So the last few days, I've been trying to use multiple apps each time I work out. Earlier this week, I did a 4.5 mile outdoor walk using the Workout app and Runkeeper simultaneously. At the end, I found that Runkeeper was about a tenth of a mile ahead of the workout app, and said I burned double the calories. Then, I did a treadmill run a couple of nights ago that the Workout app said was 1 mile, but the treadmill itself said it was only .86 miles. Just curious of anyone else's results, and any thoughts you may have.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Workout app with treadmill or RunKeeper
- Thread starter justinharris89
- Start date
- Sort by reaction score
When you say that Runkeeper reported double the calorie usage... were you comparing Active calories on both - or total calories on both (I don't use Runkeeper so I don't know what it provides). It sounds like you were looking at Active calories on the Watch and Total calories (ie Active + Resting) on Runkeeper.
As for distance.... have you done a few outdoor calibration walks with your iPhone and Watch? That is how the Watch "learns" your stride length so that when you are walking on a treadmill it can give a good approximation of the distance you walked.
If you have done that... then the distance discrepancy between the two means that you most likely are using a slightly different stride length when you walk on the treadmill vs when you walked outside. In your case you would be using a slightly longer stride on the treadmill than when you walked outside... leading to a smaller number of steps (which is really all the Watch can estimate) for the same distance... ultimately leading to an under prediction of the distance traveled by the Watch.
What I do is I look at the Pace as reported by the treadmill and I periodically glance the pace on my Watch. If my watch thinks my pace is higher than the treadmill then it means I'm taking smaller steps than I normally do outside... so I lengthen them a bit until the two match. Similarly, if my Watch pace is too low then I shorten my steps a bit until the two match. This gives me really good agreement between my Watch and treadmill at the end of a workout.
Ultimately... you have to keep in mind that the Watch is tied to your arm and is trying to infer what's happening through an accelerometer. It's never going to be perfect. The best we can do is try to give it good training data and try to keep our strides similar to that training data...
As for distance.... have you done a few outdoor calibration walks with your iPhone and Watch? That is how the Watch "learns" your stride length so that when you are walking on a treadmill it can give a good approximation of the distance you walked.
If you have done that... then the distance discrepancy between the two means that you most likely are using a slightly different stride length when you walk on the treadmill vs when you walked outside. In your case you would be using a slightly longer stride on the treadmill than when you walked outside... leading to a smaller number of steps (which is really all the Watch can estimate) for the same distance... ultimately leading to an under prediction of the distance traveled by the Watch.
What I do is I look at the Pace as reported by the treadmill and I periodically glance the pace on my Watch. If my watch thinks my pace is higher than the treadmill then it means I'm taking smaller steps than I normally do outside... so I lengthen them a bit until the two match. Similarly, if my Watch pace is too low then I shorten my steps a bit until the two match. This gives me really good agreement between my Watch and treadmill at the end of a workout.
Ultimately... you have to keep in mind that the Watch is tied to your arm and is trying to infer what's happening through an accelerometer. It's never going to be perfect. The best we can do is try to give it good training data and try to keep our strides similar to that training data...
I've walked outside several times at a brisk pace, though I haven't done an outdoor run with watch yet. The different stride thing sounds about right, I was looking at the watch while running on the treadmill and it was giving me about a minute faster per mile pace than the treadmill was. I just returned from a treadmill run this evening, and the difference was smaller this time, which leads me to believe that it may get better as I go. It just freaked me out a little bit the first time when it was a large separation, hah. Thanks for the reply.
I find the Watch always displays significantly more distance than the treadmill. I did a 6.5 mile run on the treadmill yesterday and the Watch said 9.2 miles.
I wish Apple offered better calibration options (and yes, I have run outside with watch and phone for 1 hour, two different occasions and it made zero difference).
I changed my height information in the app too hoping it would estimate different stride length but it made no difference either.
Currently I can only trust it for heart rate on the treadmill which appears very accurate.
Such a shame.
I have a comical Apple watch running calibration theory... What if Apple had no clue that typical runners have a more efficient stride on a treadmill. And, what if they would have people run outside for calibration, but then nothing would match when the same runners ran indoors on treadmills. So, what if Apple just threw in an arbitrary calibration correction factor because they could not work out the error that they got every time a runner got on a treadmill? That would pretty much ruin the watch for accurate distance.
My watch distance accuracy is horrible outdoors, but I have not tried it on a treadmill yet. Given how easy it would be to accurately calibrate the watch with GPS, it is kind of mind-blowing that it is not spot on. Once I got my Garmin foot pod calibrated, it was within 0.5% of the GPS measure or treadmill distance. Apple cannot get the watch within 2% of a known distance in my experience.
My watch distance accuracy is horrible outdoors, but I have not tried it on a treadmill yet. Given how easy it would be to accurately calibrate the watch with GPS, it is kind of mind-blowing that it is not spot on. Once I got my Garmin foot pod calibrated, it was within 0.5% of the GPS measure or treadmill distance. Apple cannot get the watch within 2% of a known distance in my experience.
They went to a lot of trouble to have such horrible calibration:-
http://www.engadget.com/2015/03/20/apple-watch-fitness-lab/
http://www.engadget.com/2015/03/20/apple-watch-fitness-lab/
I agree. I read that as well-- that is the lab I was thinking of when the outdoor runners came back in to run on the treadmills. It is what led me to the comical error theory. Apple invested in all this stuff, but missed a simple data point that undermined it all.
Last edited: