Does fitbit really work?

CSARdiver

I started with a Flex and moved onto a Flex 2 - both essentially higher tech pedometers. In the beginning it was a solid motivator. It reminds me if I don't get in 250 steps in an hour, tracks my steps, and tracks sleep quality.

I switch over to a Polar H7 for my endurance workouts.

If you're into gamification I'm assuming this will have a solid positive impact. It sets up a benchmark of 10k steps/day and you can adjust accordingly. For me it pushes me to continually improve.

PAV8888

tbright1965 wrote: »

Let's keep it confined to one number, which is what I was talking about, the expenditure.

(Edited to add, I do realize I also posted my food consumption of about 2200 calories Sunday, so perhaps that causes some confusion. Just wasn't hungry on Sunday. It was Monday where my body wanted to eat all the foods.)

My logging is pretty good, but not really relevant to the idea that the Fitbit gets an accurate calorie number.

All the Fitbit knows is the GPS data it picks up, the HR data it picks up and how much I weigh, my age and gender. So it came up with ~4800 calories in 3.6 hours of work averaging just under 15MPH.

It doesn't know anything about wind. I do believe it has terrain data from comparing the GPS data to geodata out there.

So if I rounded up my speed and my weight to be 100kg (I'm a few kg below) I'd only get to 10x100x3.6 or 3600 calories, or about 3/4rs of the value presented.

I believe my sedentary burn is around the 100 calories/hour, so even adding on another 350 for BMR, I'm still below 4k, but certainly closer.

I do agree, to get a better reading, a power meter is needed.

What I've found works for me is to only eat back 1/2 of what MMR and Fitbit suggest are my burns. That has kept me on a 1-2 pound per week weight loss.

I do weigh calorie dense foods on my food scale. So if I have a omelette, I weigh the cheese,but not the veggies I put in it.

So far, for the past 7.5 months, that approach has allowed me to drop 50#

I simply caution people to be a little skeptical of the calorie burns offered up by such tools.

It would be unwise for me to take the 4800 calorie figure as permission to eat 4800 calories of cookies just because I rode my bike.

PAV8888 wrote: »

Whether you can trust your numbers to "eat" them or not depends on the interaction of:

--the quality of your food intake log
--the quality of your caloric expenditure log
--the quality of your weight change measurement
--how close to "average" you happen to be

All but the last are somewhat under one's control. You start by semi-trusting your numbers and you adjust based on results.

Cycling is a very well researched activity. A 15mph bike ride is a MET 10 activity. The higher the speed the more the MET value increases. Inclines and wind also have an effect. Some apps try to adjust for inclines and your bike's weight. (f.e. 16-19mph without drafting is about 12 MET)

A direct power meter would give you a better approximation. Strava will probably give you the most accurate estimation possible, short of being wired directly.

This means that a 100kg person engaged in that activity for 1 hour will burn approximately 1000 Calories (10 x 100 x 1)

Alternatively an 80kg person engaged in a MET 10 activity for 0.5 hours would burn ~400 Cal (10 x 80 x 0.5).

Whether YOU can eat those calories depends on the factors I outlined initially. However, this does not negate the fact that an average person engaged in that activity would, on average, burn this amount of calories.

To complicate things further "how close to the average person you are" takes a hit when you're a habituated athlete. You may personally have a 20% lesser expenditure for the same results as compared to, for example, me just because you're used to the exercise.

Also extremely long activities that are not an everyday occurrence probably do not require a complete eat-back if the goal is weight loss. HOWEVER, daily activities require much more careful balancing to avoid side effects.

Again I am referring to considerable activity. I note that MFP's very active setting is reachable with ~3 hours of walking a day.

It is one thing to go for a 4 hour bike ride one day. It is another thing to go for two of them every day! There was a full time bicycle messenger who also exercised on top posting a few days ago....

All this does not negate basic activity estimates based on population averages which is all that fitbit does. The only "magic" is that instead of you having to tell it I engaged in a 15 mph ride for 15 minutes, it sits there and guesses that you engaged in a 10mph ride for 5 minutes, and a 15mph ride for 5 minutes and a 20mph ride for 5 minutes, and produces a slightly different, and hopefully more precise and less biased by your own perceptions, average.

You mentioned earlier that you don't want to consider the full picture, only the expenditure. You can't. You can't judge the correctness of the expenditure directly. Only indirectly through the end results you get. Which are also influenced by everything else.

tbright1965

PAV8888 wrote: »

tbright1965 wrote: »

Let's keep it confined to one number, which is what I was talking about, the expenditure.

