Data Details

As mentioned earlier, I use power data in my bike training, and I think everyone should. It allows precision and repeat-ability in workouts, and provides unerring pacing feedback in races.

I'm shooting for a critical power (or one-hour power, or threshold power...) of 200 watts for the Tahoe half IM. That will allow me to race at 160w or so and, hopefully, get off the bike inside 3 hours. I've been encouraged by my Powertap powermeter head unit -- I've had some 2+ hour rides at 160 +/- watts, and this morning's 1.5 hour workout came in at 168 watts (normalized).

But then I loaded the power file into the Apollo Raceday software (sold by Phil Skiba) and the NP came in at only 156 watts for that ride. What the heck...?

It turns out the Powertap head unit ("Joule") computes average power and normalized power by dropping out any periods where power is zero. So, in effect, the NP = 168 figure was what I was doing while pedaling, without taking into account not-pedaling. Why does this matter?

Because the whole point of NP is to take into account hard and easy riding segments, and the fact that harder segments are A LOT harder than easy segments. But the easy segments must still be accounted for since they serve to pull down the average -- and they serve to help us not be so tired. We need to include the zero segments in order for NP to work.

After a long phone call with Powertap tech support, the Powertap folks acknowledged that no, the basic Joule head unit does not have the "include zeros or not" switch that most powermeter head units have. That won't work for me...I need the accurate figures.

So, the company agreed to let me return my basic unit and pay a discounted upgrade cost for the more advanced Joule GPS unit that has the required setup switch. Plus -- it has GPS. I don't why I might need GPS on a bike computer, but I guess I'll find some use for it.

So, anyway...if you have a powermeter that includes the NP computation, set it so that it includes zeros.

But...here's a tip...if your computer does NOT have NP calculations, set the average power display to leave out the zeros. Am I contradicting myself? No...based on long experience, the average power on a rolling or hilly course without zeros ends up being pretty close to the actual NP computation. You can use that no-zeros AP on your display to approximate the actual NP of your ride, in real time. I did it for several years and had great results with that little trick.