How does Fitbit determine what you get for excercise calories?

Options
I know how Fitbit basically works and have always found it accurate for me; however, my charge hr bit the dust and I got the charge hr 2. I have had it about a week and a half. So far I love it, but I was wondering does it go by your heart rate to determine what you burn? I have mfp set to sedentary. I am actually in between sedentary and lightly active outside of excercise but my activity differs from day to day so I had rather see the big numbers of sedentary than to have some taken away overnight.

On Sunday I got over 11,000 steps when I took about an hour or so long light - moderate walk. It gave me an adjustment to mfp of about 500 calories... that is about what my other Fitbit would give me. However yesterday I worked at a yardsale all day from 5:00 am 4:00 sitting very little. I was loading and un-loading, helping people, running back and forth hanging clothes... at the end of the day I was absolutely dead on my feet. I only got 10500 steps but it gave me an adjustment to mfp of +1200 calories. I know I worked hard yesterday but it was a normal paced walking not even as fast as my hour long walk on Sunday. However, I was on my feet all day. Please help me understand thenig adjustment . Was it a fluke? Could it be accurate because I was busy all day even though I had less steps than Sunday?

Replies

  • yirara
    yirara Posts: 9,444 Member
    Options
    I find my charge 2 grossly overestimates my calories. I don't link it to MFP but just use it for fun. If you go to the website you can see what Fitbit allocates as BMR to you: just find your lowest heartrate on that graph and see what the burn per minute is. That times 60*24 gives you the BMR that fitbit assumes. Everything else is somehow based on heartrate I would guess.

    One other thing: Fitbit also doesn't use BMI plus exercise such as MFP does but it displays gross calories for workouts. So if for example you record a run and it comes back at 500 calories for one hour, then you need to take the BMR calories for one hour off this number.

    But honestly, I find those numbers ok to huge, depending on what I do. I ran 5km yesterday and fitbit gave me 311calories for it. Even if I substract my BMI from it it would still be about 275. I would give myself about 266 for it, which is fine.

    On the other hand I ran 12km the other night. Fitbit gave me 1016kcal gross/930-ish net. I would only give myself 640 max.
  • MommaGem2017
    MommaGem2017 Posts: 405 Member
    Options
    Yes, the FitBit uses your resting heart rate vs. your active heart rate to determine the calories burned. So if you casually stroll around the house for 10k steps you'll burn fewer calories than if you briskly walked those 10k steps. However, I have read that it can take several weeks for a new FitBit to "get to know you" and your actual average resting heart rate. Your calories burns may start to get more consistent over time.
  • yirara
    yirara Posts: 9,444 Member
    Options
    Yes, the FitBit uses your resting heart rate vs. your active heart rate to determine the calories burned. So if you casually stroll around the house for 10k steps you'll burn fewer calories than if you briskly walked those 10k steps. However, I have read that it can take several weeks for a new FitBit to "get to know you" and your actual average resting heart rate. Your calories burns may start to get more consistent over time.

    Interesting! Mind is still calibrating I think. But I only have it for just over two weeks now. My resting HR started out fairly high, and now it's come down to 52. A few more weeks and Fitbit will declare me dead :D

    I do hope that this has no influence on my workout calories as my maxHR is massive, around 205/210, and it does show on the calorie count even for slow walks I think. My VOmax estimate is completely rubbish as well. Hey, I'm sure I could run a half marathon (training for one just now), but Fitbit only puts me in the somewhat fit goup.
  • MommaGem2017
    MommaGem2017 Posts: 405 Member
    Options
    Earlier in the year I took a break from the FitBit. I put it back on about 3 months ago and it took a solid month to settle into my resting heart rate. It's so cool to look back at the history and see the steady decline over 30 days, then 2 more months when it settled around 55-60 and the line stayed pretty horizontal.

    Wow, if your resting is 52 and your max could be 200 - you must be getting an incredible calorie burn. What does your FitBit average as your total daily burn? I have heard some people talk about inaccuracies when it comes to irregular heart rates, so you may have to adjust the settings on your FitBit so they match your experience more. (I have no idea how to do that, but the FitBit users group has talked about it)
  • yirara
    yirara Posts: 9,444 Member
    Options
    Earlier in the year I took a break from the FitBit. I put it back on about 3 months ago and it took a solid month to settle into my resting heart rate. It's so cool to look back at the history and see the steady decline over 30 days, then 2 more months when it settled around 55-60 and the line stayed pretty horizontal.

