 10 mile long run in Brooklyn’s Prospect Park is the ugly heat of July, and I’ve cracked mile 3 in a humbling 9:40

Prospect Park Mile 3 (Garmin GPS)
Mile 3 of 10 in Prospect Park, Brooklyn, today.

Using the same mapping tool to check the accuracy of MapMyRun on the iPhone, I measured this distance, which came out to be 1.01 miles. If anything, my Garmin is robbing me, but the difference is very small.

Prospect Park Mile 3 (Measured)
Garmin stole 10 meters!

On a track, and using Garmin’s own mapping, I see the the same thing. Earlier in the week, I did 800m intervals on a track (2 laps). Here’s the distance my watch recorded for those intervals (I started and stopped the timer exactly as I crossed the line after 2 laps):

Garmin 800m Splits
The .49 on the far left is the split distance. 800m is (to be very precise) 0.497 of a mile. Again, Garmin is robbing me of about 10 meters!

So, in conclusion? Don’t believe your phone! If you’re serious about running, you probably already have a dedicated watch, but you may be leaving it at home on occasion. If you don’t have a watch, think about getting one. They’re getting better and better (I have a Garmin 225 and think it’s fantastic, more on that another time) are easier to look at while running, and actually do a better job of telling you what you’re doing. Most watches from most companies allow you to export your workouts and import them into your favorite app (MapMyRun, Strava etc) and some (like Garmin) allow you to sync everything without ever having to click a button.

While I don’t expect necessarily to run any faster this year, simply because I am using a more accurate GPS, I do expect my marathon prediction to be more accurate. In part 3, I’ll crack open the training plan and see what’s in store.

Happy Running!

Advertisements

One thought on “A Second Crack at the NYC Marathon (Part 2: Lies, Damn Lies, and Your Phone’s GPS)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s