Somehow, Athletica thinks the average power for the ride was 139W, MP4 was 147W. I did not break up the ride or take prolonged breaks, about every 20 minutes I paused pedaling for 10-15 seconds in order to sit up and reposition myself.
Where is it getting this number from? It’s completely off?
Hi @ethilo… sometimes there is a delay in syncing and analysis between Garmin and Athletica backend (parsed data plus Athletica brain). Going into your account I note that Athletica reads your average power as 164w. Please check your end.
Hey Paul,
I manually entered 164 for the avg power and MP4. In the screenshot I had the values that the AI put in which were 139 and 147 I believe
Gotcha @ethilo … if this happens again, please document it for us as you did nicely and let it be so we can get in and fix.
I had it happen again, here is what I found:
Plan was a 1h50m zone 2 aerobic development ride on the trainer. I held a power on ERG mode anywhere between 165-185W, never went below 165W save for short breaks to take lactate tests.
Garmin data:
Immediate feedback:
I noticed the analysis graph was way off and totally scattered:
I then found the “smoothing” function at the bottom and clicked “disable” and this gave me a much more accurate picture of the data that was collected by Garmin:
I don’t really know why my power meter data has so many dropouts of data. I am using a Wahoo Kickr V6 and riding with Zwift. I don’t visually see these dropouts when I’m riding in Zwift but it looks like they are showing up in the Garmin data.
I think the data that Athletica is using to perform its analysis is after it performed the “smoothing” function that incorporates in these 0 Watt moments into the averaging.
Perhaps in my situation a moving average smoothing combined with a threshold filter (that eliminates nonsensical signal dropouts) or median filter would solve the issue?
