There are a lot of new terms to learn when you start training with power.
Most are common sense. For example, average power is fairly self-explanatory. However, one term that trips a lot of people up is normalized power.
Normalized power is a score that estimates the physiological effort of a ride. The easiest way to explain this is through an example. If we look at a 25-mile time trial and a criterium, then in both events you are required to ride as hard as possible for roughly one hour. If you recorded your average heart rate for both efforts you would imagine that they would be very similar.
In the time trial you would aim to ride a consistent power for the whole hour, where, in a criterium a rider will sprint out of a corner, then not pedal through the next corner, the sprint etc. This means that in a criterium a rider’s power output is very variable. The amount of time spent not pedaling in a criterium may be as high as 30 per cent and this drags down the average power.
As a result, the average power for the time trial will be higher than the average power for the criterium but at the end of both efforts you will likely feel as though you have ridden as hard as possible for an hour.
Peaks and troughs
This is where normalized power comes in. Normalized power is a formula that smooths out the peaks and troughs to give a value that is approximately the same as if the power output was consistent for the whole effort. In essence, this means that sprints more than cancel out the time spent not pedaling and you come out with a normalized power figure very similar to your FTP power (the power you can hold for one hour).
As a result, the normalized power for a one hour criterium and a one hour time trial would be very similar, i.e. the maximum power you can sustain for one hour. As a result, this is a much better representation of how hard you have worked during the event and how you will feel afterwards.
A common misconception is that normalized power is the average power if the time when you are not putting out any power aren’t included. This isn’t actually true. I won’t go into how normalized power is calculated but essentially it gives a greater weighting to higher power outputs, as these obviously require more effort to put out.
What else do you need to know about data analysis?
