Garmin Connect Calorie Adjustment confusing!

pdavies77uk

I'm struggling to understand how MFP calculates and adjusts calories based on Garmin Connect.

I wear my Garmin constantly and obviously the amount of steps/exercise I do each day is fed into MFP. When I am under 10,000 steps for the day, MFP awards me calories. However, if I go over 10,000 steps a day, which I often do, it then starts giving me minus calories. For example, so far today I have done 15,467 steps but MFP has adjusted it so that I have -99 calories because of the amount of steps I have done. Surely this is wrong? Surely the more steps I do, the more calories I should be awarded?

Often the difference in calories between Garmin Connect and MFP can be as much as 400 calories because of this.

Anyone else had this experience or any advice how to correct it?

Lietchi

Very strange.
Do you log any workouts too through your Garmin?
Can you post print screens of your exercise diary and your calorie adjustment (tap on it in your diary to see the details) in MFP and total calorie burn for the day in Garmin?

heybales

MFP gets a daily burn value from Garmin.
MFP Subtracts what it thought you'd burn based on body stats and MFP selected Activity level.
If there are workouts MFP knows about (synced or manually entered), it subtracts that too (assuming the daily burn from Garmin already contains those calories that avoids double counting).

Garmin 2500 - 2000 MFP - 200 exercise = 300 adjustment.

Base eating goal 1500 + 300 adj + 200 exercise = 2000 new eating goal.
Still 500 cal deficit.


So you need to forget steps - nothing to do with the math, just allows you a quick figure for how active you may have been that day. Find compare days in Garmin looking at actual daily calorie burn.

In your MFP Exercise Diary, the line for Garmin Adjustment - press and hold to see more details.
Pick a past day with and without workout showing in MFP.
That's the screen shots to get.

Garmin Connect does do an odd way of working with MFP to figure your calorie burn, and their math is wrong for the intended purpose and stated eating goal, I wouldn't compare that at all.
They changed it recently for whatever reason, prior method was wrong too.
They are matching what Apple displays now, base sedentary burn, and active burn above that.

pdavies77uk

It does seem to take issue with those days I do run. On those days, once it has added that days run, it'll give me a minus calorie adjustment, based on the steps I've taken. I've attached screenshots of Sunday which may better explain it.

pdavies77uk

In addition to that, for "Calories Remaining" on that Sunday, Garmin Connect says 664 and MFP say 94. A huge difference.

Lietchi

The thing is: running = exercise. MFP is based on your activity level NOT counting exercise. So you can exercise as much as you want, but if you're not reaching the activity level you entered in MFP outside of exercise, calories will be taken away.

What activity level are you set at? And how many of those 23000 steps were part of your run (so how many outside of your run)?

Edit: Garmin also gives you gross calorie burns (including BMR calories) so those also get taken away through the MFP calorie adjustment (to avoid double counting BMR calories)

pdavies77uk

The activity level is set at "Active". Most weeks I will run for 5 or 6 days of the week, ranging from 5k up to around 12 miles. Sunday's run was 11 miles but Garmin Connect doesn't give a Step count just for that run unfortunately. Best guess would be around 10,000 steps.

yirara

So you additionally logged over 1k calories of exercise. What's going on there? Did you track hikes/walks? If so then Garmin might transfer the walk over, subtract the walk steps, and only transfer the additional steps as steps with calories over to MFP.

heybales

Again - the adjustment has NOTHING to do with the steps you take, at least you phrased it that way.
Leave steps out of any relationship, because frankly you can have big step days with little distance, and that's what matter to calorie burn. You can have fewer steps and more distance and more calorie burn.
Leave steps out - there is no math with them.

So MFP based on your selected Active level expects you to burn 3732-1190= 2542 calories already. Based on BMR about 1588 x 1.6.

Garmin reported you burned 3504 as of 10:49 pm - so 71 min left of the day, or 71/1440=4.93% of day left.

MFP expected 2542 x 4.93% = 125.3 calories left at your expected Active level you selected - that's likely not true at night - hence the recommendation to pick Sedentary on MFP.

Garmin so far 3504 + 125 MFP expected = 3629 daily estimated burn.

Garmin 3629 - 2542 MFP estimated - 1190 exercise = neg 103 adjustment

Base eating goal 2040 + 1190 exercise - 103 adj = 3127 new eating goal. 500 cal deficit still!

You ate 3033, leaves 94 remaining to eat.

Nope - all the math is correct with figures provided.


