How accurate are your ‘Energy Burned’ numbers?

[MrBosco]

Hi folks,

I’m wondering if anyone else is having problems with very inaccurate calorie burn figures.  I suspect, but am not certain, that you need to be close to an average level of cardiovascular fitness for the trackers that use heartbeat to calculate calorie burn to be close to accurate.  I’d like to know for sure.

Have you ever tested the calorie burn feature for yourself?  Was it over, under, or close to accurate?  Are you very fit, of average fitness, or unfit?  What model of tracker do you use?  Do you ‘fudge’ the numbers in any way by entering a false height or weight in order to get Fitbit to show better calorie burn numbers for you?

To determine how accurate your tracker is at calculating calorie burn you can use this method:

1:            Track and log your food intake in detail for at least two weeks.

2:            Track your weight weekly during this time period.

3:            Estimate your average daily calorie surplus/deficit for this time period.

I use a spreadsheet to do this (screenshot attached) though the math is quite simple.

4:            Compare this figure to the weekly ‘Cal in vs. Out’ graphs in the Fitbit app.

Screenshot attached showing my own numbers.

Testing tips:  Do not change your diet or exercise routine significantly during the test period.  If you are currently making major changes, wait a while for your body to adjust before trying to verify your daily calorie surplus/deficit.  Weigh yourself at the same time of day and (ideally) the same day of the week.  First thing in the morning after using the bathroom is good.  This helps to minimize big variations due to water weight – a pint of water weighs around half a kilo, so your breakfast and morning coffee can make a significant difference.  Stored body water can also make a big difference, though if your water intake has been relatively steady for a few weeks and you don’t make big changes during the test period this shouldn’t be too great an issue if you weigh yourself first thing in the morning (after you pee).

I have a Charge 6, am currently fat, unfit, and am losing weight.  In my case the app is showing me a calorie burn that is inflated by over 700 calories per day on average.  That’s a whole dinner for me.  I’d like to confirm if this is a Charge 6 problem or a general Fitbit problem, and whether this is a common problem for people who bought a Fitbit to help them lose weight.

Many thanks for reading.

[JeffSinSF]

I am new to using a device for fitness tracking, having just received a Pixel 2 watch as an early Christmas gift.  After two weeks of using the watch and looking at my data in the Fitbit app, I'm convinced that its calories burned calculation makes little sense.  Here's an example for me:

Yesterday I walked uphill on a treadmill for about twice as long as I walked on a flat surface outside (43 minutes on the treadmill versus 53 minutes outside).  My heart rate was a bit higher for the treadmill workout which makes sense  since it's harder to walk uphill than on a flat surface.  Fitbit, however, indicated that I burned slightly less than 50% more calories on the treadmill than I did walking.  I supposedly burned 6.26 calories per minute on the treadmill but 8.4 calories per minute during the walk.  I am left scratching my head.  It is using the same profile (age, weight, etc).  It accurately counted the minutes of the two activities.  It presumably accurately measured my heart rate, but even if those measurements were off it said my heart rate was slightly higher on the treadmill.  Yet, somehow, I was burning more calories per minute walking on flat terrain in the neighborhood than when deliberately walking uphill.  And this inaccuracy isn't due to it not accurately measuring steps, either, though I know that's a huge issue since it appears to heavily rely on arm movement to determine steps.

In reality, I think the calories burned per minute walking is being exaggerated greatly, and the calories burned walking uphill on the treadmill are being undercounted. 

[JeffSinSF]

Oops.  I just noticed a typo in my post and don't see how I can edit it:  I walked 53 minutes on the treadmill and 23 minutes outside.

[MrBosco]

Hi @JeffSinSF - very many thanks for the reply.

I've read on this forum that the newer trackers with optical heartrate sensors use heartrate exclusively to estimate calories burned.  Your experience suggests this might not be the case.  Could the device be using steps and/or distance in addition to heartrate to estimate calorie burn when using the Walk exercise mode?  Or is the heartrate sensor so inaccurate that you get a wide variation in the calories burned number between sessions doing exactly the same exercise at exactly the same intensity for exactly the same duration?

If you're keen on finding out you could do some experiments.  Since you have access to a treadmill a good place to start might be to track a number of sessions where you walk the same distance at the same incline setting at the same speed.  Every session should have near-identical calorie burn figures if the sessions are long enough.  Starting heartrate might be a confounding factor if the sessions are short, as it takes a while for your heartrate to pick up when you begin exercising.  If the numbers don't match between sessions then the problem must be the heartrate sensor I think.  If the numbers do match up then there's more experimenting to do to compare the Walk exercise mode with the Treadmill mode.  Fun times! 😉

[JeffSinSF]

Thanks for your reply, MrBosco.

I don't think I have enough sessions yet to do a good comparison since the first week I had the watch I didn't even start the Fitbit Exercise app to tell it to start a session.  I just relied on it to autodetect when I was moving.  That being said, what's hanging me up so far is the discrepancy between walking on a flat surface and walking uphill.  On the flat surface my heart rate didn't go as high, and yet for each step the flat walk was supposedly burning more calories than the uphill walk.  I'm a fairly logical person and this doesn't make sense to me.  I checked a website and entered my data for an uphill walk on the treadmill today to see what it said I would have burned for the same speed and number of steps without taking into account that I was walking uphill and that website said I would have burned 350 calories while the Fitbit app says I burned 240.  A fat man walking uphill for nearly 40 minutes burns only 240 calories?  It just doesn't make sense.

The other thing which has me leaning in the direction of questioning the worthiness of this device and its software is that I noticed last night when I was moving a tea bag up and down to get my tea to steep faster it was counting each arm movement as a step.  I get that the watch could be fooled into thinking I was walking if I swung my arm as much as it would naturally move when walking, but moving my forearm up and down a distance of probably 10 inches at most gets counted as a step? 

Maybe these devices are best used by people a lot more active than me (e.g., people who go for runs outside where their heart rates reach their peak zone for a half hour or more).  I'm going to continue to try to use it and see if I can do anything to get it to be more accurate for my purposes, but hopefully the watch will actually perform the function I wanted it most for --- to call an emergency contact if I happen to fall down. 

[stewie007]

fitbit is a huge trash 

dont buy premium. its trash trash trash

FITIBT = BAD

[CalvinWilkins]

Why you said that?