KJ Calculations on Various Platforms

I was looking at my ride from Saturday and I noticed a rather significant difference in the Wahoo, Strava and TrainerRoad KJ calculations. I would think KJs are a rather absolute measurement, unlike something like IF or TSS, so disappointed to see a 24% difference between headunit and TR.

Any explanations or what should be considered the source of truth? Worth submitting to TR support?

KJs / Calories / Elevaton
Wahoo- 2313 / 2325 / 5032
Strava- 2,870 / 2,315 / 5342
TR- 2,869 / 2869 / 5308

Similar issues on Friday ride

KJs / calories / Elevation
Wahoo- 4119 / 4119 / 8697
Strava- 5121 / 4119 / 8,698
TR- 5119 / 5119 / 9238 (I really like their elevation value)

The power averages are off. 194 vs 184.

Only single digits/ .3% between Strava/ Bolt/ TR for me from my spin on Saturday.

kj = kcalories is what TR use. Strava and TR pretty much matching, so I’d suggest it’s the Wahoo that’s off? What was the head unit showing at the end of the ride?

fwiw, I seem to be having lots of updates to my bolt, and when I went into the app just now, I noticed it had my as recording on a “roam” rather than “bolt”.

Mine vary too, but not enough to be worried about.

image
image image

1 Like

Anyway you can look at the raw data? Golden Cheetah or WKO? Strava and TR might be reporting based on correcting the gaps in recording. It’s an odd behavior because when I upload my RGT Cycling ride file (raw), they don’t correct for the gaps. Looks like Wahoo is uploading extrapolated/massaged dataset but not display it on their side. Never happens with my Garmin or IpBike (app).

All apps/computers calculate slightly different for a variety of reasons. If you are following numbers pick one and ignore the rest. Lol, I pick garmin, for me its always 0.1-.0.2mph faster and 1w higher and its the one thing I’ve always got with me indoors or out :wink:

It’s over 19% difference in first example and 20% the second.

1 Like

It always is, different app/computers collect slightly different base figures based on autopause and tons of other variables, they then process it with other factors which magnifies their differences in guesstimates of Calories etc. Don’t get hung up about it, its just the way it is, just pick one and follow that and ignore the others :+1:

Wahoo looks off to me. It looks like you stopped for an hour, or so, so looking at moving time, looks like about 4 hours 10 min. so time in seconds is 4 * 3600 + 10* 60 = 15000 seconds, times 194 average watts is 2,910,000 J, or 2910kJ, which is much closer to TR and Strava.

2 Likes

KJ with a power meter should be very consistent, its just average watts x time in seconds.

1 Like

Should be, but from time to time things will be different but your original power averages are different so it won’t be. Maybe its the method of recording (recording 0s) or autopause settings or simply gps interference. Don’t get to hung up about just use one thing and follow it. Night :+1:

Thanks for looking at it and the calcs.

That’s correct on stoppage time, I got colder than I think I’ve been on some early ride descents and it took me about 45min and 2+ cups of coffee to stop shaking!

So did you have two recorders, Wahoo headunit and something recording TR? TR instead of Strava because Strava can’t pair with any sensors (on my phone and as far as I know). Is it the same ride file uploaded to the three app? If so only thing that explains it is Wahoo selectively reading the recording or it’s reading the raw data without fixing gaps in recording. Gaps in recording is not the same as stopped or paused in recording. (TSS, average speed, power, etc will matter if it’s one continuous file. Wouldn’t matter for kj anyways.) Durationwise they’re typically a second or so and there’s no sensor data at all. There’s no user defined duration with Garmin (and WKO) as far as I know. Default is 1 second for Golden Cheetah. There’s no reason to extend that and for you case 45 minutes. Again wouldn’t matter for kj anyways.

No, just bolt headunit and it automatically syncs with Strava, TP, etc. I didn’t fool with anything. It’s interesting that the ‘source’ of the data may be the least accurate.

What power meter are you using?

I noticed this exact problem when riding with my Favero Assioma and Wahoo Bolt. It’s not a problem with the pm, but with Wahoo. For some reason their headline average power and kJ calcs are just wrong for this pm. When I raised a ticket with them I got a shoddy response to the effect that they knew about it and weren’t going to fix it. I don’t have this problem with my Quarq.

It’s weird, because the raw data Wahoo is collecting is correct so when I load into Golden Cheetah or TR it’s correct, but on Strava and Wahoo the maths just doesn’t work and is out by about 20%.

1 Like

I have the same issue with Garmin and my Stages XT pm. Garmin AVG Power - Equipment - TrainerRoad

check your Wahoo and try disabling auto-pause to see if it fixes the issue.

I spoke to Wahoo about it and went through their troubleshooting. They acknowledged it didn’t calc right. Just the simple kJ formula from average power and time didn’t stack up. I don’t believe it’s an auto pause issue in my case. But I also don’t know what else it is or why Wahoo weren’t planning to fix it as it seems pretty fundamental to me for an expensive headunit to display the correct average power and kJ from a popular and reliable power meter

Is it the head unit or the Wahoo App though? Given it went to Strava and TR pretty much ok, what was recorded on the head unit was correct. It’s the wahoo software thats the issue post ride. Well at least that’s what it looks like to me.

I usually display KJ (which matches calories since an update over a year ago, when paired with a power meter). You could compare what you have the end of the ride, and then with what’s uploaded to various platforms.

What does the history for the rides on the actual device show? I can see the kjoules under “power” when I go to “history” on my bolt.

I just checked head unit and it also displays 2313, so not just their app.

1 Like