Apple Watch running accuracy

I agree. Being able to manually calibrate the distance after a run would be perfect.

One thing I have noticed is that the pedometer / number of counted steps is very precise. Many times I have been running the same routes, only wearing the watch and not iPhone with me, and the number of counted steps are always within a 0,5% range. I almost run at the same pace every time and choose between 5-10 different routes, depending on wind and weather. This means that if I could be able to manually set my step length somewhere the watch would be PERFECT for me and it would be very consistent in measuring the distance.

My step length is shorter than most. I run with a 185-190 steps pr. minute where normal runners run at 120-130 steps pr. minute. This means that my Apple-watch-alone measurements are 20-25% off.

I really wish Apple would change this. It would give me opportunity to go off-route and rund into unknown paths and have a good idea of how long I have gone that day. Today I'll have to bring my iPhone (which can only happen on dry days and not rainy days) or guess the distance based on time and perceived pace.

"Today I'll have to bring my iPhone (which can only happen on dry days and not rainy days) or guess the distance based on time and perceived pace."

I always put my phone in a zip-lock plastic bag when I run (and carry it in my hand). I sweat too much and it would get pretty wet even on a sunny day. Doesn’t solve the problem with accuracy without the phone, but at least you could get “fairly” accurate distance on rainy days. Low tech (but less than ideal) solution.

I wish sweat was the problem where I live in Denmark. :-)

I do have a zip-lock plastic bag, but it is too big to fit in the pocket in my Camelback running backpack. I have an arm-band too but it is not waterproof and I prefer the backpack while running.

Also on a sidenote would believe Apple DO care about runners and that is why I have big hopes that the workout app will improve in future versions. At least the whole marketing about running with the Apple Watch, was the major reason I got myself an Apple Watch. I remember following the blog from Christy Turlington on the Apple website.

http://www.nydailynews.com/blogs/running-dialogue/christy-turlington-burns-runs- apple-watch-blog-entry-1.2143214

I also saw a documentary from the fitness labs at Apple where they made a lot of testing and had people workout while doing a ton of measurements etc. So fitness and running has to be a top selling point for a SPORTS watch like the Apple Watch Sports edition. Now I cannot find the video itself - but maybe you can recall it?

http://appleinsider.com/articles/15/03/20/apple-offers-rare-inside-look-inside-s ecret-apple-watch-health-and-fitness-testing-labs

vpnogueira wrote:

I certainly think that 2%-3% is too much if you are in an Outdoor Run with an iphone, but let's put aside the Outdoor Run for a moment, since we don't seem to agree.

I admit that my expectation for phone GPS accuracy (2-3%) was based on my experience and not rigorous data, so I did a quick search to see what info may be available. Here was an interesting summary about various devices/apps tested on a track:

http://bit.ly/10RKhIo

Basically, they found most were better than 3% and averaged about 1.5%. A significant factor was how frequently the app registered a GPS location (more frequent points have a tradeoff of faster battery drain). You may want to try different apps that may use a more frequent GPS sampling frequency to get better results. Apple (and every other app developer) makes a decision on accuracy vs battery drain. Apple probably leans toward battery life because they will take the blame if the battery drain is too fast. I know a popular iPhone running app was notorious for rapid battery drain and i suspect they were sampling much more frequently.

Another article:

http://bit.ly/1RcLDU4

This article emphasizes that the accuracy depends largely on the “view of the sky.” “View of the sky” determines how many satellites can be viewed at a time (more satellites, greater accuracy). Buildings, topography and even trees can impact this. They state that a 99% accuracy can be obtained with a clear view of the sky, but they also say that 95% accuracy is more typical (this statement is from Garmin, the GPS manufacturer). I suspect that the first article had a “clear view of the sky” and was located in ideal conditions.

Neither of these articles point out that another significant variable is your location on earth. Some parts of the earth have more satellites in view. Others (like where I live in Alaska), have fewer and GPS is inherently less accurate. Again, these errors can be greatly reduced with specialized surveying GPS units (and processing). I am talking about only about phone based GPS units.

Neither of these articles are really “scientifically rigorous” in nature, and there were many other articles I didn’t look at. However, based on this, I’d still say an accuracy of 2-3% still sounds about right to me.

Dave

burning777 wrote:

I wish sweat was the problem where I live in Denmark. :-)

Ha!

I am even further north here in Alaska (61 degrees north vs 55). I just have a real problem with sweat!

I have seen those articles, but not the video you spoke of. I assume things will improve with time, but Apple always likes to make things simple and not have too many extra settings that may be better for a specific purpose (like calibrating indoor vs outdoor runs differently). Once you try to optimize settings for every possible workout, the permutations can become unmanageable. In other words your (or my) specific wish for improvement may never happen. Apple leaves many of these options for 3rd party app developers.

