Active Calories & Daily Steps Not Adding Correctly (MATH WRONG)

Same here with FR235. Since about May.
I told MFP - they will ask Garmin, if Garmin changed anything. As soon as there is time...
I told Garmin. No reaction at all.

It seems, that Garmin changed some calculations - and nobody cares.:mad:

So I think I won't care any further too and go back to my good old fitbit. Their results were always ok for me.

It's because they changed how resting calories are calculated with recent firmware versions. Its inaccurate and significantly underestimates resting calories. All they are doing now is calculating base bmr instead of factoring what a normal person burns being sedentary.

So most people won't even get an adjustment coming across. For me garmin now underestimates my sedentary bmr by 350 calories per day.

Is any one having a problem like described in the attachment. I have been having this problem since the June 8th issues. Also I am no longer getting a MFP calorie adjustment, the only thing that syncs to MFP is my activities created. If you do not use MFP then this part does not apply to you. I am using a Forerunner 25 if it is device specific.

In short, once I create an activity the Daily Steps Calories that I accumulated prior to exercising are wiped out to 0 and I not longer accumulate Daily Steps Calories for the rest of day. Also the MATH does not add up to the calories that I burned. I have not idea what happened to this, and I am having intermittent issues with getting Garmin to actually take a proactive look at this.

Any help to put my mind at ease is greatly appreciated.

In addition to my post above I'm now seeing this too. Example if I accrue 50 calories though daily steps and then burn 500 on a run, my 50 daily step calories are wiped out too. This happened recently and not sure if it the Fenix3 software or Connect doing this.

Garmin why do you keep messing with the calorie formulas? It used to work well until the resting calorie calculation was changed. And now even daily step calories are working as designed for some reason.

And now I have this issue too, along with the other calorie issue GCBS posted about. I'm not sure when it started but did a little experiment today to test this. I recorded a 2 mile walk this morning on my Fenix 3 and noticed that I had no daily steps calories after it. I went for another recorded walk in the afternoon and checked beforehand and I had 10 daily step calories for 1.05 miles (way too low as well) and then after syncing my new walk activity it was wiped out to zero.

I wonder if users with devices with continuous heart rate are having this issue? The other calorie issue seems to mostly be a problem for devices that don't have that, though it seems from the posts I've seen like devices with HR have the opposite problem and are reporting calories that are too high.

Either way this is getting ridiculous. Part of the reason I switched from Fitbit to Garmin was because of Fitbit's unilateral changes to their software and poor communication and now it looks like the same thing is happening here. Both these issues need to be fixed, for HR and non HR units alike.

It's because they changed how resting calories are calculated with recent firmware versions. Its inaccurate and significantly underestimates resting calories. All they are doing now is calculating base bmr instead of factoring what a normal person burns being sedentary.

So most people won't even get an adjustment coming across. For me garmin now underestimates my sedentary bmr by 350 calories per day.

I am not trying to diminish what OP reported- clearly there is at least a math problem in a report as well as an open question on why step calories are shifting.

That said, GCBS is wrong about BMR, which is likely why Garmin is messing with BMR numbers.

BMR is a constant value for any particular person of given height, age and weight +/- individual metabolic differences not easily measured. Activity level has NO effect on BMR (Basal Metabolic Rate). BMR is what you would burn in a coma.

"Sedentary BMR" does not exist. GCBS is confusing "sedentary BMR" with TDEE. TDEE is total daily energy expenditure which does certainly vary based on activity. For example, let's say a true BMR is 1600 for person X. (remember, this is what you burn in a coma, without any additional activity or efforts). Remember, for that person as long as weight and age don't change, BMR does not change.

TDEE = BMR + calories burned from extra effort, be it digesting food, walking, running, elevated heart rate, you name it. GCBS mentions sedentary BMR - but the right definition for that is TDEE for a sedentary person with his demographics.

(Note I'm about 1-2 month into using Garmin, but having read the history on this issue I'm pretty close to it.

1) For some amount of time in the past, Garmin reported BMR as GCBS mentions- by taking BMR and adjusting it upwards by some factor to represent the basic burn from simple things for a sedentary person. (Or perhaps used activity level flag to moderate that adjustment.

2) Then, not too long ago, and timeframe varying on your device, Garmin made a change. They decided to be accurate with the technical definition of BMR. SO what they did was report what I'll call "True BMR" as the resting calories (also labeled step calories) under activity. This is what GCBS describes.

This makes Total Calories on the day (TDEE) = BMR + active calories from non-measured activities (your "walking" or step" calories) + extra calories burned from a measured run or walk etc. as an activity. The burn from being sedentary is included in the walking or step calories (at least for the VAHR, though should be same for everyone IMO)

This is by the book. Especially for the 24/7 OHR models, this is simple to implement.

