Calories not accurate - how to adjust (real numbers)

[Cegcd]

I've been running an experiment and the result is that calories burned are not accurate. 

Length of time: 52 days

Weight change: lost 7lbs

Total calories consumed: 110,100

Implied caloric deficit: 7 X 3500 = 24,500

Implied calories burned: 110,100 + 24,500 = 134,600

Fitbit calories burned: 173,800

Difference: I burned: 39,200 fewer calories than fitbit shows.

weight loss implied by fitbit calories: (173,800 - 110,100) / 3500 = 18.2 lbs

Actual weight loss: 7lbs

Difference: 11.2 lbs

This also doesn't factor in that about 120 calories per day are from fibre so not metabolized, and I often eat excess protein which apparently is just passed through as waste so my calories consumed is likely lower adjusted for those factors.

There is no way that I am carrying 11.2 lbs of extra water weight, or that it is just off by a random amount which is accurate on average over time. 

If you believe there is a systematic inaccuracy or any other factors I haven't considered, please share your thoughts.

How do I reduce the metabolic rate in my app?

[Baltoscott]

@Cegcd -- Fitbit's calorie burn estimate is just what an average person who is your weight/sex/age with similar movement patterns will burn over time. Very few of us are dead-on average and like you, at least when I was overweight, the app seemed to give me too much 'credit' for calories burned.  It may have been that I was not accurately recording my food. Maybe not weighing everything, or maybe the calorie estimates on the package were not accurate (legally they can be off by 20% and restaurant estimates are worse).  But I realized that it doesn't really matter.  In the end, the scale reflects the deficit. As long as it is going down -- and that is what you are aiming for, right? -- you are on target. 

There are a few other things that you should consider.  Water weight is a thing, and may have a lot to do with it.  But you can eliminate that by weighing daily -- first thing in the morning -- and then focusing on your weekly average to measure your progress.  Some daily weigh-ins will be higher because of water weight, but others over the week will be lower for the same reason, and that part of it will average out over the week.

The other issue, a little less likely over just a couple of months, is that you may be becoming fitter and actually building muscle. Muscle is denser than fat, so you can get smaller even without losing that much weight if you are building muscle. Weekly waist measurements and pictures are good non-scale metrics to help you mark your progress.

Keep up the good work.  You are losing around 3/4 lbs a week right now and eating around 2100 calories a day.  If you want to speed things up a bit try cutting back to 1800 calories a day or so for a couple of weeks and see how it goes.   

[Heybales]

Sadly the calories burned is not the only side of the equation used in your math.

Calories eaten.

Food labels in the US are allowed to be upwards of 20% off, usually given as low figures shown because that is more desired, so calories eaten easily upwards of 20% more than logged.

And that is if accurately logged by weight for everything but fluids.

No eating out since that has great potential for inaccuracy (fast food better than others for accuracy potential).

That can easily account for part of discrepancy. 

More to the burned side.

MFP gives a rate of burn during all non-step time of BMR, sleeping level rate. When awake you are burning more. When standing burning even more.

You get no credit for the TEF (Thermic Effect of Food when processed).

All that means underestimated for those items.

Known inflated calorie burns:

If your workouts are NOT steady-state aerobic, but instead HR all over the place like intervals, and anaerobic like lifting, then inflated calorie burn.

If your low level walking which would be more accurate by distance-based calories burn, instead bumps up into the HR-based area for calorie burn - known inflated burn calculations.

While the distance based calorie burn can be really good estimate for daily stuff - if you get a ton of steps and the stride-length is wrong - really magnified issue.

So you really don't know if what you need to adjust is the metabolic rate (while true the BMR is 50-70% of the daily burn but that ain't all of it), or the logging error, or other.

Mine underestimated.

I had 1 great month where I logged all my workouts for tri-training by means of other most accurate calorie burn, and every single meal was logged as best I could. Expected and actual weight loss was a tad under 5% difference - that's really good and rare frankly.

Because I am not that accurate on the food side constantly. Exercise is easier once the methods are in place.

The other thing to consider.

The 3500 calories is per lb of fat.

If because of extreme diet (7 lbs in 52 days would not imply that at 1lb a week), or poorly designed diet and workout type (going low carb when doing tons of intense cardio), people do lose muscle mass as part of the weight loss.

And muscle does not follow that 3500 cal/lb fat average for loss.

Interestingly it appears to be that much to build a pound. (3600)

And it's not that you burn it for energy often (need endurance cardio or real damage), but rather muscle is broken down everyday anyway (tad more under heavy work), and built back if you have the protein, and not too much deficit that the body requires the needed amino acids elsewhere, and you are using the muscle to tell the body it is desired.

But rather if you have that low carb but intense workouts, eaten protein will be used ineffectively for energy before being sent to rebuild muscle.

It's estimated that many dieters lose upwards of 25% of their weight as muscle mass, and their math doesn't equal out to 3500 cal/lb due to that.