(Edited to add, I do realize I also posted my food consumption of about 2200 calories Sunday, so perhaps that causes some confusion. Just wasn't hungry on Sunday. It was Monday where my body wanted to eat all the foods.)

My logging is pretty good, but not really relevant to the idea that the Fitbit gets an accurate calorie number.

All the Fitbit knows is the GPS data it picks up, the HR data it picks up and how much I weigh, my age and gender. So it came up with ~4800 calories in 3.6 hours of work averaging just under 15MPH.

It doesn't know anything about wind. I do believe it has terrain data from comparing the GPS data to geodata out there.

So if I rounded up my speed and my weight to be 100kg (I'm a few kg below) I'd only get to 10x100x3.6 or 3600 calories, or about 3/4rs of the value presented.

I believe my sedentary burn is around the 100 calories/hour, so even adding on another 350 for BMR, I'm still below 4k, but certainly closer.

I do agree, to get a better reading, a power meter is needed.

What I've found works for me is to only eat back 1/2 of what MMR and Fitbit suggest are my burns. That has kept me on a 1-2 pound per week weight loss.

I do weigh calorie dense foods on my food scale. So if I have a omelette, I weigh the cheese,but not the veggies I put in it.

So far, for the past 7.5 months, that approach has allowed me to drop 50#

I simply caution people to be a little skeptical of the calorie burns offered up by such tools.

It would be unwise for me to take the 4800 calorie figure as permission to eat 4800 calories of cookies just because I rode my bike.

PAV8888 wrote: »

Whether you can trust your numbers to "eat" them or not depends on the interaction of:

--the quality of your food intake log
--the quality of your caloric expenditure log
--the quality of your weight change measurement
--how close to "average" you happen to be

All but the last are somewhat under one's control. You start by semi-trusting your numbers and you adjust based on results.

Cycling is a very well researched activity. A 15mph bike ride is a MET 10 activity. The higher the speed the more the MET value increases. Inclines and wind also have an effect. Some apps try to adjust for inclines and your bike's weight. (f.e. 16-19mph without drafting is about 12 MET)

A direct power meter would give you a better approximation. Strava will probably give you the most accurate estimation possible, short of being wired directly.

This means that a 100kg person engaged in that activity for 1 hour will burn approximately 1000 Calories (10 x 100 x 1)

Alternatively an 80kg person engaged in a MET 10 activity for 0.5 hours would burn ~400 Cal (10 x 80 x 0.5).

Whether YOU can eat those calories depends on the factors I outlined initially. However, this does not negate the fact that an average person engaged in that activity would, on average, burn this amount of calories.

To complicate things further "how close to the average person you are" takes a hit when you're a habituated athlete. You may personally have a 20% lesser expenditure for the same results as compared to, for example, me just because you're used to the exercise.

Also extremely long activities that are not an everyday occurrence probably do not require a complete eat-back if the goal is weight loss. HOWEVER, daily activities require much more careful balancing to avoid side effects.

Again I am referring to considerable activity. I note that MFP's very active setting is reachable with ~3 hours of walking a day.

It is one thing to go for a 4 hour bike ride one day. It is another thing to go for two of them every day! There was a full time bicycle messenger who also exercised on top posting a few days ago....

All this does not negate basic activity estimates based on population averages which is all that fitbit does. The only "magic" is that instead of you having to tell it I engaged in a 15 mph ride for 15 minutes, it sits there and guesses that you engaged in a 10mph ride for 5 minutes, and a 15mph ride for 5 minutes and a 20mph ride for 5 minutes, and produces a slightly different, and hopefully more precise and less biased by your own perceptions, average.

You mentioned earlier that you don't want to consider the full picture, only the expenditure. You can't. You can't judge the correctness of the expenditure directly. Only indirectly through the end results you get. Which are also influenced by everything else.

Actually, I can in terms of the subject.

The subject is Does Fitbit Really Work.

Since fitbit doesn't look at my fork or weigh my food, the only real things we can answer regarding the fitbit is does it accurately track my activity.

Therefore, in the context of the question presented, I can and rightfully SHOULD limit the discussion to those things the fitbit purports to measure.

Francl27

I used to do some bike in front of my TV. Then I got a fitbit... and I pretty much stopped. I was burning maybe 200 calories in 1 hour - absolutely pathetic. Either that or my fitbit was really completely wrong (it has HR), but I suppose that it kinda makes sense compared to walking because you really only move your legs... (talking stationary bike here).

I'm still not sure how people get those 400 calorie numbers - I was KILLING my legs for less of a burn than a 1hour walk.