3) Unfortunately, this confused some people with OHR models and perhaps also confused people with non OHR devices since the measurement scheme is different. so...

As of last week, unnanounced, Garmin pushed ANOTHER algorithm change -at least to the VAHR. It appears that they have reverted to representing BMR the way GCBS wants it- claiming that BMR should now be about 20% higher than seen before for VAHR users in a post buried deep in a forum.

Now, Fenix is a different unit and firmware, but what you guys describe actually makes some sense if you bear with me.


On the VAHR, they jacked BMR by 20% (400 cal for me). However, given my definition in point 2, they would need to reduce active calories (which include step cals) by 400 over the day in order for there to be no net change in the total for two identical days. THeyh ave that all wrong for vahr.

**However, I wonder if what is happening in this thread is that you burn cals in a recorded activity...and they got it correct that they need to back these cals out of step cals to make up for the way BMR definition changed?


Here is the test to see if they are doing what I think or merely bungled the implementation of it. Try not measuring a run or any other activity until the evening - by which time you should have a decent number of "step" calories built up. Then record a run or walk or something and make sure that you burn fewer cal on that run/walk than you had showing for step cals.

If yo had 900 step cals, burned 400 on your run, I'm curious if step cals go to zero or if they reduce down to 900-400=500.


Love to know the answer

That said, GCBS is wrong about BMR, which is likely why Garmin is messing with BMR numbers.

BMR is a constant value for any particular person of given height, age and weight +/- individual metabolic differences not easily measured. Activity level has NO effect on BMR (Basal Metabolic Rate). BMR is what you would burn in a coma.

"Sedentary BMR" does not exist. GCBS is confusing "sedentary BMR" with TDEE. TDEE is total daily energy expenditure which does certainly vary based on activity. For example, let's say a true BMR is 1600 for person X. (remember, this is what you burn in a coma, without any additional activity or efforts). Remember, for that person as long as weight and age don't change, BMR does not change.

TDEE = BMR + calories burned from extra effort, be it digesting food, walking, running, elevated heart rate, you name it. GCBS mentions sedentary BMR - but the right definition for that is TDEE for a sedentary person with his demographics.

This whole "resting" calories lingo needs to be removed and perhaps they call it "baseline" or "sedentary" calories. I've had the Fenix3 non-HR version for 6 months. Was only recently that the "resting" calories calculation changed. I term Garmin resting calories as base BMR x a sedentary TDEE multiplier as that's how it was done up until recently.

Shouldn't the baseline for daily calories be BMR x a sedentary TDEE multiplier? If that isn't considered than it really messes up myfitnesspal adjustments (MFP uses BMR x 1.25 to determine sedentary TDEE to maintain your weight before exercise, and that seems to work for a lots of people...). Which I feel is accurate. Garmin was using 1.2 which is the traditional sedentary TDEE multiplier so the adjustment was always off by a bit....but now the difference is huge w/o a multiplier at all.

Then, ontop of that baseline, any significant walking steps and recorded exercise should be added in addition to the this baseline to determine a total TDEE. If you only use BMR and not factor in enough NEAT etc before exercise (which is what the typical 1.2 or 1.25 value accounts for), it underestimates your daily burn and of course completely messes up MFP adjustments.

That's why I've been suggesting Garmin give us the option to have a custom baseline multiplier or at least a few options. Myself I go by the Katch-Mcardle formula so my base BMR is a bit higher than normal....but I still want a sedentary TDEE multiplier added before any logged steps/exercise are added.

Daily step calories outside of recorded exercise were typically 100 cals/5000 steps which seems fair, but now even that is messed up as well.

How many step calories is the VAHR giving you a day?

I'm honestly just frustrated that, even with there being no calories from 'daily steps', the two activities I have don't add up correctly.

Running activity = 702 cals
Walking activity = 64 cals

Total should be = 766
What GC says = 751

I have an FR235 with OHR. This shouldn't be difficult!

This whole "resting" calories lingo needs to be removed and perhaps they call it "baseline" or "sedentary" calories. I've had the Fenix3 non-HR version for 6 months. Was only recently that the "resting" calories calculation changed. I term Garmin resting calories as base BMR x a sedentary TDEE multiplier as that's how it was done up until recently.

Shouldn't the baseline for daily calories be BMR x a sedentary TDEE multiplier? If that isn't considered than it really messes up myfitnesspal adjustments (MFP uses BMR x 1.25 to determine sedentary TDEE to maintain your weight before exercise, and that seems to work for a lots of people...). Which I feel is accurate. Garmin was using 1.2 which is the traditional sedentary TDEE multiplier so the adjustment was always off by a bit....but now the difference is huge w/o a multiplier at all.

