How much percentage should I take away from Fitbit Charger Two calorie Burn Workouts?

jenniator
jenniator Posts: 475 Member
edited May 2018 in Fitness and Exercise
Hey everyone,

Two days ago I bought the Fitbit charger 2 which I absolutely love. It's the first Fitness tracker I've ever used and bought, so I'm still learning about it. I've always wanted one, so I've been having a lot of fun. The main reason I bought it was to give me a closer estimate of how many calories I'm burning though out my workout on machines, walking, and group lessons. Yesterday I did a BodyCombat class and it says I burned 621 calories in 55 minutes.

I'm 227 pounds and my height is 5'4, so it might be accurate. Later today I have a Zumba class and a Bodypump class on Wednesday. Some days I also use the elliptical, exercise bike, or rowing machine. During the weekends, I walk around. So it would be nice to have a formula that would not over estimate my calories no matter what type of activity I'm doing. I would rather slightly underestimate than over estimate them my calorie burn.

I know that Fitness trackers are not always 100% accurate and that they can over estimate calories. So my question is, by how much percentage should I take away from my total calorie burn to get a more accurate estimate and to make sure that I'm not over estimating my calorie burn? Should I take off 10%, 15%, 20%, or another percentage away from every workout?

Replies

  • MeanderingMammal
    MeanderingMammal Posts: 7,866 Member
    jenniator wrote: »
    Yesterday I did a BodyCombat class and it says I burned 621 calories in 55 minutes.

    You might have done about 200, possibly.

    Any wrist based tracker is meaningless for those types of activity as it's largely measuring movement and assuming that you're walking. So lots of false positives in that type of class.

    I'd also note that the HR measure is meaningless as well as you're not in a steady state, aerobic, situation.
    I know that Fitness trackers are not always 100% accurate and that they can over estimate calories.

    Possibly a question of semantics, and I'm a control systems engineer by training, but when people say not 100% accurate, that suggests c90% or better. In the situation you described when you asked for a recommendation a FitBit is likely to have an error of ± 75-80%.
  • biscuitnow
    biscuitnow Posts: 141 Member
    edited May 2018
    While the Charge 2 can be used to track different types of exercises (I use it and I'm very happy with it), I would be very careful about its calorie burn estimations. Like most trackers, gym machines etc., it tends to overestimate. I always halve the numbers (and it works really well for me).
  • jenniator
    jenniator Posts: 475 Member
    edited May 2018
    Thanks for the answers everyone. Wow I didn't realize how inaccurate they could be. I always heard that they could be off from 10-20%! I guess I will have to be more careful.
  • shadow2soul
    shadow2soul Posts: 7,692 Member
    edited May 2018
    jenniator wrote: »
    Thanks for the answers everyone. Wow I didn't realize how inaccurate they could be. I always heard that they could be off from 10-20%! I guess I will have to be more careful.
    It actually really depends on the individual. They don’t overestimate for everyone. For me:

    Fitbit’s without HR (Zip, Flex, & Flex2) underestimate my total daily energy expenditure by an average of 250 calories a day.

    Fitbit’s with HR (Surge, Blaze, & Ionic) are close enough at estimating my total daily energy expenditure that I lose/maintain/gain as I would expect based on my intake vs Fitbit calories burned.

    So for me there is no need to subtract calories from Fitbit’s burn. I do use a food scale and I try to be super careful about the entries I use from the database to make sure my logging is accurate as I can get it.

    Give it 30 Days and your Fitbit profile will have your 30 day average burn. Then do the math for your average intake (used to be on the Fitbit profile, but a glitch they have yet to fix means we have to calculate it ourselves).

    30 Day Fitbit avg - 30 Day intake = 30 Day Avg Deficit

    30 Day Avg Deficit * 30 = Total Deficit for the 30 days

    Total Deficit /3500 = Expected weight loss based on Fitbit calories burned

    Compare the Expected number to your actual loss.

  • jjpptt2
    jjpptt2 Posts: 5,650 Member
    I'm sure you're going to get the usual "numbers from trackers for those types of activities are useless" response... but I don't think that's particularly helpful. Those numbers are unreliable, yes... but that doesn't mean they are necessarily unrealistic or inaccurate.

    Most people suggest using 50% of your estimated burn as a starting point, which most of the time is as reasonable as anything. But ultimately, it's just a starting point for your own estimating/logging/tracking. Once you have a few weeks' worth of data, you can compare your actual results to your expected results. THEN you'll know how accurate you are or aren't being, and can tweak things as needed.
  • MeanderingMammal
    MeanderingMammal Posts: 7,866 Member
    jjpptt2 wrote: »
    I'm sure you're going to get the usual "numbers from trackers for those types of activities are useless" response... but I don't think that's particularly helpful. Those numbers are unreliable, yes... but that doesn't mean they are necessarily unrealistic or inaccurate.

    The observation I'd make is that this approach would be reasonable if the tracker is likely to be consistent. Given the activities that the originator has identified, a FitBit is both inaccurate, and inconsistent.
  • floppybackend
    floppybackend Posts: 52 Member
    Michael Mosley BBC Fitness/diet Dr said they are at least 25 - 50% inaccurate reason why people eat more and don't lose weight.
  • jjpptt2
    jjpptt2 Posts: 5,650 Member
    edited May 2018
    MeanderingMammal wrote: »
    jjpptt2 wrote: »
    I'm sure you're going to get the usual "numbers from trackers for those types of activities are useless" response... but I don't think that's particularly helpful. Those numbers are unreliable, yes... but that doesn't mean they are necessarily unrealistic or inaccurate.

    The observation I'd make is that this approach would be reasonable if the tracker is likely to be consistent. Given the activities that the originator has identified, a FitBit is both inaccurate, and inconsistent.

    meh.

    It's not like there are monkeys inside the tracker throwing darts at a miniature dart board to come up with a number. The number is derived from a formula based on averages and assumptions and certain data. It's as reasonable a starting point as anything is.

    Don't get me wrong... I'm not trying to say that these estimates are right or reliable or accurate or anything else. I just don't think the opposing blanket statement of "they are worthless" is entirely appropriate either.

    I could be missing something though... or giving trackers too much benefit of the doubt due to my own experiences with them.
  • MeanderingMammal
    MeanderingMammal Posts: 7,866 Member
    edited May 2018
    jjpptt2 wrote: »
    The observation I'd make is that this approach would be reasonable if the tracker is likely to be consistent. Given the activities that the originator has identified, a FitBit is both inaccurate, and inconsistent.

    meh.

    It's not like there are monkeys inside the tracker throwing darts at a miniature dart board to come up with a number. The number is derived from a formula based on averages and assumptions and certain data. It's as reasonable a starting point as anything is.

    There are accelorometers in them that measure movement in three dimensions, and extrapolate that to give an approximate umber of steps. They then work on the basis of an average step length to give an approximate distance travelled, and extrapolate a calorie approximation from there.

    As the FitBit is the most basic of processors the use of HR as a corroboration is meaningless, they don't correlate that data with a particularly sophisticated algorithm. Garmin, for exmple, uses the FirstBeat algorithms to try to accommodate the error, by correlating direction and rate of movement with intra-beat deviation

    So if you're walking, that's fine.

    Zumba, for example, will assume a reasonable distance travelled, when what you've been doing is waving the arms around. Bodypump and BodyCombat are the same. Arm movements inject error.

    Might as well roll a couple of dice, then apply a factor based on the moon phase on the day your dog was born.

    They have a place, but not for the types of classes that the originator is taking.

Categories