Edge 520/Garmin Connect and Strava shows different values for average power

The problem is on the Edge. The activity summary record shows average power for the ride at 141 watts.

Try changing the setting on the Edge to exclude zeros and then back to include zeros to make sure that the value is set correctly.

I did notice that the radio reception rate from your Stages power meter is very low and this could cause issues with the quality of the data, but it would not affect the way average power is being computed.

I will try to set the settings to exclude and then to include zeros.

Last days I have been using clip-on aerobars that probably interferes with the reception, with aerobars Im having trouble with calibrating stages when the Edge is in the mount.

The problem with different average power value in connect/strava started before I started using the aerobars.

And as you say, that shouldnt affect how the average is calculated.

Did new testride after exluding and including zeros. Still the same problem.
Connect and Strava still shows the same avg. power for uphill laps where Im pedaling all the time.
Its when I stop pedaling they somehow calculates avg. power different.

Thanks. Let me log some data with my 520 and see what happens. It could be that something is corrupt in memory and the unit will need to be reset.

So I did some testing on this and GC/Edge and Strava can show different results depending on if you are using auto-pause or not. This I think is different to what you are seeing.

Strava computes your average power using all power values that are captured when you are moving. The Edge computes your average power using all power values that are captured. If you don't have auto-pause enabled then the Edge will also include any zero values that the power meter is sending when you are stopped if the timer is running and the power meter is not asleep.

The average power value I got did correctly include zero values.


I think the next thing for you to try is to backup your settings, reset the unit and then restore your settings.

If you would like instructions on how to do that please let me know.

So I tried to reset to factory settings, dont know if i did it right because I dont think the problem is solved.

Did 3 testrides today:
- 1 short ride with auto pause turned off: GC_avgP=111W, Strava_avgP=107W
- 1 short ride with auto pause turned on: GC_avgP=111W, Strava_avgP=106W
- 1 longer ride with auto pause turned on: GC_avgP=129W, Strava_avgP=120W

Here is the rides:

https://www.strava.com/activities/708248727
https://www.strava.com/activities/708248768
https://www.strava.com/activities/708359636

Couldnt see that turning autopause on/off had any effect, had only resting time in testride#3 though.
The longest laps in testride #1/#2 matched well with strava, so was optimistic before taking testride #3, but results wasnt good. Had one lap match perfect in testride#3, that was uphill pedaling all the time, so I still belive that its the zerovalues that mess up the calculation of average power.

Maybe I need instructions to reset and restore settings.

Ok I see what the problem is. It is a side affect of the low radio reception rate from the Stages power meter.

I looked at the longer ride.

The power meter transmits, torque, cadence and accumulated torque. The Edge using the cadence data converts torque to power. If there was no missing data then if you summed the individual power records that value should match the accumalted value reported by the power meter. The FIT file records the power values that have been assigned to each record along with the accumulated power value.

Strava does not make use of the accumalted power value information, but looks at the individual power records to compute average power. The 520 is using the accumalted power values to compute the average. If the radio reception (RX) rate is good then the sum of the individual power reords would equal the accumalted power value. In this case the RX is low. The Edge uses a backfilling method to try and fill in some of the missing data using the accumulated data, but if the gap is beyond a certain duration it will not do that.

For this ride if you sum the individual power records you get total work of 254,432 kJ, the reported accumulated power value is 274,088 kJ. The sum of the power records is 7.17% below that of the accumulated value from the power meter.

The activity time is 2117 seconds.

Average power over this duration is

Using sum of the power values = 254,432/2117 = 120 watts
Using the accumulated value = 274,088/2117 = 129 watts

Ok, dont know if I understood all, but if low radio reception rate from the Stages power meter is the solution here, how come that uphill laps show no different avg power? Is this a problem that appears for lower values of power?

And how can i see the reception rate for different rides?

Did a new testride, maybe too short to get good average powerdata. GC/strava gives 99W/96W.

Did also change my garmin mount, can you see if radio reception is better now?

https://www.strava.com/activities/708719701/overview