Then, ontop of that baseline, any significant walking steps and recorded exercise should be added in addition to the this baseline to determine a total TDEE. If you only use BMR and not factor in enough NEAT etc before exercise (which is what the typical 1.2 or 1.25 value accounts for), it underestimates your daily burn and of course completely messes up MFP adjustments.

That's why I've been suggesting Garmin give us the option to have a custom baseline multiplier or at least a few options. Myself I go by the Katch-Mcardle formula so my base BMR is a bit higher than normal....but I still want a sedentary TDEE multiplier added before any logged steps/exercise are added.

Daily step calories outside of recorded exercise were typically 100 cals/5000 steps which seems fair, but now even that is messed up as well.

How many step calories is the VAHR giving you a day?

The problem with Garmin's change to using no multiplier with BMR (I agree that it could be labeled better) is that for non-OHR models Garmin did not increase the calories per step along with removing the multiplier. That means the burn from being sedentary is not included in the step calories on the non-OHR models. The previous calories per step value likely took into account the multiplier and so it was lower than it should be with no multiplier and the results seem way too low and as GCBS mentioned really mess up Myfitnesspal syncing.

One extreme example I mentioned in another thread was where I spent an entire day on my feet and walked 42,450 steps (21 miles) but had no recorded activities that day. Garmin Connect claims I only burned 2,911 calories that day. That is way too low...

I could write 25 pages in response but will refrain :)

1) Garmin needs to realize they are no longer a step-tracking company. There was a time when simple pedometers were the state of the art, but today they aren't. While some people may really care about their steps, many do not. Especially those who have the more advanced running watches or who put in high volume workouts of whatever sport.

They need to figure out how to remove any semblance of step-lingo, except where it makes absolute sense. Steps taken during a walk? sure. During a run? makes sense for calculating stride length and efficiency, but is no measure of effort. So when it comes to calories burned outside of measured activites- they need to de-couple at least the phrasing from the word step, since it so often does not apply. (Plus the devices in most cases use far more than number of steps taken to estimate exertion (calories). By reporting burn as step burn that has nothing to do with steps, noone knows what the hell is going on.

2) GCBS - your 100 cals/500 steps metric is likely off by a factor of 2. and,
Pauldn - your 21 miles at 2911 calories actually is quite reasonable.

http://knightsofknee.com/calculators/walking-calories-pedometer-steps.html This link lets you calc calorie burn estimates based on steps/mile (stride length) and weight. If you don't know your steps/mile they have this link that provides a standardized table. https://www.verywell.com/how-many-walking-steps-are-in-a-mile-3435916

According to that site, I should burn 215 cals for every 5,000 actual steps.
Enter your pedometer steps: 5000
What are your steps per mile? 2218
Enter your weight: 180
Select your pace from the list below: I chose: 3 mph - 20 minutes per mile
Result: You walked 2.25 miles, and burned 215 calories.

Unless GCBS is a very thin/light guy, his estimate is under per this tool.


I re-ran it for pauls' scenario as well. Assuming Paul didn't round, 42450 step over 21 miles is 2021 steps per mile

Enter your pedometer steps: 42450
What are your steps per mile? 2021
Enter your weight: 180 (I used mine)
Select your pace from the list below: 3 mph - 20 minutes per mile
You walked 21.00 miles, and burned 2005 calories.

This seems reasonable since I can''t imagine you did much else all day than walk and it gives 900 calories for the rest of the day.

3) In terms of how non-active calories are measured I agree with GCBS that the user option to choose the BMR model would be great. Even better if user decides if it reports pure BMR or BMR + sedentary multiplier.

However, if the user is stuck* with pure BMR and no multiplier, GCBS is incorrect re: MFP adjustments. I use MFP, allow my exercise adjustments to get pushed over and have no problem despite the fact that I've been lazy and left my MFP starting point at a different calorie number than the Garmin BMR (or even the Garmin BMR+multiplier that is now appearing this week).

The whole point of shipping timed exercise over to MFP and then recording an adjustment in MFP is to 100% sync GC and MFP projected total day calorie burn to the same number. It works perfectly well in doing that, especially if you allow negative adjustments in MFP. It can cause GC to show more calories remaining to be eaten if the start points are different, but I rely on mfp for the number.

And if you want MFP to always = GC, then just manually set the goal in MFP to equal the True BMR in GC or BMR+multiplier in GC if that's what your device uses. You don't need to be mfp premier to do that.

