Power Analysis

[NJK]

I completed a criterium race which the CPU (powertap) stated i averaged 255 watts for the 1hr. After downloading the file it stated my NP power for the 1hr was 234 watts. I presume the 255 watts average is pedalling only as obviously there are plenty of 0-20 watts in the race. I think it was 30%.

Would increasing the tyre circumference number have an effect on the CPU reading but not the download figure. I say this because my average for the same race, course last year was 205 average with 234 NP. 50 watt average power difference, same NP.

 

[Alex Simmons]

I completed a criterium race which the CPU (powertap) stated i averaged 255 watts for the 1hr. After downloading the file it stated my NP power for the 1hr was 234 watts. I presume the 255 watts average is pedalling only as obviously there are plenty of 0-20 watts in the race. I think it was 30%.

Would increasing the tyre circumference number have an effect on the CPU reading but not the download figure. I say this because my average for the same race, course last year was 205 average with 234 NP. 50 watt average power difference, same NP.

Circumference input will only affect your speed and distance reading.

Getting different APs for same NPs for same circuit/duration in a race is not uncommon. It basically depends on the race. See example here:
http://alex-cycle.blogspot.com/2006/10/2-races-in-one-day.html

Non zero average power doesn't really tell you much. It's more for curiosity value than anything else.

 

[daveryanwyoming]

I completed a criterium race which the CPU (powertap) stated i averaged 255 watts for the 1hr. After downloading the file it stated my NP power for the 1hr was 234 watts. I presume the 255 watts average is pedalling only as obviously there are plenty of 0-20 watts in the race. I think it was 30%.

Would increasing the tyre circumference number have an effect on the CPU reading but not the download figure. I say this because my average for the same race, course last year was 205 average with 234 NP. 50 watt average power difference, same NP.

Alex's response is right on target, but to be more explicit. It sounds like you have the PT CPU set up to ignore periods of zero power while calculating its display average power. It won't change anything while riding steady efforts like time trials but will give you much higher average power readings on the CPU display for things like crits where you coast through frequent corners. Go into the PT setup menu and make sure you include zeros in your display average if you want it to match your uploaded files.

-Dave

 

[NJK]

Yeah i think it is set-up as pedal only. So the big difference between the two comparison races 205av 234NP / 255av 234NP is purely down to when i pedalled i pedalled harder in race 2. It certainly felt like it

 

[daveryanwyoming]

Yeah i think it is set-up as pedal only. So the big difference between the two comparison races 205av 234NP / 255av 234NP is purely down to when i pedalled i pedalled harder in race 2. It certainly felt like it

I think you've basically got it, but your terminology is a bit mixed up.

It's impossible to have an actual AP of 255 and NP of 234 watts. Your situation is CPU displayed AP vs. CyclingPeaks AP or perhaps CPU displayed AP vs. CyclingPeaks NP, I'm not sure which you're referring to.

Anyway, set the CPU to include zeros in averaging and your CPU displayed AP and AP in CyclingPeaks after upload will agree. That's step one if you ride events like crits with substantial coasting.

The second question of NP vs. AP (true average power, not displayed average ignoring zeros) is as you've observed. It's possible to go easier overall (relatively low AP) but go harder during the peak power efforts and get a higher NP but with the exception of very short efforts in the 5 minute range your NP will always be equal to or higher than your AP, not lower.

Anyway you didn't have an AP of 255 and an NP of 234. It sounds like you had a CPU displayed AP of 255 (when you ignore zeros), an actual AP in the 230 range and an NP of 234.

-Dave

 

[NJK]

I think you've basically got it, but your terminology is a bit mixed up.

It's impossible to have an actual AP of 255 and NP of 234 watts. Your situation is CPU displayed AP vs. CyclingPeaks AP or perhaps CPU displayed AP vs. CyclingPeaks NP, I'm not sure which you're referring to.

Anyway, set the CPU to include zeros in averaging and your CPU displayed AP and AP in CyclingPeaks after upload will agree. That's step one if you ride events like crits with substantial coasting.

The second question of NP vs. AP (true average power, not displayed average ignoring zeros) is as you've observed. It's possible to go easier overall (relatively low AP) but go harder during the peak power efforts and get a higher NP but with the exception of very short efforts in the 5 minute range your NP will always be equal to or higher than your AP, not lower.

Anyway you didn't have an AP of 255 and an NP of 234. It sounds like you had a CPU displayed AP of 255 (when you ignore zeros), an actual AP in the 230 range and an NP of 234.

-Dave

Cheers Dave. It just looked strange as you say its impossible to have those figures. I haven't had any odd looking data in the 18 months i've had the powertap until this one. I haven't altered any settings and every road race or criterium the AP CPU figure has always been 20-30 watts below NP figure, until this.