Again - ignore Garmin Connect math because they are not using the same figures expect for pulling from MFP what your eating goal is.
But they don't know your expected level of burn, they don't know the activity level, they don't know the deficit you selected.
That's why their math can't work.

Just look at their totals for daily burn and eating goal - they would have you eating at maintenance.

I can explain why their math fouls up if interested and what they are doing.
But the end of the matter is - don't use their eating goal.

Garmin for movement goals and logging.
MFP for eating goals and logging.

heybales

Meant to add - the reason you are getting negative adjustment, and I get them all the time especially on big exercise days.

You are less active outside of your workout than your Selected activity level - that's all.

Not surprising - many people have a hard workout wipe them out, and they move less than they would have if there had been no workout.

Fitbit folks always have this complaint it seems - Fitbit doesn't sync workouts over, so for them they only see the adjustment, and when it's lower than the Fitbit calorie burn on the workout - they think something is messed up.

Nope, the workout may have burned 1190 calories, but the net effect to the day was only 1087.

Be glad - I'm usually 200-300 neg on those days, and I'm set to sedentary!
You are set to active and only got neg 100!!!!!
Incredible.

pdavies77uk

Thanks for the advice. As you suggest, to get a more accurate reflection I'll have a look at dropping my activity level.

yirara

(not wanting to confused TO even more, but anyway... I had to reinstall the Connect app the other day, which really messed with my settings. Everything's messed up. But what annoyed me most is that the HR zones for general/walking had reverted to the standard values, which totally don't work for me. Running/Cycling/Swimming was still ok. So anyway, I walked 1km to run some errands. And was wondering why I got vigorous intensity minutes for that short walk. That's when I noticed the missing custom HR zones. Also noticed that I got nearly 300 calories for this short walk. Thus yes, it looks like Garmin takes HR into account when calculating calorie burns, and if the real ones deviate too much from 220-age then totally crazy non-existing calorie burns might appear, even for a short walk to a shop)

heybales

yirara wrote: »

(not wanting to confused TO even more, but anyway... I had to reinstall the Connect app the other day, which really messed with my settings. Everything's messed up. But what annoyed me most is that the HR zones for general/walking had reverted to the standard values, which totally don't work for me. Running/Cycling/Swimming was still ok. So anyway, I walked 1km to run some errands. And was wondering why I got vigorous intensity minutes for that short walk. That's when I noticed the missing custom HR zones. Also noticed that I got nearly 300 calories for this short walk. Thus yes, it looks like Garmin takes HR into account when calculating calorie burns, and if the real ones deviate too much from 220-age then totally crazy non-existing calorie burns might appear, even for a short walk to a shop)

Garmin is like the others - if you start a workout, or it thinks one started due to HR and steps and auto-starts one - HR-based calorie burn is used.
And that is inflated calorie burn at the bottom of the aerobic range above daily activity level, inflated at the top too right below anaerobic.

And yes HRmax is an important figure in HR calculations. Not the zones themselves, but those are based on HRmax also by default.

I would be curious if you adjust the HR-zones if it would NOT auto-start a workout. Or if that was due to another setting reset to default.

yirara

I did start a walking workout I think. My watch doesn't have the auto start of activities thingy.

heybales

pdavies77uk wrote: »

Thanks for the advice. As you suggest, to get a more accurate reflection I'll have a look at dropping my activity level.

That is mainly to control the nightly effect when you hit the couch and bed.

If you looked at calories remaining and ate to goal at say 8 pm, with 4 hrs of day left, MFP would have a decent amount left based on Active level, which of course the next morning Garmin will sync and tell MFP you burned BMR rate for the 4 hrs actually.
New daily burn given for prior day, MFP redoes math, and your eating goal drops by BMR/24x4hrsx0.6, and now you ate over goal.

With Sedentary set at least it's 0.2 rate over - and you can either calculate that, or see how much it typically changes by just noting the difference.
Then again if you stay up until midnight or don't go to bed until 11 and see last update then - doesn't matter.

So you'll see bigger adjustment set to Sedentary - but you'll typically learn about how many calories to plan your day around, knowing some adjustment in the evening will occur.

Mteud

heybales wrote: »

Again - the adjustment has NOTHING to do with the steps you take, at least you phrased it that way.
Leave steps out of any relationship, because frankly you can have big step days with little distance, and that's what matter to calorie burn. You can have fewer steps and more distance and more calorie burn.
Leave steps out - there is no math with them.

So MFP based on your selected Active level expects you to burn 3732-1190= 2542 calories already. Based on BMR about 1588 x 1.6.

Garmin reported you burned 3504 as of 10:49 pm - so 71 min left of the day, or 71/1440=4.93% of day left.

MFP expected 2542 x 4.93% = 125.3 calories left at your expected Active level you selected - that's likely not true at night - hence the recommendation to pick Sedentary on MFP.

Garmin so far 3504 + 125 MFP expected = 3629 daily estimated burn.

Garmin 3629 - 2542 MFP estimated - 1190 exercise = neg 103 adjustment

Base eating goal 2040 + 1190 exercise - 103 adj = 3127 new eating goal. 500 cal deficit still!

You ate 3033, leaves 94 remaining to eat.

Nope - all the math is correct with figures provided.


Again - ignore Garmin Connect math because they are not using the same figures expect for pulling from MFP what your eating goal is.
But they don't know your expected level of burn, they don't know the activity level, they don't know the deficit you selected.
That's why their math can't work.

Just look at their totals for daily burn and eating goal - they would have you eating at maintenance.

I can explain why their math fouls up if interested and what they are doing.
But the end of the matter is - don't use their eating goal.

Garmin for movement goals and logging.
MFP for eating goals and logging.

Thanks for the detailed explanation.
I would be curious however to know why is Garmin’s calculation so wrong.
Connect pulls from MFP the calorie goal. It then adds active calories and gives you an adjusted goal. I would expect that the active calories in Garmin Connect to be the same as Exercise calories in MFP but clearly they are not.

Jthanmyfitnesspal

@heybales is rather an expert on this. I've spent way too much time thinking about it as well. I looked through what @heybales says, but I though a little more explanation might help.

It isn't that Garmin is wrong and MFP is right, it's just that the two platforms are doing the calculations differently. When you put your stats into MFP, including your "activity level," MFP calculates a daily calorie plan for you. Garmin does the same thing. If you set MFP to "sedentary," then you will have the base calorie plan matching on both platforms. This is a good way to set things up.

Garmin then attempts to calculate your added calories due to exercise and incidental activity. One little foible is that, when you do an activity, Garmin gives the "gross calorie burn" during that activity, including whatever you would have expended during the time period if inactive. It then sends this number to MFP and also sends your total calories burned for the day. When you record a longer workout, there can be a few 100 double-counted calories. Garmin fixes it on their site, but still sends MFP the total for the day (corrected) plus the (gross) calories for exercise. MFP is trying to match Garmin's total, and thus shows a negative correction. The negative corrections \will be larger if your plan is set higher, so it works best when your plan is set to "sedentary."

Now, let me add that Garmin's calorie estimate is not perfect and you have to do a little estimation yourself. It gets easier with practice!

heybales

Mteud wrote: »

heybales wrote: »

Again - the adjustment has NOTHING to do with the steps you take, at least you phrased it that way.
Leave steps out of any relationship, because frankly you can have big step days with little distance, and that's what matter to calorie burn. You can have fewer steps and more distance and more calorie burn.
Leave steps out - there is no math with them.

So MFP based on your selected Active level expects you to burn 3732-1190= 2542 calories already. Based on BMR about 1588 x 1.6.

Garmin reported you burned 3504 as of 10:49 pm - so 71 min left of the day, or 71/1440=4.93% of day left.

MFP expected 2542 x 4.93% = 125.3 calories left at your expected Active level you selected - that's likely not true at night - hence the recommendation to pick Sedentary on MFP.

Garmin so far 3504 + 125 MFP expected = 3629 daily estimated burn.

Garmin 3629 - 2542 MFP estimated - 1190 exercise = neg 103 adjustment

Base eating goal 2040 + 1190 exercise - 103 adj = 3127 new eating goal. 500 cal deficit still!

You ate 3033, leaves 94 remaining to eat.

Nope - all the math is correct with figures provided.


Again - ignore Garmin Connect math because they are not using the same figures expect for pulling from MFP what your eating goal is.
But they don't know your expected level of burn, they don't know the activity level, they don't know the deficit you selected.
That's why their math can't work.

Just look at their totals for daily burn and eating goal - they would have you eating at maintenance.

I can explain why their math fouls up if interested and what they are doing.
But the end of the matter is - don't use their eating goal.

Garmin for movement goals and logging.
MFP for eating goals and logging.

