Fitness trackers. Over calculating daily calories burnt.

temple_girl

Hi there,

Looking for a little advice on fitness trackers. I used to have the Fitbit HR charge 2 had it for about 18 months. Had no issues with it apart from I found that it over calculated calories during a workout.

I have recently bought an iwatch series 5 I find this is a lot better calculating calories burnt during exercise. However I find the daily calories burnt rather excessive. I wore my Fitbit on the same day as my iwatch to compare and the iwatch was about 700 calories higher than the iwatch. I am quite over weight and I’m trying to lose weight normally and healthily but trying to work out a calorie deficit when I’m not sure what is accurate is difficult. I know no fitness tracker is perfect and I won’t take every calorie as gospel but just a ball park figure would be nice.

Any advice please. Thank you.

Lillymoo01

Is it possible that you are overestimating the food you are eating rather than the Fitness watches overestimating the calories burned? Which one gave you 700 calories more as you mentioned the Apple watch twice in that sentence? One figure could be total calories burned so far for the day and the other additional calories burned through exercise.

My suggestion is to tighten up on your logging. Weigh everything before you eat it, including prepackaged food and condiments. Check that all entries are correct by comparing them with food labels of the USDA website for fresh produce. If you get this as accurate as possible it is much easier to work out inaccuracies with Fitness watches.

AnnPT77

Have you double checked your user profile entries/personalization to make sure they're correct in both? Usually there are settings for age, weight, etc., but possibly there may also be things like stride length that take a little more to get right. There may also be "self tests" that you can set up and run.

(I can't be more specific for iWatch or Fitbit, because I have a Garmin, not either of those, but there are some things that are generally common across the different makes/models.)

It's also the case that depending on how a tracker is designed, it will take a little time to measure your activity and adapt based on what it sees over a period of time. In effect, some models "learn" you. So, your iWatch could get more accurate with "experience".

Regardless, all that any of these devices do is estimate calories. Depending on the model, they measure different things: Heart rate, step-type impact, altitude changes, speed of geographic movement (via GPS), arm movement of various sorts, etc.

Some of those things correlate loosely with calorie expenditure, but none of them match up exactly with calorie burn. So they use research on other people, statistics, and the measurements to make educated guesses. If they know explicitly what activity you're doing (because you tell them via a setting, or they guess right), they have the potential to make better estimates, but whether they do so depends on the features of their programming.

What you can do is log your eating carefully, and compare that plus your weight change to what your tracker says, over a period of 4-6 weeks, at least. (Premenopausal women need to compare the same point in two or more different menstrual cycles.) That will give you enough data to adjust your intake to a sensible level, and decide whether your device is close enough to synch to MFP, or you need to adjust its estimate in some way to be accurate enough to use for all-day burn.

My device, set up as accurately as I can get it, materially understimates my daily calorie expenditure (by several hundred calories) as compared with 4+ years of logging and weight data. This is rare, but it happens. Since I already had an established logging habit, and a good idea of my average TDEE (all day calorie burn) when I got it, I don't bother to synch it. For me, it would be more trouble than it's worth. I do use it for calorie esimates for some types of exercise, where it seems to be close enough to be useful.

Duck_Puddle

I have the same experience with my Apple Watch series 4.

Workout calories are quite reasonable and in line with other reasonable estimates (slightly on the low side even).

Overall calories for the day is consistently about 20% over my actual TDEE (as measured via weighed food logging and weight change).

My Apple health settings (height, weight, etc.) are fed from Garmin-which managed to use those same numbers on its device to produce a more reasonable estimate of TDEE.

I’ve had various trackers linked to mfp over the years. What I’ve done for those that overestimate TDEE is set my activity level on mfp higher than it should be - which lowers the amount of the calorie adjustment (the calorie adjustment is only the amount over whatever mfp expects you to burn).

I do highly recommend very carefully logging absolutely everything you eat, and weighing all your solid foods (even prepackaged), and using a weight trending app for at least 6 weeks. Compare what you’ve eaten to the change in your trended weight to figure out your actual TDEE. Then compare that to what your Apple Watch said.

