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

How are others setting their power?

3999384
over 3 years ago

So it took me a while to figure out why my power numbers in Garmin Connect and Strava were completely different for the same ride.  What I think is happening is that the 530's default for power is to exclude 0's whereas Strava takes into account the 0's (freewheeling).  Can anyone confirm this?

Additionally, I would be curious how other avid cyclists have this setting.  Are they including or excluding zeros?  What about the Pros?  I would suspect they include the zeros, but I don't know for sure. 

  • 0 over 3 years ago

    If you upload the rides to RidewithGPS (Free account) it will give you both numbers (with & without 0s) in the ride metrics but as with the way of most bike things the metrics will probably disagree with both the Garmin and Strava, just so you have more to ponder in your search for the "truth"

  • 0 over 3 years ago in reply to L.Rouge

    Boy, you certainly said it there.  I uploaded the ride to RIDEWITHGPS and now they are all different.

    • Garmin Connect:  Average Power = 221 watts
    • Ride with GPS:  Average Power = 225 watts
    • Ride with GPS:  WR (work reflected) Power = 208 watts
    • Strava: Average Power = 181 watts
    • Strava:  Weighted Average Power = 195 watts

    What a mess.  Strava's Weighted Average Power is the only one I throw out because they use their own formulas to come up with that number.  The rest should reflect some kind of consistency, but they don't.

    This is frustrating.  It is confusing enough that Strava's Average Power isn't even close to what Garmin Connect reflects.  I guess I'll have to exclude zeros and see if that will possibly match up Garmin Connect with Strava.  If they don't then I don't have a clue what Strava is using to calculate power.

    Additionally, this now calls into question so many other riders and trying to compare your power to theirs.  Yes, there could be a little variation between power meters, but personally, I don't think Garmin should even have the option to include zeros.  It makes matters worse that the default is to exclude zeros.  You have no idea who has this set to exclude zeros and who is including zeros so there is no point in trying to compare.    

  • 0 over 3 years ago in reply to 3999384

    I was viewing a GPLAMA Youtube video in which he mentioned the differences between what people's Garmins report and upload / simulation sites like Zwift etc reports, His view was that this is just as it will be as different sites / devices will use different formula and measuring algorithms. Assuming the "average" power is an average for the same sample periods (instantaneous, 3 second, 10 second,...) then they should be close eg 221 & 225 are a good match but the "work related", "Weighted average" and Garmins "normalised power" all introduce other varying interpretations and other formula variables.

    Your best bet is to do a few rides and settle on one figure that you have a bit more faith in and stick with that and use it to compare you performance over time. Important is to compare yourself to your previous performance and don't worry about other people as they will have different devices and be measuring different things. Unless you are a solid pro then the including zeros will result in a (much?) lower average number, good luck

  • 0 over 3 years ago in reply to 3999384

    I would just use the power numbers from Garmin.  That is directly based on what was recorded during the rides.  Then you don't have to worry about how other services interpret the data.

    It is best to include zeros in the power data recording and turn off auto pause so all the metrics based off power are more accurate.

  • 0 over 3 years ago in reply to Wongataa

    It's probably the auto pause causing the variances (or much of it) as this can give different readings for average speeds, especially when there is a lot of stopping and starting. 

    MB

  • 0 over 3 years ago in reply to meltonboy

    Thanks meltonboy, but that can't be it.  This was on an enclosed 5-mile loop (total of 30 miles) with no stopping. 

  • 0 over 3 years ago in reply to 3999384

    Can you post the FIT activity file?

  • 0 over 3 years ago

    I just had a look at mine, GC/Strava/TrainingPeaks- Avg power is identical or close enough (within 3-5w) for all platforms. Settings are:

    Recording interval: 1 sec
    Cadence Avg: Do NOT include 0s
    Power Avg: Include 0s

    Also Auto Pause: When stopped

    Not sure why Power and Cadence are different or if that was set by default, but I don't remember specifically setting them that way and its working as expected. Avg cadence data also matches similarily across platforms.

  • 0 over 3 years ago in reply to Notsofast...

    I think the differences you are seeing are based on whether zero values are being used in the averages.

    Using the data in you file I get

    Average power for the duration of the activity of 181 including zero power values. If I exclude zero power values I get 224. 

    This I think this matches the spread in values you are seeing.

    I would check the data recoding options on your Edge to see what this setting is configured to 

    Settings -> System -> Data Recording -> Power Averaging (set it to include zeros)