Should I retroactively scale power?

I got my power meter back in January, so I haven’t had the chance to do many outdoor rides with it yet. Today I did a ~40m TT effort after some warm up. Before the ride my calibration value read -88 and afterwards it read -99 (I believe the unit is 1/8 N*m). Since power = torque * speed, and I may have had a torque offset for some or all of the ride, should I re-scale my power data by the ratio of the pre- and post- calibration factors? As in, just multiply power by 88/99? (or take an average, and multiply by 88/((99-88/2)). Am I thinking about that right or will it make less of a difference than that? Do most people just leave their data alone when this happens?

(I know I can do this via “Fix Data” on this website, just wondering whether I “should”). Thanks!

If you are new to this and still on the up power-wise I would leave it alone. Otherwise you might want to fix those rides so you can still be motivated by “best power” achievements and so on.

You could also start a new season in Intervals.icu to have a fresh start.

88/99 would not be the right way to scale it. The calibration value for a power meter is the offset, not the slope. A 10% change in offset does NOT mean 10% change in power. If the unit is in fact 1/8 Nm, then your observed shift of 11 would translate to a 11*1/8 Nm = 1.375 Nm TORQUE offset. The impact on power would need to be recalculated with a new torque value as the input. For 40 min effort a reasonable torque value would be roughly 30 Nm, so you’re looking at 1.375/30 = <5% impact on your numbers.

1 Like

D’oh. Yeah I actually realized that while thinking in the shower this morning lol. I was more like ~20Nm, but yeah, should work out to 5% or less, so I’m going to leave it alone. Thanks.