How is Calorie burnout calculated? Is it accurate?


The amount of energy burnt by a person to do an activity should be different for another person doing the same activity. How does the smart watches and similar devices calculate calories burnt?

In: 1

They use an algorithm that is generally based on an input of effort (steps on a pedometer, distance and elevation on a treadmill, heart rate and distance for a typical running watch, etc.) and weight (generally input by the user). Any calculation will be off based on inaccuracy of the data and estimation factors inherent to the algorithm. For example, we have a pretty good estimate of how many calories a 200 lb. person burns per mile run, but a variety of factors (running form, running economy, shoes, elevation, etc.) will cause small differences from person to person.

Depending on the activity, you can estimate it pretty well based on e.g. a person’s weight. For instance, the amount of calories you burn running a mile depends mainly on your weight. It also depends a bit on your speed, but not much. Of course, other factors do play a role, like how efficient your movements are (e.g. if you have a lot of vertical motion, bobbing up and down, that “wastes” more energy), as well as different physiological and anatomical factors. But those other factors are more marginal. Based on weight alone, you can get a fairly good estimate. That is, as long as you’re running on flat terrain with not too much wind. If there’s lots of climbs on your route, that can make a substantial difference that you need to account for, which is possible (in principle) using GPS. Wind of course is much harder to factor in.

This is just one example. Not every activity is so easy to estimate. Cycling is easy. Anything you do on a machine (like a treadmill, elliptical, rowing machine, etc.) is especially easy as the machine knows more of the variables (like the exact resistance on your stationary bike). But something like boxing, or playing basketball, that’s much harder as there’s a wider array of different movements involved. You can still do studies to get average estimates, but they won’t be as accurate if you apply them to different individuals.

They guess, and most of them are pretty bad, because people like big numbers and therefore like devices that do that. Some devices also measure both basal metabolism – how much you would burn if not exercising – and exercise metabolism.

The only fairly accurate ones are calibrated exercise bikes and bicycles with power meters; those are probably within 10%.