Adjust as necessary.

I have done that part but wanted to share that my Apple Watch also has a crazy high total calories burned number even though the workout calories are very reasonable.

PAV8888

Just a quick note that assuming that integration is working correctly--which is definitely not a gimme on a day-to-day basis with every tracker, and definitely and additionally there have been serious questions about the correctness of direct Apple watch synchronization with MFP without the use of an intermediary app--the activity level you choose on mfp is immaterial in terms of your total calories number at the end of the day.

If integration is working correctly, the adjustment number will change such as that at the end of the day your MFP total calories expended number will be equal to what your tracker detected.

The only difference will be whether the adjustment will be smaller or larger, in all cases arriving at the same final number.

Again this is assuming that integration works

Duck_Puddle

PAV8888 wrote: »

Just a quick note that assuming that integration is working correctly--which is definitely not a gimme on a day-to-day basis with every tracker, and definitely and additionally there have been serious questions about the correctness of direct Apple watch synchronization with MFP without the use of an intermediary app--the activity level you choose on mfp is immaterial in terms of your total calories number at the end of the day.

If integration is working correctly, the adjustment number will change such as that at the end of the day your MFP total calories expended number will be equal to what your tracker detected.

The only difference will be whether the adjustment will be smaller or larger, in all cases arriving at the same final number.

Again this is assuming that integration works

If your base calorie goal goes up when you increase your activity level-that’s true. If you keep the original base calorie goal (I forgot that part), overstating your activity level on mfp covers the overage on the tracker.

My calorie goal is 1200 (my actual NEAT is 1600).

My tracker says I burned 2500 for the day. I know from experience/tracking that this is about 300 calories too high.

If I leave myself as sedentary (which is my actual activity level), my adjustment will be 900 (2500-1600) and my total calories will be 2100 (1200+900).

If I set myself as lightly active (and keep the same base calorie goal), mfp now thinks my NEAT is 1900 and my adjustment is will be 600 (2500-1900) and my total calories will be 1800 (1200+600).

If you set your activity level on mfp to one where the difference between that and your actual neat is roughly the same as the amount the tracker overstates (and keep the calorie goal for your correct activity level)-it all works just right.

brittlb07

I should note that I don’t work out THAT much, but I don’t eat much extra even when I do. This seemed to work well for me. I’ve read to eat back about 50% of the calories that these devices claim because of the issues you are stating.

PAV8888

Hmmm.

I think you may be playing/benefiting/ impacted by 1200 being floor calories below which MFP does not go.

Otherwise your goal eating calories rise in tandem and in lockstep with your activity level.

F.e. if at -500 and sedentary my eating goal is 1800 and expected MFP NEAT 2300, at -500 and very active my eating goal will be 2300 and expected MFP NEAT will be 2800, each incrementing by the same 500.

If my Fitbit TDEE is 3000, positive adjustment will be 700 and 200 respectively and eating goal the same 2500.

If my Fitbit TDEE is 2500, it will be a positive +200 and a negative -300 adjustment respectively, with eating goal being the same 2000

The example is made up; but the mechanism was tested/verified about 3-4 years ago and I've not had an indication it has changed.... (has it?)

Duck_Puddle

PAV8888 wrote: »

Hmmm.

I think you may be playing/benefiting/ impacted by 1200 being floor calories below which MFP does not go.

Otherwise your goal eating calories rise in tandem and in lockstep with your activity level.

F.e. if at -500 and sedentary my eating goal is 1800 and expected MFP NEAT 2300, at -500 and very active my eating goal will be 2300 and expected MFP NEAT will be 2800, each incrementing by the same 500.

If my Fitbit TDEE is 3000, positive adjustment will be 700 and 200 respectively and eating goal the same 2500.

If my Fitbit TDEE is 2500, it will be a positive +200 and a negative -300 adjustment respectively, with eating goal being the same 2000

The example is made up; but the mechanism was tested/verified about 3-4 years ago and I've not had an indication it has changed.... (has it?)

Right. That’s why I said keep the base calories (eating goal) from the correct (lower) activity setting. I forgot that in my original mention and that’s paramount.

