Distance totally wrong

I am also experiencing this issue with the Garmin edge 830 that I purchased just over 1 week ago. Yesterday I went for a hike and it recorded 4.25km, today I did exactly the same hike, same weather and it recorded my distance at 5.5km. iPhone using Gaia gps recorded the route at 7.75km. I will plot the route manually later in base camp but I expect the Gaia distance to be correct. I was walking at the same pace and the same total time was recorded. Also I notice that my max speed was recorded as 23km/h. It seems like I am not alone with this. I was mostly walking in dense trees but I was on designated trails that in summer will be MTB trails. I specifically purchased this unit as it was marketed and advertised with MTB route credentials. In my experience mountain biking goes on in densely wooded areas often at slow speeds. Why can’t a 400 euro product that is specifically designed for the purpose of route tracking and recording do what a free phone app has no problem doing? The phone app also has far better definition of my local area and actually shows the mountain bike trails - the Garmin does not. I will check the settings and repeat the route tomorrow. If accuracy is not up to par I will return the device as it is an absolute requirement that this can match a phone for accuracy as that is essentially what I am looking to replace. At the moment this is feeling like a road riding gps unit that has been repurposed as an MTB gps thanks to the addition of some pointless jump and ‘flow/grit’ metrics? I won’t be adding a wheel speed or cadence sensor to fix this. Garmin response??

Today, 2nd outdoor ride the same, but longer stable. It started after carriing the bike upstairs. Lost ~15km of 58 km. This time when switching the unit off/on, it was working again for the rest of the tour. Only the cadence sensor (of the spd/cad unit) is used.

Ok, first of all, it's a cycling computer.  But did you disable auto-pause, and did you have the recording interval (System > Data Recording > Recording Interval) set to 1/sec (instead of the "smart" rate)?  Otherwise, the Edge will only record at longer intervals and will miss a large amount of your distance, particularly since there is no speed sensor to tell it that you are moving.

Hi, thanks for the reply and appreciate the setting advice. Interval is set at 1 sec and auto pause is off. Agreed it’s a cycling computer, that is what I purchased it for. However 3 days after getting it we got a meter of snow so it’s getting tested on foot right now. Even though I am on foot I have 2 huskies strapped to me so we are moving at a good rate and trekking the same trails that I would on my bike - In some areas faster. I have not yet tested the integrity of the gps recording by exporting and re-importing I will do that later just for curiosity. The issue that I have is that the device has recorded 2 identical routes with a resulting distance that is 20% off. My phone, that is not a cycling or hiking computer can record this with a higher level of accuracy and the tracking on that is sideline not it’s main function. The idea of ‘needing’ a speed sensor to achieve an acceptable level of accuracy is honestly not what I expected from a premium device. Maybe for dialing in the last 2-3% but not 20-30%. I could be jumping the gun a bit as I have not run this route with Galileo + gps yet. I will try that tomorrow but I am not holding out much hope. 

I went for another hike today with the edge 830.  This time I changed to gps + Galileo. I measured with both phone and Garmin. This time on almost the same trail as yesterday the distances were 0.12km apart. Based on experience the distance recorded was definitely in the right ball park. This is a positive result. When I exported the route and imported it back in I got a distance that was about 0.6km higher than both phone and Garmin reported. I notice that on the Garmin track there is a hell of a lot of ‘noise’ on the track. By that I mean it is very jagged. This looks far more pronounced when in heavy tree cover so I guess it’s due to the gps signal being shadowed by the trees. It seems like the software is ironing out some of this noise which is useful. I will try the route again tomorrow just to see if the accuracy repeats. The trails are also looking packed enough to ride so I may also try the same route by bike to see what the dense tree areas look like at bike speed. I am hoping this level of accuracy isn’t a fluke

Note that there is also a current problem with GPS location offset caused by a data problem with Sony GPS chipsets (as explained by DCRainmaker).  It's possible that your earlier efforts were subject to this problem.  It doesn't affect iPhone (if that's what you have) because Apple uses a different chipset.

Thanks for the info. That was an interesting read. I don’t think that was the cause of the early issues as the route recorded on the phone and the route on the 830 were pretty close (other than the 830 having it’s very jagged line). It seems like there is/was some conversation issue between the FIT file and the gps track as the raw gps track file was more like the correct distance. However it then makes no sense that switching to gps + Galileo would have any effect. I will run some more tests on the same route on and off bike to see how things are shaping up. 

I'm having similar issues.  Following and trying the suggestions.  I'm on a Fat Bike and noticing distance off.  

My lunch ride was open farm fields.  Garmin recorded 3.91 km and Apple Watch (6) with iphone 8 plus recorded 4.03km.  I was set to Smart and GPS + Galileo.  Just changed interval to 1 second - will see tomorrow if that helps.

If you have a wheel sensor, and you only rode 4.03km, the distance may not be calibrated to the circumference of your fat-tires.  You can measure this and set it directly (I think) or if you ride far enough it might get automatically calibrated.  I suspect the manual has better information than me.

As an aside: I'm usually not that interested in distance (or the precision of distance).  More interested in accurate elevation gain, power numbers, and Strava segment times.  But I only ride on the road, where distance/speed are not key metrics (whether accurate or not).