Originally Posted by daveryanwyoming .
The PowerTap or any power meter measures watts and over the course of a ride measures average watts which is much more closely tied to Calories burned than any HR or speed based estimate. It breaks down like this:
kilojoules of energy = average power (watts) * seconds/1000
So in terms of hours:
kj = AP (watts)*3.6*hours (there's 3.6 kilo-seconds per hour)
Calories combusted by your metabolism = 4.184*kj
So far that's straight physics and unit conversions and does not vary between individuals. But individual riders each have their own Gross Metabolic Efficiency which isn't about how pretty a circle you spin while riding, it's a measure of how much of that combustion energy makes it down to the pedals in terms of useful power. Most of the remaining energy gets burned off as waste heat. The variation across riders is very small and in the range of 19-26% or and typically shown in the published research to be normally distributed with a mean near 22 or 23%. Interestingly GME statistics are the same for both trained and untrained populations so even though individuals vary there's little evidence that GME necessarily increases with increased training as even grand tour pro cyclists have GME distributions statistically similar to untrained populations.
So in terms of dietary calories burned during cycling it comes down to the average power you sustain in watts, the time you sustained it and a narrow range of variation based on your own GME that can be measured during a lab VO2 Max test if gas exchange measurements are taken which means the full breathing mask rig. Since most folks don't do gas exchange testing and haven't nailed down their GME it becomes simpler to assume their GME is ~23.9% or the reciprocal of 4.184 (the kj to Calorie conversion factor). When you do that you're basically estimating dietary Calories as straight kj burned in exercise which is what power meters measure. So you end up with:
Calories ~ AP (watts) * 3.6 * hours with a normally distributed error with a variance of ~ +/- 8%
So if you maintain an average power of 200 watts for an hour you'll burn approximately 720 Calories. Do that for a five hour century and you'll burn approximately 3,600 Calories. Sustain 300 watts for an hour and you'll burn approximately 1,080 Calories and so on.
Short of riding with a gas exchange mask that's the most accurate way of estimating dietary calories burned while cycling (or rowing on a power equipped rowing erg). Heart rate based estimates must make a similar GME assumption but they also are forced to estimate the power you sustain at a given HR. They try to do this by having you enter weight, perhaps tracking speed, perhaps taking into consideration an estimate of your HR at LT or HR max but they're forced to layer estimates on estimates which only increases error.
For instance take two riders who both have an LT HR of say 160 bpm and even assume their weight and aerodynamics and tire rolling resistances are the same but one is a new rider who barely sustains 150 watts at LT and the other a Cat 3 cyclist that sustains 300 watts at LT. They both go out and ride at LT for an hour and both come back with an average HR of 160 bpm for the ride. A HR based estimate of Calories burned is going to be way off in one or both cases as the Cat 3 rider will have burned roughly twice as many Calories per hour as the untrained rider.
Bottom Line, dietary Calories are a measurement of energy, power is defined as energy per unit time or IOW, energy is defined as power integrated over time. A PM directly measures both power and energy which is what we're after.
-Dave