I’m not changing anything with Fitbit (or any tracker). Just changing what my activity level setting is on mfp and keeping the original base calorie goal.

My base is 1200 at sedentary-neat is 1600.

If I up my mfp setting to lightly active (and don’t keep the original calorie goal), my base goes up to 1400 my neat to 1900 (base would also go up 300 if I wasn’t hitting the 1200 floor on sedentary).

If I just change my activity level on mfp (and it raises base calories)-then we end up with exactly the same total calories - just one has lower base/higher adjustment, one has higher base/lower adjustment. That’s what your examples show.

The mechanism hasn’t changed. What I’m manipulating is what mfp thinks my neat is (without changing my eating goal). I get the lower adjustment (based on my higher neat) but keeping original eating goal. So my total calories is lower-by the difference in neat between the original and higher activity level.

So in your example-you change your setting to very active on mfp AND keep your original base goal of 1800. Mfp thinks your neat is 2800. It makes the adjustment based on 2800. Your Fitbit still records a TDeE of 3000 or 2500. You’ll have the same adjustment amounts that you calculated based on the 2800 neat (+200 or -300) but with the lower original eating goal, your total goal to eat will be 2000 or 1500. If your Fitbit is overestimating by 500-ish, this puts you at the right deficit.

I’ve done it with 4 of the 9 or so trackers I’ve had linked over the year’s. It allows me to keep the tracker linked and benefit from that without having to do a bunch of math and/or manually lower my base calorie goal to a number that mfp doesn’t allow.

psychod787

My charge 2 would have me eating an added 1500 calories over sedentary today. Lol yeah right!

PAV8888

I think I got ya!

I have less than a 5% divergence over the long term if I am logging carefully (so in the 100 range, 200 as an outside value), and I also use my target calories as a warning to slow down as opposed to limit (I am perpetually in the red)

So I am hand-set at -400 from very active, and after taking into account the Fitbit error, it is maybe enough to slow me down when I am in cookies mode (generally more than 500 over goal), and generally enough to real me in slowly when I am being "careful" and only hover in the +200 to +400 range.

Though I think that we are getting into individualized manipulation of a system we both understand fairly well, vs things useful to people starting out.

The biggest issues I've heard about (no first hand knowledge) with Apple watches is that exercise calories sometimes are erroneously removed as opposed to added during integration with MFP

In any case, given that most trackers generally use similar formulas (MET values and mifflin-st jeor RMR) as their starting values, it would be curious to see why there is such a large divergence between the Fitbit and apple watch worn at the same time.

I was under the IMPRESSION that Fitbit and Apple had similar TDEE values with Garmin being more conservative with both exercise and TDEE.

IMHO Fitbit gives a touch too much credit for detected movement; but, also gives less than enough credit for no detection, and the two tend to average out fairly well for most with "subtle encouragement" for people to exercise more and be more active.

temple_girl

Thanks everyone. Really good advice and a lot to take on board there.
My profile statistics are all correct and up to date so it should be calculating correctly as per those.
I don’t tend to do by daily calories burnt I’m logging and tracking everyone and sticking to what MFP says I should be eating daily for my height and weight to lose a certain amount of weight weekly.
I just wondered what issues everyone else was having. My total daily calories burnt yesterday was 2677 which when sticking to what MFP says I should be having of 1610 a day is a big deficit.
I wish I really did burn that amount a calories a day. 😂🙌🏼
Thanks everyone.

Duck_Puddle

PAV8888 wrote: »

I think I got ya!

I have less than a 5% divergence over the long term if I am logging carefully (so in the 100 range, 200 as an outside value), and I also use my target calories as a warning to slow down as opposed to limit (I am perpetually in the red)

So I am hand-set at -400 from very active, and after taking into account the Fitbit error, it is maybe enough to slow me down when I am in cookies mode (generally more than 500 over goal), and generally enough to real me in slowly when I am being "careful" and only hover in the +200 to +400 range.

Though I think that we are getting into individualized manipulation of a system we both understand fairly well, vs things useful to people starting out.

