Question Re Calibration/Running Discrepancy

[iPhysicist]

I saw a good article on cult of Mac and how poorly the Apple Watch performed without the iPhone along for the ride. I calibrated a bunch of times but run or walk the same route so the watch by itself has been accurate. When I do intervals with more pronounced arm movements it seems to start to be off. I normally barely move my arms when I run and it seems accurate.

I would never trust the Apple Watch for serious running though.

You have to calibrate with different pace or styles of running. That includes your arm movement. Stride and arm movement belongs together.

I don't know how you think calibration works. Apple states to do different paces to get better accuracy. I did it with 6'/km down to 4:30'/km and don't have issues with paces in between those marks. I always run the same route and got always good accuracy.
The arm movement most likely is the culprit in your case because that is what causes my accuracy to drop whenever I do uphill running. Fortunately it compensates a little when going downhill but yes I assume the arm movement is taken into equation for calculating stride length. When you look at professional runners and their arm movements I would say it is correlating to their stride and running style.
The Apple Watch reads the accelerometer to learn your arm movement that correlates to your stride. The GPS from the iPhone the gives the needed path length to do the rest of the math.

That is - in my eyes - why the reading is off for some runners that have inconsistent arm movement for different velocities.

 

[eastamherstbias]

Calibration Sucks, I run on a measured course every day. It is 4 miles. The watch always logs 3.78-3.82. It blows.

 

[exxxviii]

Update: it is as bad as it ever was...

I ran 28 miles last week with my watch and phone together. The cumulative difference between my AW and my Garmin was 0.02 miles. The biggest discrepancy in a single run was 0.03 miles. Overall, that is pretty freaking awesome. But...

Today I ran without the phone. My Garmin measured 5.06 miles (on a 5.07 mile course), and the watch measured 5.66 miles. This discrepancy is consistent with what I have been experiencing since the wOS 2 update.

 

[eastamherstbias]

Update: it is as bad as it ever was...

I ran 28 miles last week with my watch and phone together. The cumulative difference between my AW and my Garmin was 0.02 miles. The biggest discrepancy in a single run was 0.03 miles. Overall, that is pretty freaking awesome. But...

Today I ran without the phone. My Garmin measured 5.06 miles (on a 5.07 mile course), and the watch measured 5.66 miles. This discrepancy is consistent with what I have been experiencing since the wOS 2 update.

Super annoying. Happens to me daily. My garmin was near perfect. The apple watch and iPhone together or separate are always massively off.

Is there a way at the end of a run for it to ask you to calibrate?

That would be super helpful.

 

[exxxviii]

Is there a way at the end of a run for it to ask you to calibrate?

The AW is supposed to dynamically automatically calibrate itself using GPS data from the phone when the phone is present.

 

[BarracksSi]

Is there a way at the end of a run for it to ask you to calibrate?

That would be super helpful.

Are you asking if you could manually input the distance you knew you traveled, so the watch could calibrate to your step size?

I remember Nike+ on the old iPods could do this (although I never figured out how to do it on my current-gen Nano). Either that, or you could tell Nike+ you were going to jog a chosen distance, then you'd tell it "stop" when you finished.

I think it would be cool, too, to bring such a function back. But, it also adds a bit of complexity to a device that's trying to be simple to use.

 

[exxxviii]

I think it would be cool, too, to bring such a function back. But, it also adds a bit of complexity to a device that's trying to be simple to use.

And, adding this feature would effectively be a concession that Apple's dynamic calibration does not work well.

 

[eastamherstbias]

Are you asking if you could manually input the distance you knew you traveled, so the watch could calibrate to your step size?

I remember Nike+ on the old iPods could do this (although I never figured out how to do it on my current-gen Nano). Either that, or you could tell Nike+ you were going to jog a chosen distance, then you'd tell it "stop" when you finished.

I think it would be cool, too, to bring such a function back. But, it also adds a bit of complexity to a device that's trying to be simple to use.

Well it's trying to be simple to use and not working correctly. The Nike App at the end of the run lets you modify the end distance.

 

[eastamherstbias]

And, adding this feature would effectively be a concession that Apple's dynamic calibration does not work well.

Well it doesn't work correctly. They could just enable this to be a variable in the formula.