This discussion has been locked.

You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Pace not matching watch / GC / Strava

I noticed some strange behavior today after my run. I was doing a fartlek going from a slow relaxed jog into a faster interval (i used my lap lap button for). The watch was showing a 4:19 m/k average for the 600m lap, i pressed again and jogged on in a different pace.

After the run the laps were shown properly on the watch and the app e.g. with a 4:19 average for that lap. The Graph however shows absurd momentary pace(s) when hovering over with the mouse over the graph on GCM and GC Web. They could never add up to that average. Strava shows other momentary paces, some that make much more sense. Maybe this is not a 945 problem, more a Connect issue? Has anyone experienced this? I unfortunately don't how strava and GC get the momentary paces (from the same watch) - maybe some is able to shed some light on this. Had this experience twice on a fartlek coming from slow to faster intervals. Not so much when doing intervals with less of a speed difference between work/rest.

0 Garmin-Blake over 5 years ago

Would you be willing to post public links to these activities or screenshots?

0 DanielM over 5 years ago in reply to Garmin-Blake

yes, here are some screen grabs :

there is another inconsistency in connect: its showing the best pace for that lap lower than the average. this highest pace is shown in the sreengrab at the corner of the field.

first one is showing the best pace at 4:27 at the corner of the field, strava is showing 4:02 m/km.

i just have it setup that connect pushes the activity to strava - there shouldn't be different values, should there?

0 DanielM over 5 years ago in reply to DanielM

videos are not shown for me, here a link to dropbox:

https://www.dropbox.com/s/k93nv34rxba13sr/Bildschirmvideo%20aufnehmen%202019-10-04%20um%2007.58.54.mov?dl=0

https://www.dropbox.com/s/fmqb0sj85rsuvtw/Bildschirmvideo%20aufnehmen%202019-10-04%20um%2007.57.03.mov?dl=0

0 AllanOlesen67 over 5 years ago

DanielM said:
The Graph however shows absurd momentary pace(s) when hovering over with the mouse over the graph on GCM and GC Web. They could never add up to that average.

The average of instant paces (or rather velocities) during a session will quite often not match the reported average.

As far as I can see, the reported averages are based on accumulated distance, not on instant pace. The accumulated distance per session or per lap will often be different from what can be calculated from the logging of instant pace.

For fun I made a custom metric in Golden Cheetah, which calculated distance as an integration of time*velocity and got differences of more than 20% for some splits. This usually happens when running under tree cover where my instant pace will appear slower, even though my accumulated distance will still be true.

So the answer to your observation can probably be found in differences in the algorithms for turning GPS data into instant pace and into distance.

0 DanielM over 5 years ago in reply to AllanOlesen67

Thanks for explaining - so the averages are more accurate, thats what i figured.

How does that explain the differences between GC and Strava? Are they using different algorithms from the raw data from the watch?

0 AllanOlesen67 over 5 years ago in reply to DanielM

DanielM said:
How does that explain the differences between GC and Strava? Are they using different algorithms from the raw data from the watch?

I have never dived into that. I have a Strava account, but I am not using it much.

But as far as I know, Strava can do some kind of distance correction where they apply their own distance algorithm to the raw GPS positions. However, I don't know if it will cause correction of both pace and distance, and I don't know if it can be enabled/disabled by the user or is always happening.

0 DanielM over 5 years ago in reply to AllanOlesen67

AllanOlesen67 thanks for answering - maybe Former Member can shed some light on this?