Accuracy of calorie estimates on fitness apps

[Anotherespressoplease]

Something occurred to me recently about the calorie expenditure estimates on fitness apps. I think many people would assume it refers to additional calories burned rather than total calories burned over that time.

I did a long route (~3 hours) and fitness app said it burnt a huge number of calories, far more than I would actually need to eat to replace it (otherwise looking at my weekly exercise vs food intake, I would have wasted away by now and clearly haven't).

It struck me that when calorie expenditure is measured, presumably it doesn't account for the basal metabolism over that time period. For example, very approximately, you would use around 100kcal/hour just for basal function if you used 2400kcal daily intake as your baseline (for easy maths). That's a very rough estimate given some hours you are sleeping.

Anyway, over a 3 hour activity, that means my fitness app presumably includes in the calorie estimate approx 300 calories that I would have used anyway in that time, had I just been watching TV or mooching about in the house. This obviously adds up significantly over time if people are using their calorie estimates on exercise apps to justify another piece of cake etc, and I wonder if this is a factor in why some people find they are exercising lots but not losing weight.

That's without factoring in that much data is probably gathered from default male population, and so may already overestimate calorie expenditure for women given their average weight will be lower.

I appreciate different apps may use different algorithms or have different amounts of baseline data (e.g. would a wearable be more accurate?). However I still think it is likely to include the basal rate and lead to the user over-estimating their additional calorie expenditure from the exercise.

What are others' experiences of calorie estimates and do they vary significantly if you use different apps?

Does anyone know if apps make attempts to count the additional calories expended through the exercise, or if they just consider all calories likely to be expended during that time?

Different apps have different levels of accuracy depending on what data they have. If they are used in combination with a watch and HRM and know sex, age, weight, heart rate etc and know how fast you are running and on what gradient they are probably reasonably accurate at population level (i.e. assuming you are a typical (eg) 9 stone 45 year old woman running 5.30min/km with 1000ft of V over 10k. But some people just run much more efficiently than others because of their biomechanics/gait so it's not going to be exact.

But yes, I think consensus is that generally they overestimate. Critically, evidence has suggested that (generally) when you exercise you tend to compensate with more rest for the remainder of the day so that also has an effect on total energy expenditure.

Way out. The only time my weight loss stalled was when I added exercise to MFP and it gave me the calories back to eat. In theory, as I was mostly still eating under my daily calorie targets according to the app, I should have still lost weight. I didn’t. I stopped adding exercise (& the calories) and my weight loss was back on track.
I’ve see loads of comments on fitness forums saying the same & that exercise trackers in general massively overestimate calories burned.

Also depends on what exercise they are tracking- I think they're generally better at things where they can track speed than things like weight training/ yoga etc.

Garmin does diferentiate between "active" calories and "resting" calories.
While it's not going to be completely accurate, with knowing height, weight, heartrate and activity type, it's as good an estimate as a casual athlete is going to get.

I wore a heart rate monitor (polar) constantly at same time as a Garmin & a Fitbit activity tracker.
Did that Several times, a few full days of data.
From the heart rate numbers & my biological basics (age, sex, mass) I can estimate likely calorie needs.
The daily calorie needs by all 3 methods were very similar, within 3-8%, so to me that's not a big difference between them.

None of the methods consistently was higher or lower than others.

I don't really think about individual activity estimates, it's the daily total that matters imho.
I would say the device estimates+App are very accurate for daily totals and must be pretty accurate for individual activity, which probably includes being alive calorie needs, too.