Power reading anomalies

turbotommy

Hi guys

Just finished a ride and noticed that strava is showing a different average power to my wahoo element bolt. The bolt is taking readings from my power meter (p2m ng). And strava is linked to the bolt. In strava it’s saying the reading is from power meter rather than their own guesstimate system, so I’m a little confused.

Bolt is showing 255w and strava 220w

38special

Have you checked your Bolt settings and made sure 'Include Zeros in Average Power' is turned 'ON'? This would account for the different Average Power. Normalized Power on the Bolt and Weighted Average Power on Strava will almost always be different because they use different formulas.

turbotommy

38special wrote:

Have you checked your Bolt settings and made sure 'Include Zeros in Average Power' is turned 'ON'? This would account for the different Average Power. Normalized Power on the Bolt and Weighted Average Power on Strava will almost always be different because they use different formulas.

Hi 38special

I’ll check that settting, what does it mean in simpleton terms?

It terms of formula shouldn’t The bolt just be reporting power output from my power meter, and then passing on the same number to strava?

maryka

Are you talking about the display on the Bolt or the data that you upload and look at afterwards? And if you're just comparing what's displayed on the Bolt after the ride to what Strava says, how does that compare with what you see later when you upload the file for analysis in the software? (I have no idea what software Wahoo uses, but you could use Golden Cheetah for free, or maybe you have WKO or Training Peaks?)

Removing zeroes means that every time you freewheel, those aren't included in the calculations which displays average power. For Garmins, the zeroes are still included in the recorded data though, so when you look at the file later, what you see is not what you saw when you were out riding. Not sure what the Wahoo does. Personally I think it's a bit pointless to remove the zeroes as every time you freewheel or stop pedalling, that's recovery time for your legs that should be included in any power calculations. So e.g., if you rode 20 min with a handful of short downhills and junctions where you had to stop pedalling, then you rode 20 min on the turbo non-stop, those two sets of numbers would be useless to compare with each other if you'd removed all the zeroes from the outdoor ride.

turbotommy

maryka wrote:

Are you talking about the display on the Bolt or the data that you upload and look at afterwards? And if you're just comparing what's displayed on the Bolt after the ride to what Strava says, how does that compare with what you see later when you upload the file for analysis in the software? (I have no idea what software Wahoo uses, but you could use Golden Cheetah for free, or maybe you have WKO or Training Peaks?)

Removing zeroes means that every time you freewheel, those aren't included in the calculations which displays average power. For Garmins, the zeroes are still included in the recorded data though, so when you look at the file later, what you see is not what you saw when you were out riding. Not sure what the Wahoo does. Personally I think it's a bit pointless to remove the zeroes as every time you freewheel or stop pedalling, that's recovery time for your legs that should be included in any power calculations. So e.g., if you rode 20 min with a handful of short downhills and junctions where you had to stop pedalling, then you rode 20 min on the turbo non-stop, those two sets of numbers would be useless to compare with each other if you'd removed all the zeroes from the outdoor ride.

Thanks for the response. Yeah just comparing data on the bolt screen vs strava. I don’t normally look at average power for an entire ride but it was the last thing I saw on the screen as I turned the bolt off. I use golden cheetah so I’ll have a look when I’m back home and uploaded the ride. I agree about including the zeros. Although I’m normally looking at my data with regards intervals or efforts rather than overall averages.

I suppose I just assumed that strava was taking the information from the bolt at face value as opposed to analysing data from the pm differently.

Thanks again for both of your responses.

maryka

Nope, Strava uses the entire file as GC does. It's only the display you're altering when you choose to remove zeroes, have a different rolling average power displayed, etc. Which is a good thing, you don't want the choices for display to actually be reflected in the raw data for analysis!

MishMash95

If you are looking at Strava on the phone app, the displayed average power is the mean average of the ride, whereas if you use the desktop version of Strava, it displays "Weighted Average Power" which is the normalized power that better represents the physiological cost of your ride. Alternatively to the potential include-zero issue, It could be the case that the Wahoo was simply showing the normalized power for the ride, while the strava app was showing mean avg.
(Also note that weighted average power, or normalized power, is not the same as removing zeroes).

maryka

Strava desktop also displays average power in the "show more" drop-down, which shows an AP that includes freewheeling zeroes but doesn't include stopped time zeroes. So if you spent 2-3 min waiting for someone and left your Wahoo/Garmin running, it wouldn't include that. This AP will be different to what shows up in GC or WKO where the stopped time zeroes contribute to the overall average power for the ride unless you cut them out.