its certainly no gold standard but it is not practical to do much better =)

I didn't change my opinion/argument at all. I simply demonstrated that academia is in agreement with my argument. I give you a B- for reading comprehension skills. Oh wait, English isn't your first language? How long are you going to lean on that crutch Paulo? You seem to write fluently enough, so it's time to start taking ownership of your reading comprehension/verbal reasoning. You misunderstood what I wrote; I didn't misrepresent or misunderstand science--don't confuse the two.

I didn't "get" anything--I simply showed that I know my stuff. I'll gladly go toe to toe or frontal lobe to frontal lobe with any UNC or Duke student.

So Zack, would you say that my average heart rate for an hour on my trainer is a better measure of my Calorie consumption than my average wattage for that hour? I have often seen my average heart rate be more than 10 beats higher when the room is warmer than usual or the fan isn't turned up.

Am I burning less calories running when my fitness is up and I run the same pace at a lower heart rate?

I'd still like to believe that on an individual basis, with well-calibrated models mapping HR and rear-wheel power to calorie expenditure, that HR would be a better predictor than rear-wheel power, especially because of the reason you highlighted. Rear wheel power doesn't change with room temperature, while heart rate and heat exchange (therefore by extension net calorie expenditure) do.

Right you are.
PT=<$1000
Indirect Calorimetry equipment for measuring gas exchange= ~$10,000
Direct Calorimetry equipment ~$1,000,000+

If the PT is accurate to within 10% of either the direct or the indirect cal equipment then it's totally fine (especially considering this feature is pretty trivial compared to the primary features).

I have been, and continue to be of the opinion that a power based prediction of caloric expenditure will be more accurate than a HR based prediction in the majority of cases.

The powertap isn't any better at "measuring" than your watch is I think we've established this is essentially true.. In fact, neither "measures" your calorie expenditure in the strictest sense of the word--both take one or more input parameters (which is not calorie expenditure again, the only thing that can measure calorie expenditure directly is a calorimeter), do some math (which may or may not be based specifically on you and your body type and energy expenditure rates here, I address the importance of the calibration of the model), then give you a number. I'm not sure what data you had to plug into the powertap, but if the answer is "none" then I'd take the data from the suunto over the PT any day Here I admit uncertainty as to the input parameters or quality of the powertap model, as used by FTDA. Powertap very well may have a great system, but if all of the inputs aren't being taken into account then it's probably not that great.. In fact, if the PT isn't using HR to calculate expenditure, then I'd say the data is about as useful as a poopoo flavored lollipop An admittedly bold statement, but still not totally incorrect. The large range of documented muscle efficiencies makes this method inherently uncertain, and I have yet to find an article detailing how this model can be improved (i.e how to determine which muscle efficiency the PT user has at a given level of power output (it isn't the same across all levels either!!! some people are inherently more efficient than others in certain regions of power output!). A small fraction of the rider's total energy expenditure and time rate of energy expenditure (power) is accounted for by the rear wheel power, with the remaining majority spent simply venting heat through your skin to the air, heating up your internals, etc. True, and undeniable. Thermo doesn't lie. Obviously, using some sort of "typical efficiency" figure would attempt to sort out the fractions, and extrapolate how many calories you burn overall, but I would have serious doubts about the accuracy of this method because of the extent to which the correlation between rear-wheel power output and net power output is related to your own body I think I did a decent job of showing that my doubts about the method were justified. What I'm trying to get at is the HR-based calculation is a lot more direct This may not have been correct, but I have yet to find evidence to the contrary. HR-based calculations aren't bad, provided the model is tailored to the individual, and I would sooner use that as my "accepted value" than the PT number Especially after reading what I've read I would definitely still use HR based calculations over one that only took into account rear wheel power. I just don't understand how a model based on rear wheel power could take into the account the atmospheric conditions in a stationary environment. It doesn't seem obvious to me that the atmospheric conditions don't matter, but I suppose this is a possibility. It just seems that the rate of heat exchange between your body and the environment, as well as the rate of heating of your own body would be dependent on the atmospheric conditions. .