GPS error?

[Sturmey_Hub]

Today I went for a short walk and took along all mapping devices.
Satmap Active 12 recorded 3.31km
Strava recorded 3.55km
Ride with GPS recorded 3.6km
Garmin 820 recorded 2.87km
That's a difference of approx 700m in just over 3km.
Which do I trust????
That's a massive error on a 100km ride.

 

[vickster]

Choose one and stick with it and enjoy the ride. They'll have different algorithms, different speeds at which autopause kicks in etc

For Strava... My phone has me going slower and further while the Wahoo measures faster but less distance. I mix and match to balance it out!

 

[Slick]

I believe whichever device I happen to be using at the time.

 

[presta]

Plotaroute.com: 14.32m
Cateye Mity: 14.36m (goes up to 14.39 when the tyres need pumping up)

 

[Ian H]

I've noticed that Strava appears to down-sample any file I upload. The differences are probably just down to how often each device records your position. When walking, your route is likely to be more convoluted than when cycling, so recording less often will have a more pronounced 'smoothing' effect – straightening and thus shortening the track.

 

[Sturmey_Hub]

I just found it surprising. When I'm cycling along, the Garmin's so accurate (or so I thought), that it knows whether I'm on the road or cycle path which is just 3m on my left.
Surely then, the further I cycle, that 3 or 3m becomes a smaller and smaller percentage of the total.
At this rate, my old Cateye Mitty 2 will be more accurate.

 

[welsh dragon]

I don't trust any of them as most around where I live record Incorrect distances. I know from point A to point B is 5 miles yet the distance between the various aps can vary as much as 1.5 miles. With that in mind I just use one (Strava) as a guide and don't take what it says as gospel. Good job i am not anal about the distance I travel.

 

[Supersuperleeds]

The Satmap as it is by far the superior product

 

[Milkfloat]

I just found it surprising. When I'm cycling along, the Garmin's so accurate (or so I thought), that it knows whether I'm on the road or cycle path which is just 3m on my left.
Surely then, the further I cycle, that 3 or 3m becomes a smaller and smaller percentage of the total.
At this rate, my old Cateye Mitty 2 will be more accurate.

The Garmin is the outlier. Does it have auto pause turned on and what is the sampling rate set to? I would suggest that when you are cycling you won’t see such a difference.

 

[Mike_P]

Also Strava can have differences recording off two different devices as I found out using a Wahoo Elemnt Bolt and my phone on the same ride. Some differences are also down to incorrect data held by the relevant body - an example of this on Strava is in a number of locations where it includes a decent down and climb back up a river valley that you have crossed over by bridge. Stick to whichever is the most convenient, although I do know of cyclists who choose whichever set of data is in their favour per ride.

 

[Pale Rider]

I believe at one time GPS data was deliberately skewed by the US Department of Defense to prevent those commies from accurately targeting their missiles.

 

[T.M.H.N.E.T]

I would take the dedicated cycle GPS to be the data point and ignore everything else inc apps.

 

[Sturmey_Hub]

It would appear that this is a well known thing.
I'll go for a longer ride tomorrow and see how it turns out - a flat one.
Thanks for all the comments.

 

[Ming the Merciless]

There's two things, knowing your immediate position, how many of those it saves into the GPX track. If you have smart recording on, the result will be less accurate as it throws track points away. So it'll give an accurate outline of your route but not necessarily distance.

What did you come up with, from measuring on a paper map?

 

[I like Skol]

I don't trust any GPS device to record distance with any real certainty.
The only absolute record in my mind is a wheel counting bicycle computer that has been calibrated against a known measured distance.