Running Distance Usually Short?

Hello, 

New Fenix 6 Pro owner here.  I've noticed that when using the device to record a Run activity, the distance measured is always a little shorter than Strava on my phone records the same run. 

I've set the GPS to record every second, and looking at the track on a satellite map, GPS accuracy seems quite acceptable.  Certainly the line sometimes deviates from the road a bit, but for a watch, accuracy seems darn good.  The GPS course follows the actual route, turns around at the correct places etc. 

So what I don't understand is, with these sorts of small deviations, I would expect that if anything, the watch would be measuring a slightly LONGER distance than Strava (which is using the better antenna in a cell phone, and therefore shows a straighter track with fewer deviations).

So why is the Fenix measuring a shorter distance? 

I've found other threads discussing this but they seem to mostly refer to the "trail run" activity.  I'm seeing this behavior on the regular "run" activity. 

Thanks for any insight! 

  • At a recent parkrun, it seemed almost everyone was wearing Garmin watches, and while checking out other runners Strava feeds, I'd say the vast majority were coming up around 3.07 - 3.09 rather than 3.11 - they claim it's measured with a measuring wheel and I see no reason to doubt them, its a nice out and back on a decent tarmac track with not much scope to wander off course or even deviate much. I couldn't see an obvious correlation between the Fenix 6 being short vs. other Garmin watches.

    I can't remember where I heard this, but one course owner here in the UK added in a bit extra as Garmin users were always complaining his course was too short (it wasn't apparently)

  • Or use a Forerunner.

    ...Or buy other brand of sports watch. The only way Garmin will ever notice something's up is at the quarterly financial reports.

  • Has this same problem been reported on the new Forerunner 945 as well? Anyone knows?
    I've been short up to 300m on a measured 5k track, and that's just bloody ridiculous.

  • Whats your GPS settings, as if anything I'm seeing a pretty good correlation - Strava on the left if anything is slightly shorter?

    (BTW There's a mix of "run" and "trail run" here.)

    (GPS Only, one second recording here)

  • (GPS Only, one second recording here)

    FWIW the GPS setting of 1s or Smart only affects the plotted track. It has no bearing on the distance or pace accuracy.

    What is Smart Recording...

    And my results generally agree with 

    And applying Strava's so called correction changes little either.

      

    As always, it works for some it appears not to for others. But also some people do have unrealistic expectations. 

  • Just sorted my activities by distance, working on the assumption that should magnify the differences, and yes for me Strava does tend to be shorter, Only problem is, some of these will be with my Fenix 5x+ but thats for activities < Dec 2019. Only one or two in this list
     - Just noticed the Strava list is slightly longer! Sorry


    If anyone notices some of these have different times by a way, they will be marked on Strava as races so will show elapsed time rather than moving time. When you mark an activiity as a race on strava time paused (such as at Checkpoints) still counts towards your final time (as it should) but Garmin shows moving time))

    OK EDIT: Yeah in the strava activity list it shows moving time, garmin always shows elapsed time in the list - so the above is bass ackwards lol)


  • I'm not sure that's exactly true... the GPS pos seems plenty accurate.  Any spot position check I do seems quite acceptable, considering a wrist-mounted antenna.  Plenty acceptable for navigation and track recording. 

    No, it doesn't seem that GPS accuracy is the problem.  It seems that whatever the Garmin software is DOING with this position data is the problem.  Which means the fix isn't complicated; it's just a software tweak.

  • And as someone mentioned above, I think the Fenix is optimised towards walk/hike/trail run/ultra - naturally slower paced than runs under <20K

    What would be interested is if someone run for say a month with a Forerunner on one wrist, and a Fenix for another - would definitely identify an algorithm issue

    Garmin, If you're interested I'll PM you my address for a free F945 :)