Determining whether exercise calories are accurate (Fitbit)

Good morning folks,

I'm trying to get a handle on just how accurate my logging and measuring is, so that I can better understand my progress (or lack of it, should that transpire).

Anyway, I've established that (for me) my Charge HR seems bang on accurate for HRM. Good, that's a big plus.

However, I'm not so sure about the accuracy of the calculations. For example, yesterday I set out to push myself a bit - 7K run in the morning, 4K walk before work, 6K walk at lunch, 3K walk home followed by an 7K walk in the evening. This amounted to 31.5K steps, total climb of 139 floors and a total burn (based on my height/weight) of 4109 KCal for the day.

During this time the 'adjustment' from Fitbit to MFP was 1903 KCal - does that seem reasonable for this level of activity? Or is it exaggerating?

On less active days I've been taken to eating most if not all of the exercise calories, but as the numbers increase I get very nervous about doing so. Yesterday I ate 2119 KCal (according to my fairly accurate logging & measurement - I'd expect an error of +/- 5% on this) so I was left with a net deficit of 1449 KCal for the day (my target being 1660 KCal). Which, if this is in any way accurate, means that I got by on a net 200 KCal. Unlikely! But that was the result of eating back some (not all) of those exercise calories....

So you can see why I'm a bit skeptical. Or am I about to crash any time now from a complete lack of energy? I'm mildly fatigued, but otherwise fine today (apart from slightly sore toes!).

FWIW in the previous week I logged ~150K steps (121 Km), ate 25K KCal, and had a balance of -3447 KCal over the month (should be ~0.5Kg-ish reduction). And saw no appreciable weight change. Not that that bothers me overly much, but I am curious.

Sorry for the long post!!!

Replies

  • zyxst
    zyxst Posts: 9,149 Member
    Are you losing weight at the rate you expect? Like if you have MFP set to 1# per week, is your average loss #1 pound? Average as in taking a month or 6 weeks of data and seeing if it works out to your predicted loss rate.

    I had a Fitbit Zip that gave me very high burn. If I'd eaten those calories, I'd have been gaining weight instead of losing. I knew it was off because I'd only had 1 day of clocking a 3,000 burn and I'd been walking outside on hills for 6 hours & weighed 280#.
  • eIIekay
    eIIekay Posts: 166 Member
    I'll try to help...on my vivofit, I noticed that it displays calories burned and will rise throughout the day. It is telling me that this is how many calories (resting and active) that I have burned so far in the day. I think your 4109 is telling you that.

    Then, the 1903 adjustment that you got is the actual calories burned from exercise. If you add this to the calories you need to maintain your weight, should be 4109 for the day.

    If you are trying to lose weight, don't eat the calories your fitbit tells you that you've burned. You will maintain.
    Does this make sense?
  • _piaffe
    _piaffe Posts: 163 Member
    Funny, I was wondering the same thing - accuracy of calories earned. I religiously track my timed workouts using the stopwatch function on my Charge HR. Here are recent comparisons between what Fitbit calculated the isolated activity to be, vs. what MFP would calculate for the same activity:
    • Running/jogging at a 6.2 min per km pace for 35 mins: 378 Fitbit calories vs. 393 MFP calories
    • HIIT elliptical for 30 mins: avg 144 bpm HR = 320 Fitbit calories vs. 303 MFP calories
    • Hill incline treadmill for 25 mins at 4.0 mph, 13% incline: avg 162 bpm HR = 262 Fitbit calories vs. "very very brisk walking" 177 MFP calories
    More or less equivalent on the jogging and elliptical, but I think my Fitbit is closer to the real burn on the incline work. MFP doesn't allow me to add the significant incline variable.
  • tahxirez
    tahxirez Posts: 270 Member
    All you can do is eat what MFP gives you and see how you react to it. For me, my Fitbit calorie adjustment is dead on. I have lost and maintained exactly as expected per what MFP and Fitbit give me (almost 2 years now.) I do not pay attention or log calories in Fitbit however. Try it for 6 weeks or so, see if you're losing at the rate you'd expect. There's really no question you're at a deficit so at this point its just figuring out how much of one.
  • Jeniccm
    Jeniccm Posts: 44 Member
    You walked 27km - that's 16 miles (more than a half marathon!). I'm not sure of your stats but to be honest 1900 calories for walking that distance (plus your additional calories for just living and moving around) doesn't sound that far off. Remember Fitbit is giving you a base BMR and the calories it gives you is anything over and above that.

