cancel
Showing results for 
Search instead for 
Did you mean: 

Incorrect power meter data

erlendostbye
Mt. Kenya

I have a 4iiii power meter, and i have discovered that there is a discrepancy between the average power i displayed in Garmin Connect and Strava. On my last ride Garmin claimed 223W, while Strava claimed 167W (Not too big numbers... hehe). That 56W in difference. I have checked that the power meter is registered, and that Strava does not estimate the power. Does anyone know how to fix this? 

1 ACCEPTED SOLUTION

Jane
Moderator Moderator
Moderator

Hello @erlendostbye 

Thanks for posting about this and apologies for any confusion.

Weighted Average Power is similar to - but NOT the same as - Garmin's Normalized Power.
Both numbers attempt to capture the average effort of a workout by taking into account lactate accumulation and other types of fatigue. Strava uses an algorithm very similar to xPower for computing Weighted Average Power.   Find out more here. 

There will be differences between the Weighted Average Power and Normalized Power. How much and which direction will depend on the ride. Typically, for shorter rides, Weighted Average Power will be lower than Normalized Power.

This is a result of the different types of smoothing used by the algorithms. Both numbers have been used successfully by professional coaches.

Another thing to keep in mind is that on Strava we do not remove any sections of zero power from your data before we do our calculations.  Whether or not your Garmin power calculations are including sections of zero power depends on your Garmin settings.  You can check the Garmin Support site to find out how to adjust this setting if necessary.

I hope that provides some insight into this.  Please let us know if you have any further questions.


Jane (she/her)
STRAVA | Community Hub Team

View solution in original post

13 REPLIES 13

Johndinis
Mt. Kenya

Hi. 
completed my ride this morning, avr power was 190w and calories 847 

two hours later I checked on the app and the avr power has now dropped to 97 and calories dropped to 420 

 

how is this possible after morning ride has been done and completed 

rafiniv
Mt. Kenya

Yes. I changed my setup in the Garmin to include zeros when calculating power!
I set the same settings as my friends.
It seems to my friends that the data from the Garmin app goes to Strava exactly as it is on the Garmin without any additional calculation on Strava.
For me, they are probably calculated in Strava, so I get a much lower result.

Jane
Moderator Moderator
Moderator

Hello @gerardgonzales and @rafiniv 

Thanks to your both for posting.  Have you tried changing the setting on your recording device to include zeros when calculating power?  If not, I would recommend giving that a try.   

 


Jane (she/her)
STRAVA | Community Hub Team

Jane
Moderator Moderator
Moderator

Hello @erlendostbye 

Thanks for posting about this and apologies for any confusion.

Weighted Average Power is similar to - but NOT the same as - Garmin's Normalized Power.
Both numbers attempt to capture the average effort of a workout by taking into account lactate accumulation and other types of fatigue. Strava uses an algorithm very similar to xPower for computing Weighted Average Power.   Find out more here. 

There will be differences between the Weighted Average Power and Normalized Power. How much and which direction will depend on the ride. Typically, for shorter rides, Weighted Average Power will be lower than Normalized Power.

This is a result of the different types of smoothing used by the algorithms. Both numbers have been used successfully by professional coaches.

Another thing to keep in mind is that on Strava we do not remove any sections of zero power from your data before we do our calculations.  Whether or not your Garmin power calculations are including sections of zero power depends on your Garmin settings.  You can check the Garmin Support site to find out how to adjust this setting if necessary.

I hope that provides some insight into this.  Please let us know if you have any further questions.


Jane (she/her)
STRAVA | Community Hub Team

This isn't good, we want to exclude zeros as we want to see our workout for real in terms of average power produced, it doesn't matter about the rest period which in turn Strava is lowering our real numbers, please add the option like Garmin to exclude zero, and not change our data.

This is not the entire story on the data differences though.  Setting the zero records on power averaging to be included is a start to resolving the problem, but there is also something else happening in strava that is not being addressed here and that is actual ride pauses and how different head units and software processes those vs what strava does.  I have tested this extensively by recording rides with structured intervals using multiple head units (all set to include zeros in power averaging) and while other software (training peaks, today's plan, garmin connect, golden cheetah, etc...) all summarize the data in almost exactly the same way with the same results from different computers, the same cannot be said of strava.  For example, just the other day did 3x15 min intervals and used both a Pioneer SGX CA600 computer as well as a Stages L200.  For the first two intervals I had almost no traffic stops but on the 3rd I hit several stoplights and was careful to manually pause/start recording on both computers at the same time at every stop (GPS auto pause turned off on both since that can also cause problems with strava).  During the ride both computers tell me the lap avg power when I complete a lap and both agreed within 1-2 watts.  After the ride the other ride analysis sites showed each interval at roughly the same power, speed, time, hr, cadence, etc..  but not Strava.  The last 15 minute interval (the one with ride pauses due to traffic stops) Strava for some reason clipped over 25 watts from the average power despite showing that interval being exactly 15 minutes, but only did this to the recording from the Stages L200.  When I deleted that activity and uploaded the file recorded on the Pioneer computer, Strava showed power averages for the intervals all being correct (agreeing with the computer as well as all the other ride analysis software sites).  So strava is definitely doing something different and it has to do with both the computer somebody is using as well as how many stops, or recording pauses, a ride has.  But why, Strava?  This makes very little sense.  The example I gave above was just one, but i have done this a lot and have also tested with a Garmin 820 and Wahoo Roam.  The only computer i have found that seems to process in strava exactly as the computer shows as well as other sites is the Pioneer CA600.  Rides recorded with all other head units seem to have some sort of data processing differences in strava that don't happen anywhere else.  So the answer to the OP's question could also be "it may be that strava and your bike computer simply will not agree".  I would love it if someone at Strava could explain why this phenomenon exists.  I have tested and seen this too many times so it's definitely not a one-off fluke.  Have other riding buddies see it too.

Appreciate there are a lot of variables involved, but I have often suspected that there is something in the way Strava handles different computers, as wahoo computers often seem to average higher watts than people with Garmins. I am getting this information off a very small population so can't say it is factual!

I dont think this is the right solution. Why is Strava counting the 0W? Being penalized for stopping at the light? Doesn't make a whole lot of sense to me..

Strava doesn't count 0 W when stopping, only when riding.