Also, I just got used to carrying my phone in my hand (in a ziplock bag). I don't run far enough to carry a pack with me!

Dave

Alaska - ok that sounds COLD :-)

I would be happy with a 2-3% accuracy. But for me when I bring my iPhone with me it measures perfectly. The distance is the same as when I use runkeeper or any other app. Small variations at least. I can live with that. Also I have tried measuring the distance of some of my routes with my bike which has an odometer. It confirmed the Runkeeper distance.

The problem here is that the watch alone will not calibrate to the measured distance. My watch simply forgets my step length when I do not bring my iPhone with me. So even though I go through the calibration proces over some time, over 20 minutes, over several short and long runs - the next time I go without the iPhone my watch alone distance is 25-30 % off again. The number of steps As seen in the bottom of a day in the activity app on the iPhone - and at the end of a workout on the watch, is always very close to the number fo steps when I ran the same route the day before on the same pace.

Why will the watch not learn the distance? That is the real error here. Let's hope that they will just fix the watch workout app so it will learn from the GPS distance and calibrate the step-length.

David,

In your comments, you keep focusing on Outdoor Run with the iPhone: "You may want to try different apps that may use a more frequent GPS sampling frequency to get better results". We are asking that Apple fix their Workout app, because the calibration error level for distance/speed is completely unacceptable, often 10% or more, if one is running without a GPS/iPhone. Their calibration simply does not work properly so they have to fix it, quickly.

The simplest CORRECTION FOR THIS MAJOR ERROR is to let the user calibrate the accelerometer, inputting the distance after the run. If this does not solve the problem, because maybe a watch/wrist based accelerometer can not measure distance correctly, then Apple should enable Workout to read a foot based Bluetooth accelerometer such as the one produced by Polar.

VPN

My point:

I was responding to the comment that stated 3% accuracy (with the iPhone GPS) was “unacceptable.” Even Garmin (GPS manufacturer) says typical accuracy may be 5% in real life conditions for running. People who expect the Watch (without iPhone GPS) to be better than 5-10% accurate will be disappointed. If you get 2-3% accuracy while using the Watch with the iPhone (as I do), you are doing pretty good. Getting 10% accuracy without the phone sounds about right to me.

Your point:

The Watch will need to have access to other technology (like an ankle accelerometer) if you expect to improve on the 10% accuracy (or allow you to enter a “fudge factor”).

I couldn’t agree more.

I read the most parts of this thread. I'm too facing some difficulties when I go for a run.

I even think it has an bad effect on the daily measurements of the calories and activities.

And even after updating to OS2.1 I think I still have the same issues.

I Waiting for all your experiences n

Well, after months of trying and being frustrated with Apple, I have recently given up on using the Workout App for the Apple Watch. I am really dissapointed that Apple marketed the Apple Watch as a running watch, and it isn't. It is quite inferior to real running watches. One can not manualy calibrate the distance measurement, the accelerometer in the wrist seems really bad (10% error!!) but not sure because I can not calibrate it, the distance/speed for outdoor runs is often bad to and it does not show in a map like other apps, and one can not manually set splits while running (for example in a regular 400m running track).

I have, unfortunately, given this lack of focus on runners from Apple, back to using other solutions for both indoor running and outdoor running. I hope that at some time in the future Apple fix this product and make it a competitive product for runners, in line with what Apple had advertised.

Regards,

VPN

Im having a terrible time getting running apps to work properly. Ive tried Nike+ running, which in the past have found my iPhone to be very accurate. But I can't even get the watch to register the run if Im not holding it. This is the same with Runkepper. From the research Ive done. The Apple Watch is very inaccurate in tracking which is highly disappointing, considering half the reason I bought it was to use as a fitness tracker.

After reading this post, I am no longer buying an apple watch.

I'm a strong supporter of apple products, and recently wanted to switch from the Polar products to the Apple products to complete the homestead of apple products.... but with such in-accuracies in the apple watch for runners, its a no go for me! I had it in a cart online, ready to buy!

I live an area (North Eastern Canada) where running outdoors is not an all-year long activity. Most times your lucky to 6 months of it. I need something that can be used for indoor training, with accuracy. So far I find my Polar products very reliable and accurate. I'll stick with it.

Toxic1979 wrote:

So far I find my Polar products very reliable and accurate. I'll stick with it.

Sounds like a good decision. If you only want it for exercise, there is no need to get a general purpose wearable like the Watch. Stick with a device that specializes in exercise and you know you are happy with.

I like the Watch because I use it for many other things, but it also does what I want it to for exercise. Even though I use the Watch every day for exercise, it probably only represents about 20% of what I use it for.