Does Fitbit One estimate calories accurately?

The basic answer is yes, at least if your main exercise is walking.

Tools of the trade

Tools of the trade

Recently, after a health issue reduced me to only walking, I purchased a Fitbit One to help me manage my weight. One of the gizmo’s selling points was its much heralded website, where I could keep track of calories consumed as well as those expended. From December 13 to January 21, I meticulously recorded everything I ate (literally counting each peanut) and wore the Fitbit at all times – even on my wrist as I slept. I also recorded my weight each morning before eating. Now I’ve taken this data and crunched some numbers.

My method focuses on long-term changes, rather than daily variations. Temporary issues such as water-retention might cause a short-term spike or dip in weight but should have less importance over many days. Unfortunately, on two days I had incomplete data. The record for Dec. 31 lacks a large number of calories consumed and expended, and the data for Jan. 8 lack a few stairs climbed and steps taken, but it’s not a major problem. So I’ve compared the balance of calories to changes in weight for Dec. 12-30, Jan. 1-21, Jan. 1-7, and Jan. 9-21.

In each period, the result is similar: Fitbit estimated that I burned moderately more calories than I consumed, and my weight declined moderately. That is, Fitbit seems to work.

  • Dec. 12-30
    • Calories in – calories out = 42,577 – 46,504 = -3,927 calories
    • Weight change = -1.6 pounds
  • Jan. 1-21
    • Calories in – calories out = 51,046 – 52,651 = -1,605
    • Weight change = -1.4 pounds

And, just in case Jan. 8’s missing calories were consequential:

  • Jan. 1-7
    • Calories in – calories out = 16,801 – 18,214 = -1,413
    • Weight change = -1.6 pounds
  • Jan. 9-21
    • Calories in – calories out = 31,399 – 31,942 = -543
    • Weight change = -.2

So, I repeat: the Fitbit One’s estimate of my calories expended, when compared to my estimate of my calories consumed, correlates pretty well with changes in my weight. In short, it works for me.

Here are some caveats:

  • I barely performed any other sort of activity, such as running, cycling, or weightlifting. Fitbit’s estimates might be less accurate for those exercises.
  • Other Fitbit devices work differently, so their estimates probably will differ as well.
  • While the direction of change always matched, the ratio of caloric deficit to pounds lost differed for each period. Probably this was due to Fitbit’s inaccuracy in combination with the vagaries of digestion. Here’s the data:
    • Dec. 13-30: 2454 calories/lb.
    • Jan. 1-7: 883 calories/lb.
    • Jan. 9-21: 2715 calories/lb.
    • Combining these three periods, the ‘weighted’ average is: 2254 calories/lb.
    • That is, for each pound lost, I burned 2254 more calories than I consumed. Essentially, if I didn’t consume any calories for a day but maintained my level of activity, I’d lose a little more than a pound.

2 responses to “Does Fitbit One estimate calories accurately?

  1. Wow, so technical, impressive! What I’d like is some way to track calories “automatically.” I am challenged to take the time to write stuff down and then figure out how many calories is each peanut…

    Liked by 1 person

    • Yeah, the data ended when my commitment ended. But, after a month of peanut counting, I have a pretty good idea of how much I can eat if I don’t want to look like a marshmallow.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s