    Wow, if your resting is 52 and your max could be 200 - you must be getting an incredible calorie burn. What does your FitBit average as your total daily burn? I have heard some people talk about inaccuracies when it comes to irregular heart rates, so you may have to adjust the settings on your FitBit so they match your experience more. (I have no idea how to do that, but the FitBit users group has talked about it)

    Yes, I get fairly high calorie burns, and it sucks as my calorie burns are not in fact that high. Yesterday was a pretty sedentary day with just a bit of walking (about 4km, plus a bit at work) and fitbit gave me nearly 2000kcal. My maintenance cals are at around 1750. Walking longer distances is waaay off, and running depends. For the shorter runs it's not even that bad, but the longer they get the more off they become. If heartrate would really correlate with calorieburn then I could eat constantly :D
  • holderh1
    holderh1 Posts: 41 Member
    Options
    CyberTone wrote: »
    Since you are a previous Fitbit user, Fitbit will use your running average variables already refined from your history. The estimate with the Fitbit model replacement will be fairly accurate, because of your Fitbit history. A new Fitbit user does not have a history, so the variables used start out as general population estimates for a person of similar stats; as Fitbit collects a history of a new person's movements, it uses the past data to refine the calculations - in effect the Fitbit "learns" about your daily habits.

    The introduction of the Fitbit model with an HR monitor will help to refine the calculations for the type of day you had working at the yard sale - i.e. an active day on your feet with mostly incidental walking instead of a long, slow purposeful walk. The accelerometer and HR monitors work in conjunction to estimate how intense your activity was; loading/unloading burns more Calories than just standing or walking, short bursts of "running around" burns more Calories than slow walking.

    For comparison, I am 55, 5'8", 150 lbs, in maintenance, have a Fitbit Charge 2, and have MFP set at Sedentary. The last time I got 10,717 steps, I earned 1031 extra Calories as my Fitbit Calorie adjustment. Sedentary for me is about 2500 steps to get a zero Fitbit Calorie adjustment.

    I always eat back 100% of my earned Calories. I ate back 90% of them when I was losing a few pounds and had no problem netting 1700 Cals per day to lose those pounds. I do use a food scale to weigh all solids, measuring cups and spoons to measure liquids, and verify all food items I log through outside web sources and Nutrition Facts labels. I trust that my Calorie Intake is pretty accurate, and I trust my Calorie Output is pretty accurate using the Fitbit.

    Thank you ! This was very helpful for my particular question.

    I know my body felt exhausted yesterday but I was amazed at the calorie adjustment. When I looked at yesterday's heart rate graph it showed that I was in fat burn mode for more than 5 hours... my normal is 1-2 hours a day. Do that sounds like it was accurate yesterday. What do you think?
  • holderh1
    holderh1 Posts: 41 Member
    Options
    yirara wrote: »
    I find my charge 2 grossly overestimates my calories. I don't link it to MFP but just use it for fun. If you go to the website you can see what Fitbit allocates as BMR to you: just find your lowest heartrate on that graph and see what the burn per minute is. That times 60*24 gives you the BMR that fitbit assumes. Everything else is somehow based on heartrate I would guess.

    One other thing: Fitbit also doesn't use BMI plus exercise such as MFP does but it displays gross calories for workouts. So if for example you record a run and it comes back at 500 calories for one hour, then you need to take the BMR calories for one hour off this number.

    But honestly, I find those numbers ok to huge, depending on what I do. I ran 5km yesterday and fitbit gave me 311calories for it. Even if I substract my BMI from it it would still be about 275. I would give myself about 266 for it, which is fine.

    On the other hand I ran 12km the other night. Fitbit gave me 1016kcal gross/930-ish net. I would only give myself 640 max.

    Thank you! I will take a closer look at the graphs and see what I can come up with.