The biggest issues I've heard about (no first hand knowledge) with Apple watches is that exercise calories sometimes are erroneously removed as opposed to added during integration with MFP

In any case, given that most trackers generally use similar formulas (MET values and mifflin-st jeor RMR) as their starting values, it would be curious to see why there is such a large divergence between the Fitbit and apple watch worn at the same time.

I was under the IMPRESSION that Fitbit and Apple had similar TDEE values with Garmin being more conservative with both exercise and TDEE.

IMHO Fitbit gives a touch too much credit for detected movement; but, also gives less than enough credit for no detection, and the two tend to average out fairly well for most with "subtle encouragement" for people to exercise more and be more active.

Agreed! I felt bad taking this so far off track on that.

I think it’s a bit more device specific than brand specific. I’ve had 4 Garmin devices as trackers (ranging from lowest to about highest price). My vivomove (~$100) significantly underestimated, my currently Fenix 5s plus overestimates. My Fenix 5S was about dead on. My vivoactive clearly thought I was a 600 pound man (I’m a 160 pound woman). But others have had different experiences with the same models. So maybe it’s the types of movements I do? Who knows.

I’ve had similar results with assorted Fitbit’s (although I’ve not had any that underestimated) and Polar.

Because my TDEE is often a bit under 2000 and my calorie intake averages around 1600, a 200+ calorie difference can be significant over time. When it’s upwards of 300-500, it’s significant.

I’m fascinated by my Apple Watch-which tends to give very reasonable workout estimates, but vastly overestimates TDEE (for me). I’m not sure why or where the difference is. I don’t have the diligence to wade through minute by minute reports in Apple health to try and figure it out. I haven’t seen many reports like that (Workouts right but overall is high)-so my original intention was just to offer support that I am finding that same experience.

Then I drove the bus 1700 miles off course.

DoctahJenn

Honestly, to simplify things, for a while I removed my tracker link and got a good old fashioned chest strap HRM for my dedicated workouts. Then I set my goals on MFP to my normal activity level and just manually added my HRM burn. It worked incredibly well and saved me the headache of mental calculations.

Now I’m too lazy to put on a chest strap, so I have a set calorie goal that assumes I’m going to be reasonably active and tell myself to work out each day without adding calorie burns. 😜 After a while you get a feel for how much you burn during a proper workout, so it became easier.

Noahsmum1975

Sorry ...but what is TDEE and NEAT? I'm so confused !

angmarie28

I love my fitbit, but mine also over estimates what I burn, If I did burn that amount daily, Id be skinny, because I 100% dont eat that much. I use it as a goal to burn a certainnumber daily, but dont use it in any of my actual weightloss tracking

AnnPT77

Noahsmum1975 wrote: »

Sorry ...but what is TDEE and NEAT? I'm so confused !

New here? This may be a useful post:

http://community.myfitnesspal.com/en/discussion/1069278/acronyms-and-terms-for-new-mfp-members-v-6/p1

Relevant here:

BMR is basal metabolic rate, pretty much what you'd burn lying around, not moving.

NEAT is non-exercise activity thermogenesis, the number of calories you'd burn via BMR plus the activities of normal everyday life (work, home chores, non-exercise hobbies, etc.) but not including intentional exercise. If you follow the instructions for MFP set-up, the calorie estimate it gives you is for NEAT (minus a deficit or plus a surplus, if you said you wanted to lose or gain weight, rather than maintain your current weight).

TDEE is total daily energy expenditure, the number of calories you burn per day, total, from BMR plus everything you do, including both non-exercise activity, and intentional exercise. There are calculators outside MFP that will give you a TDEE estimate, which averages in your expected exercise. Some people use these because they like to eat the same amount every day, exercise or no. The catch is that if you don't do the expected exercise, your weight management goals may be compromised.

Fflpnari

I had the charge 2 and found it pretty accurate for calorie burn. I ate 2400 calories a day for a few months and calculated with the calories burned on fitbit. Averaging 3000-3200 calories burned daily.
It worked well, and I lost 40 pounds