    Nice job!
  • Jeniccm
    Jeniccm Posts: 44 Member
    I want to add that I did track Fitbit for a few months to see how accurate it was vs my weight loss and for me it actually slightly underestimated.
  • Samone_nah
    Samone_nah Posts: 1 Member
    The total calories your fitbit is registering is the total amount you have logged all day, both at active and just your normal day. I know with my apple watch, it tells me, based on my HR and how much I have moved around during the day, the total amount of Calories I have burnt. When I workout, I turn on the activity monitor. That logs the calories burnt during that time period as a 'workout' which I then put into MFP as exercise. My total calories for the day is always heaps more than what I ate, which is what I want. I want to be burning off the fat my body has stored. The exercise component of calories is what MFP is taking off your total approx value of what you should be eating. If you are wanting to lose weight, you still need to be burning, in total for the whole day more than what you are consuming. Hopefully this makes sense.
  • Sued0nim
    Sued0nim Posts: 17,456 Member
    edited August 2016
    It's about biofeedback over time ...how much weight you lose against how much food you logged give you an idea of your actual TDEE over time

    I've had a Fitbit for 2 years ...10k steps = just under 5 miles = 400-500 extra calories over MFP sedentary (1750), giving me a TDEE of 2150-2250

    Last Monday I walked 2 lots of about 10k steps (one in morning, one in evening) so just under 10 miles and got an 827 adjustment on sedentary ..giving me a TDEE of 2750

    My average TDEE is 2200-2400

    So you've done 31k steps ...for me that would be probably around 1200 - 1500 calories over sedentary

    The difference is determined by intensity
  • Maxematics
    Maxematics Posts: 2,287 Member
    edited August 2016
    I have a Charge HR, so I'd like to weigh in.

    "However, I'm not so sure about the accuracy of the calculations. For example, yesterday I set out to push myself a bit - 7K run in the morning, 4K walk before work, 6K walk at lunch, 3K walk home followed by an 7K walk in the evening. This amounted to 31.5K steps, total climb of 139 floors and a total burn (based on my height/weight) of 4109 KCal for the day."

    That's a lot of activity; a 4109 calorie TDEE for the day does not seem unreasonable at all. I'm 5'3" and 108 pounds and with 20K steps in a day, I get anywhere between 2000 and 2400 calories to eat depending in the level of intensity in my day.

    "During this time the 'adjustment' from Fitbit to MFP was 1903 KCal - does that seem reasonable for this level of activity? Or is it exaggerating?"

    If you are set to Sedentary and have negative adjustments enabled, then yes this is reasonable. Compare it to your usual daily adjustments. More activity = more food needed. Yesterday I went on a run for an hour, walked around a good portion of my day, and ended the day with 23K steps. My Fitbit adjustment was 904 exercise calories and I ate all of them. I've been eating my calories for months and I haven't gained weight.

    "Yesterday I ate 2119 KCal (according to my fairly accurate logging & measurement - I'd expect an error of +/- 5% on this) so I was left with a net deficit of 1449 KCal for the day (my target being 1660 KCal). Which, if this is in any way accurate, means that I got by on a net 200 KCal. Unlikely! But that was the result of eating back some (not all) of those exercise calories....

    So you can see why I'm a bit skeptical. Or am I about to crash any time now from a complete lack of energy? I'm mildly fatigued, but otherwise fine today (apart from slightly sore toes!)."


    I went back in my data to a time where I walked around as many steps as you in a day. This was all just walking, no runs. I walked for 2.5 hours into the city, did shopping, and walked 2.5 hours back home so I was pretty much on my feet from 10am until 5pm. A bit over 35K steps and an adjustment of 878 exercise calories. I ate 1688 calories that day and had 1080 remaining. I remember that day I was just so tired from all the walking and the heat outside that I didn't have the appetite to eat my calories back. In fact, whenever I do that, it happens; my appetite doesn't catch up for a while. Your body doesn't immediately input/output like that.

    "FWIW in the previous week I logged ~150K steps (121 Km), ate 25K KCal, and had a balance of -3447 KCal over the month (should be ~0.5Kg-ish reduction). And saw no appreciable weight change. Not that that bothers me overly much, but I am curious."

    Again, the body will not give you immediate input/output. It may be weeks until you see the scale jump down around a pound due to that 3447 calorie deficit. A pound loss is easily disguised by water retention for various reasons. The best way to assess the accuracy of your Fitbit is to take a 30 day average of your CICO data versus your actual weight loss.

    ETA: I don't know if I'm linking this properly or explaining it properly but @PAV8888 posted this spreadsheet a while back that's useful for Fitbit data: https://docs.google.com/spreadsheets/d/1VDmqNpLPu7sbQSochUJNXdp2F7AN15AGgkvS3zLw1GU/

    I'm trying to maintain my weight and noticed I've still been slightly dropping and had days once in a while where I needed to eat around 1000 calories over my TDEE just to feel satiated. I entered my data all the way from May and it's telling me that my TDEE is being underestimated by 13% according to my trending weight and 15.5% according to my scale weight. It explains much of what's been going on with my body lately.
  • jtcedinburgh
    jtcedinburgh Posts: 117 Member
    Thanks folks, that helps. BTW, I have my activity level set to sedentary (as, if I didn't do the exercise, my desk job would fall into that category) and I have negative adjustments activated. Also, I made a small mistake - my balance of -3447 KCal was over the week, not the month.