  • holderh1
    holderh1 Posts: 41 Member
    Options
    Yes, the FitBit uses your resting heart rate vs. your active heart rate to determine the calories burned. So if you casually stroll around the house for 10k steps you'll burn fewer calories than if you briskly walked those 10k steps. However, I have read that it can take several weeks for a new FitBit to "get to know you" and your actual average resting heart rate. Your calories burns may start to get more consistent over time.


    Thank you! This is what I was thinking but I wasn't 100% sure.
  • CyberTone
    CyberTone Posts: 7,337 Member
    edited August 2017
    Options
    holderh1 wrote: »
    CyberTone wrote: »
    Since you are a previous Fitbit user, Fitbit will use your running average variables already refined from your history. The estimate with the Fitbit model replacement will be fairly accurate, because of your Fitbit history. A new Fitbit user does not have a history, so the variables used start out as general population estimates for a person of similar stats; as Fitbit collects a history of a new person's movements, it uses the past data to refine the calculations - in effect the Fitbit "learns" about your daily habits.

    The introduction of the Fitbit model with an HR monitor will help to refine the calculations for the type of day you had working at the yard sale - i.e. an active day on your feet with mostly incidental walking instead of a long, slow purposeful walk. The accelerometer and HR monitors work in conjunction to estimate how intense your activity was; loading/unloading burns more Calories than just standing or walking, short bursts of "running around" burns more Calories than slow walking.

    For comparison, I am 55, 5'8", 150 lbs, in maintenance, have a Fitbit Charge 2, and have MFP set at Sedentary. The last time I got 10,717 steps, I earned 1031 extra Calories as my Fitbit Calorie adjustment. Sedentary for me is about 2500 steps to get a zero Fitbit Calorie adjustment.

    I always eat back 100% of my earned Calories. I ate back 90% of them when I was losing a few pounds and had no problem netting 1700 Cals per day to lose those pounds. I do use a food scale to weigh all solids, measuring cups and spoons to measure liquids, and verify all food items I log through outside web sources and Nutrition Facts labels. I trust that my Calorie Intake is pretty accurate, and I trust my Calorie Output is pretty accurate using the Fitbit.

    Thank you ! This was very helpful for my particular question.

    I know my body felt exhausted yesterday but I was amazed at the calorie adjustment. When I looked at yesterday's heart rate graph it showed that I was in fat burn mode for more than 5 hours... my normal is 1-2 hours a day. Do that sounds like it was accurate yesterday. What do you think?

    Since your activity was step-based, it is most likely providing a good estimate. Fitbit will not be quite as good for non-step based activity, such as horseback riding, swimming, cycling, or resistance/strength training.

    My activity is mostly step-based, so I can trust my Fitbit. I wear it on my left wrist (non-dominant) and have it set in Fitbit for the non-dominant hand. The only time I remove the Fitbit is when I am playing/practicing piano for more than a few minutes, so that it does not count the movement as steps. I usually will switch the Fitbit from my wrist to my ankle when I am practicing the piano.
  • NorthCascades
    NorthCascades Posts: 10,970 Member
    Options
    Earlier in the year I took a break from the FitBit. I put it back on about 3 months ago and it took a solid month to settle into my resting heart rate. It's so cool to look back at the history and see the steady decline over 30 days, then 2 more months when it settled around 55-60 and the line stayed pretty horizontal.

    It took your Fitbit a month to be able to measure your heart rate???
  • MommaGem2017
    MommaGem2017 Posts: 405 Member
    Options
    It took a month to figure my average Resting Heart Rate
  • yirara
    yirara Posts: 9,444 Member
    Options
    Ok, some more fitbit data: ran 14km tonight in 1hrs, 48 mins (yes, I'm that slow). Fitbit gave me 1125kcal gross calories. That's 1027kcal net. Uploading to MFP from my running app gave me a fairly similar number. Using lbs*0.68*miles I'd only give myself 746kcal net. That's a massive difference.
  • yirara
    yirara Posts: 9,444 Member
    Options
    Right, some more rubbish data. I walked today for 15km and fitbit gave me 1166 gross calories. As a reminder, yesterday I got 1125 gross calories for running 14km. Running burns more than twice as much calories as walking thus there is something very, VERY off here.

    Using weight * miles * 0.68 for running and weight * miles * 0.3 for walking I'd only get about 751 and 344 net calories respectively.