4) Lastly- very simplistically - THere are a limited number of calorie burns that exist. BMR + exercise/activity you time/record + exercise/activity you do not specifically record + what I'll call background calories (NEAT as described by GCBS) = Total calories. That is it, unarguably. the only issue is how the devices estimate/compute each of those chunks and how Garmin reports it back.

Garmin needs to pick one methodology and not change it. Maybe one methodology per device if their capabilities demand it, but still, Once decided....TELL THE USER BASE!

5) GCBS asked about my VAHR step cals. That question doesn't make sense with the VAHR. I'm not sure whether it is just different than the fenix 3 or if you might misunderstand what the watch is truly doing (and Garmin might be misleading in explaining this).

-let's remember that "step calories" or "walking calories" is ALL info from the tracker after subtracting timed activities. I know that my VAHR records sufficient information that I am confident aboutthat definition and that means there is no correlation whatsoever between number of steps and step calories between any 2 given days.

- One problem with trying to see a cals/5000 step metric be right is that Garmin Connect also sucks recording total steps on the day. Why? because it includes steps in timed activities in the day total, which is beyond stupid unless the timed activity is itself a walk. If on a given day you time an activity (like a run) and it totalled 1500 steps and GC shows 2000 total steps, realize that means you only took 500 steps outside of your run. This is so confusing.

-at 12:51 am this morning I show 29 steps...and 122 calories burned. Obviously that is calculating based on more than just taking a step.

-Full day yesterday: 1621 step calories. 2578 total steps. One recorded activity for 441 cal additional, no steps for that since I was rowing,

-Day before: 1827 step calories (must be wrong- too high!) 2671 total steps. 3 bike rides recorded for 879 cals more. it looks like a few hundred steps were recorded during my bike ride (perhaps I walked it across an intersection or something?)

-Day before that is very confusing: 805 step cals. 3974 total steps. remember though that steps are double counted and on this day I had some activities recorded as runs (they weren't runs but that is irrelevant). This means steps outside recorded activites are far below that 3974 number, but not disclosed.

Time of Day ActCal Duration Distance Activity Type Activity Name
10:26 AM 26 3:26.8 0.00 km Indoor Cycling Indoor Cycling
10:30 AM 709 1:28:05 3.14 km Running Brighton Running
12:03 PM 154 24:18 1.70 km Running Brighton Running
12:28 PM 65 10:11 0.00 km Other Other
Daily Steps 805 -- 1.08 km -- --
Totals 1,759 2:06:01 5.91 km -- --

-June 14. As of this date my watch was using BMR as true BMR, not yet bumped up.
705 step cals, 2690 steps (though at least 1000 of those are measured during the strength activity, so double counted)

Time of Day Active Calories Duration Distance Activity Type Activity Name
6:04 PM 69 7:52.7 0.00 km Indoor Cycling Indoor Cycling
6:12 PM 838 2:05:58 6.55 km Strength Training Oly Class
8:19 PM 292 52:50 0.00 km Other Recovery time
Daily Steps 705 -- 0.85 km -- --
Totals 1,904 3:06:41 7.40 km -- --

-And lastly, June 16, first day of BMR reported +20%. I did almost the same things all day as on June 14:
1486 step cals, 3367 steps,

Time of Day Active Calories Duration Distance Activity Type Activity Name
6:04 PM 1,036 2:31:32 0.00 km Strength Training Strength Training
Daily Steps 1,486 -- 1.57 km -- --
Totals 2,522 2:31:32 1.57 km -- --

The link you provided does not explain the methodology used to calculate calories by stride length but I will agree that the amount for 21 miles seems reasonable. However, I disagree that 900 for the rest of the day is reasonable. Based on my experiences with using various activity trackers over the last 3 years and tracking calories it definitely should be higher, at least for me. Obviously we are all experiments of one so for some that might be accurate and I may be an outlier. Like you said the best solution would be for Garmin to allow us to choose how we want the calorie math done.

I completely disagree with you on the MFP issue though. Maybe it's different for the device you use since it has OHR but for the Fenix 3 what happens is timed activities get pushed to MFP correctly, but since the Fenix 3 is underestimating non timed activity calories (probably since it has no continuous heart rate and only has the accelerometer data to rely on) it also sends a huge negative adjustment to MFP that cancels out a lot of the exercise calories. Like I said, I think the problem is different for you since you have OHR on your unit.

Ultimately, the best would be for Garmin to follow the example of some of its competitors and allow the user to select how their calorie goal is calculated or at the very least clearly explain how they are doing it, for both HR and non-HR units so we know and can work around it if needed.