desert dude wrote:
If only someone had done research on this very topic: OMG they did!!!!
Accuracy of Cycling Power Meters against a Mathematical Model of Treadmill Cycling
Authors Thomas Maier, Lucas Schmid, Beat Müller, Thomas Steiner, Jon Peter Wehrlin
http://dx.doi.org/10.1055/s-0043-102945 Published online: 8.5.2017 Int J Sports Med 2017; 38: 456–461
As a test of
quasi steady state power accuracy it was an interesting exercise in attempting to provide an absolute standard as opposed to comparing one or more meters with one another. It's actually not an easy task to provide an absolute standard that at least bears some resemblance to they way people actually ride.
While the accuracy of quasi steady state power scenario is of interest to most in the world of triathlon, it's by no means an indicator of measurement accuracy in other riding scenarios, e.g. during accelerations and for peak power measurement. And this is still a limitation to consider as I'll explain further down.
As has been said already, it helps to understand the manner in which you plan to use the data in order to choose a meter appropriate to your needs (or whether indeed a power meter is needed at all).
Sometimes you don't know what you might like to do with the data, or whether such things as accurate peak power measurement matters. Even for a triathlete I think it does matter, especially if you are considering ways of applying the more recent power duration modelling since the peak power measurement does affect all the model's parameters.
And a final comment - it's not just the meter but also the devices used to capture and upload the data, the firmware/software doing the post processing of data and how they all interact with one another. Sometimes errors are not due to the meter itself.
http://www.cyclecoach.com http://www.aerocoach.com.au