Thanks for the detailed explanation.
I would be curious however to know why is Garmin’s calculation so wrong.
Connect pulls from MFP the calorie goal. It then adds active calories and gives you an adjusted goal. I would expect that the active calories in Garmin Connect to be the same as Exercise calories in MFP but clearly they are not.

Garmin does the math wrong.

The eating goal on MFP was a deficit from some estimated daily burn figure - say 2000, deficit 500, eating goal 1500.

Garmin receives the 1500 as info from MFP - it a requestable field in the API. I think Deficit is too - they should have asked for that.

MFP finds out from Garmin you actually burned say 2400 for the day - maybe some exercise, and maybe some daily activity above the sedentary that estimated 2000.

400 extra burned, to keep 500 cal deficit, eating goal should go up by 400 too.

There is extra math there depending on workouts showing up on MFP or not - but the end result is the same.

Garmin 2400 - MFP 2000 = 400 cal adjustment.

1500 base eating goal + 400 adj = 1900 new eating goal (500 cal deficit as desired)


Garmin builds the other direction though, and it doesn't work out correctly even with themselves - because they don't know what the deficit was to reach that eating goal.

They split that daily burn 2400 the device comes up with into Resting calories, which is tad more than BMR but less than Sedentary on MFP. (figures at home on paper, I think 1.2 x BMR that is close to MFP's BMR), so 1920 in this case.
Any extra burn, workouts or extra activity, is added to Active Calories less the Resting calories for that chunk of time.
So say 400 in exercise would be shown as say 320 Active Calories for the hour it took to get them, or less if it took longer, say 300 then.

1500 Daily Goal + 300 Active Calories = 1800 Adjusted Goal

600 cal deficit just looking at the math on the screen of Total Calories Burned less Adjusted Goal.

This is totally a Garmin error not understanding how you should get an adjusted eating goal.
At least I think a deficit from a daily burn is the correct way, the MFP method.


Here are my exact figures from a big workout on Dec 14th.
Total Calories Burned
1,083 Active Calories + 2,038 Resting Calories = 3,121 Total Calories Burned
Adjusted Goal
2,100 Daily Goal + 1,083 Active Calories = 3,183 Adjusted Goal
Calories Remaining
3,183 Adjusted Goal − 0 Consumed = 3,183 Remaining

MFP on that day was set at maintain with indeed 2100 daily eating goal.
Active calories was mostly from an about 1.5 hr bike ride Garmin had at 1181 calories - 127 of the Resting calories already included there = 1054 Active.
30 extra calories from somewhere - hmm must have moved extra outside the workout.
So Garmin would have me eating 60 extra - not too bad in this example - I've seen much worse.

MFP had for math:
Garmin 3121 - MFP 2100 - 1181 known workout = neg 160 Adjustment

base eating goal 2100 + 1181 workout - 160 adj = 3121 (no deficit)

MFP's math much better. There is no double-counting in here.

nedliug

Ok, reading through the thread, this sounds messed up.
So "activity level" has nothing to do with workouts? So, if I'm training for an ironman and doing 2-3 workouts a day, but otherwise working a desk job or resting, then my "activity level" should be "sedentary"?

AnnPT77

nedliug wrote: »

Ok, reading through the thread, this sounds messed up.
So "activity level" has nothing to do with workouts? So, if I'm training for an ironman and doing 2-3 workouts a day, but otherwise working a desk job or resting, then my "activity level" should be "sedentary"?

Technically, yes.

Then you add log your exercise when you do it, which gives you credit for the exercise calories, and you eat those, too. If you're synching a tracker, and training as you describe, you'll get a large calorie adjustment most days from the tracker/MFP synch process.

Are you actually training that much? (Most people aren't.) If so, you'll get those exercise calories when you log the exercise, or when the tracker synch happens. It's not like you don't get credit for the workouts, it's just a question of how the math to include the workouts happens. Most other calculators average in planned exercise over the week; MFP credits you with the exercise when it actually happens, that's all.

Your comment/question really doesn't have a lot to do with the core of OP's question, though. Maybe start your own thread?

matteolavaggi172

I'm reading everything I can about this problem, but I can't make sense of it. Garmin receives daily calorie goals from mfp and the consumed with meals always mfp. So how is it possible that on a typical day there are 600 calories of difference between MFP and Garmin on the remaining ones? mfp uses Garmin data to calculate any additional or negative calories based on the day so far, but the totals should be the same! how is it possible that Garmin reports a total calories remaining that is 2 times higher than mfp?