Accuracy of Running Apps – What Gives?

As you probably know, I do a lot of running and training for races with my friend Matt. (You can find him over at: http://thebigrunner.wordpress.com/). We use different apps for our iphones to keep track of our running.  He uses an app called Strava and sometimes his fancy GPS watch. I have always used RunKeeper.  Please pause for a short disclaimer.

Disclaimer: I realize that while training for a race, pace is not the most important thing (unless you are trying to win or qualify for something, in which case, it is the important thing – and this post is probably not applicable to you). I know that a run can be faster or slower depending on weather, your mood, whether you are sore or not, etc.) For the most part, I run because I like to run. 

HOWEVER, it absolutely drives me nuts that our apps *consistently* show different information at the end of our runs, which we run together. I’m not talking about a few seconds here or there, but more than 1/4 of mile differences and pace differences of up to 1 min or more per mile.  Here is a perfect example of the differences for a 5 mile run that we ran together 2 Sundays ago. (Definition of together: When I say together, I don’t just mean we started together. We actually ran the 5 miles within 1/2 block from each other the entire time. Most of the time, we were right next to/in front each other.)

image

Matt’s Strava app screenshot of our run

image-1

My RunKeeper app screenshotsimage-2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Due to my teacher nature, I’d like to start this by asking you to compare the two versions of this run (that was run together: see definition of together above).

Here’s what I noticed:

1. Strava counts “moving time” and RunKeeper counts “duration”.  “Moving Time” is nice because it accounts for traffic lights.  17 seconds shouldn’t make much of a difference in this particular run, however.

2.  Strava says we ran 4.9 miles, while RunKeeper says 4.79. Again, not a major difference… but annoying nonetheless, especially when we were on roads and a trail. No stairs, no switchbacks, nothing tricky.

3.  He burned A LOT more calories than me. Stupid boys. (This is irrelevant.)

4. His average pace: 10:54. My average pace: 11:15

5. Our splits were VERY different, especially miles 2 and 4. Mile 2- 9:40 pace vs 11:04 pace. Mile 4 – 10:37 pace vs 12:38 pace.

Now, WHY is this?  That 4th mile really bothers me, since I was even a little ahead of him at that point.

If you look closely at his splits, they don’t really make sense. How is his average pace more than any of his splits? And if you add up his splits, they only add up to 51: something, and not the 53:39 of total moving time.  My average pace makes sense considering my splits, and my splits add up to about 2 minutes more than my total time, because it’s accounting for a whole 5th mile.   So, maybe Strava is incorrect?

For the most part, I think that RunKeeper’s problem is that it doesn’t account for traffic lights, which there can be a lot of in Philadelphia. So, I downloaded one called iSmoothRun that has an auto-pause feature. Anytime you stop for 5 seconds, it automatically pauses it and then deducts that 5 seconds from your time. It restarts when you begin running again.  Well, even today – we ran an easy 3 miles with my sister – and our run still wasn’t in sync.

Untitled3

My run – iSmoothRun

Screen shot 2013-08-14 at 7.31.32 PM

Matt’s run – Strava

 

Distance was pretty similar, but Matt’s was still longer. Even with the auto pause feature, his moving time was significantly less than mine, and his pace was lower.

Untitled

My run – iSmoothRun

Screen shot 2013-08-14 at 7.31.44 PM

Matt’s run – Strava

 

 

 

 

 

 

 

 

 

Our splits were off too, first the first and second mile by almost 30 seconds!

SO …. what gives? Is Strava inaccurate? Are they all inaccurate? Thoughts? Experiences?

Advertisements

6 thoughts on “Accuracy of Running Apps – What Gives?

    • Thanks for the tip! I didn’t realize that was possible; it must have been an update or something along the way.

  1. First off, excellent post! Well written and you bring up some great points. As your friend and fore mentioned running partner, I believe every running app will show some degree of variation on splits, pace, etc. let me first point out that I never gauge the calories burned on any app. A year ago I entered my weight in Strava so it is going by that. I always figure I burn 100 calories per mile. I’m also curious to see what would happen if the next time we ran I used MapMyRun just to see how far off the numbers would be.
    Today’s run was a bit closer considering I stopped mine a good 30 seconds before you did. Anyway, it was a good run!

  2. I bet you’d still get different results even if you used the same app. For one thing, I think the apps aren’t exactly running all the time or they’d drain your battery too fast. So they wake up and have to get position from GPS every once in a while. So it’s sort of an approximation of your run. If you want to find out which app has the most accurate results, you could try running a known distance (e.g. between mile markers on the Skuylkill trail) and timing it with a watch, and do this more than once so you can average it out or exclude outliers.

    • That’s a good idea, actually. Maybe I”ll try to test it against some known distances. You really think that the app isn’t running all the time? Interesting idea that it’s really sleeping most of the time.

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