Here is a calculator that takes into consideration whether you are walking and running, and the slope. It is based on the MET approach.

ExRx.net : Walk / Run Metabolic Calculator

Sorry for the gollowing long post.

From the research that I have done, the pace on the treadmill is frequently and substantially off. They have not been calibrated, or kept calibrated. So of course the calculated calories burned can be way off too for this and other reasons. This also assumes their calories calculation is valid, which IMO is problematic itself.

How does the treadmill calculate calories burned? There are several models to go by each providing a different result. One method basically uses weight, pace, and time. IMO including BMI may lead to more accurate results. Either way is simple to calculate but IMO calculates values that are useless for what are obvious reasons, some outlined next. Another method uses a proprietary value in the calculation for a given exercise chosen. This is equivalent to their own MET value. I imagine this helps allot. I think a great many devices use this approach. It is reltively easier to investigate and implement. Another way includes the heart rate in its calculation. I believe this requires allot of research and difficult to pin down for a particular individual as to how heart rate is impacting the calories burned calculation. I think there is a “better” scientific basis to this, but there are just too many variables to make this approach reliable.

Then there is now a new state-of-the-art approach. The calculation basically uses your physical fitness level, which I imagine includes an accurate evaluation of an individuals metoboic rate. There appears to be a much better scientific basis for this “fitness level” approach that can be calculated. But this is IMO unrealistic. One would need to go to a lab to determine these values accurately.

So to summarize, lets assume that the treadmill determines pace and time correctly. Once again, I am finding that in many cases this can be far from being accurate. Next the determination of calories burned is based on this information, assuming the formulas used provide worthwhile results.

Anyway, for the above reasons, I no longer go by what a device calculates as calories burned. I think it is futile. It comes down to numerically evaluating “effort” placed in an exercised. I see that there is no numerically accurate way to determine the actual **effort** that you made during your particular days exercise. I think the only method that is worth using, and as only a guide, is the approach that uses a type of MET value. This has been a well examined and formulated approach. Still this approach has its problems. However maybe it can be calibrated for an individuals particular situation?

I think your intuitive analisys of the actual effort you place in a particular exercise may have more significance. You are in the best position to determine this. Perhaps use a value calculated by a MET based approach as a starting point? You have been essentially doing this but in the past with other sources of information. I think the trick is knowing your body well enough and choosing what guide to use. For that matter, ultimately, changes in weight over a period of time is what matters. Perhaps using changes in weight as feedback into your personal “equation” will allow you to more accurately gauge calories burned for future exercises?

FWIW IMVVHO (In My Very Very Humble Opinion)