FitBit calories 1,100 daily calories lower than my calculated TDEE

MandaLeigh123

(reposted from a status update)
I am still not sold on the accuracy of this FitBit Flex. I simply do not have the time to adjust every time I walk 5-15 minutes with my 37lb nursing bags.... and I wonder if that's whats causing this huge difference. I wore my HR monitor around campus one day, and while carrying my bags and taking stairs my HR would jump up 140-160 for the times I was carrying them. FitBit seems to be way underestimating my NEAT. I'll keep using it for a month to see what it says, just out curiosity. Its telling me my average TDEE for last week would be 1, 764/day! That's about 1,100 under my estimate of 2,850. I haven't calculated the accuracy of my HR monitor, but the HR burn seems about right and I do make sure to enter that in manually to adjust my TDEE on FitBit. Going forward, I'll probably only use the information to track my steps and sleep. Will be interested in getting one of these new activity trackers coming out soon. I know a lot of people like these, but for whatever reason, it doesn't seem accurate for me. Just like the spreadsheet, my actual TDEE number isn't close.... I wonder how my numbers are just so different.

heybales Do you think you have more LBM than average gal your age/weight/height? That could be one reason, because boy do you use that muscle daily! Comparing what Fitbit is using as BMR, to say Katch BMR using BF%, can show enough difference that tweaking Fitbit would be useful. That is accomplished by increasing your height, which increases the BMR that Fitbit uses for all non-moving time, and as start of calculation for step-based calorie burn. Wouldn't effect HR calorie burn much, depending on what formula they use to get stats used in the calculations. If curious, let me know in PM if desired, or if public that's fine too.

MandaLeigh123 Yeah, I'm definitely interested in tweaking it for accuracy. I'd like some tool to help me with this bc my acitvity level is going to be changing every four months due to school for the next two years of this nursing program. I don't really know if my LBM is more than average for someone with my stats. Actually hadn't given it too much thought. I don't even know how to calculate that. Guessing I'd need to know my body fat %, which I don't know but could figure out. Would a cheapy caliper work for this? I need something affordable- a DEXA scan isn't in the budget. I'm googling stuff on body fat percent right now. I am okay to post all publicly, just to share the info. Actually this 25% gal looks closer to me but I'm more muscley: http://extrastrong.ru/wp-content/uploads/2013/01/body-fat-percentage-women-abs-and-bossom.jpg definitely not 20-22%
edit: I'm actually looking at lots of pics of body fat %, and I'd say I'm not a good person to judge what I look like. A lot of the other 25% look much bigger than me.... so I don't know.

The other thing I was thinking this morning @heybales, is that I know you've mentioned that calorie burn is inflated on a HR monitor unless cirucuit training, which I understand and agree with you. But I decided to go ahead and keep adding the inflated lifting numbers into my FitBit because if I didn't, the number would have been even lower than 1,764.

I think it's awesome that many other folks just seem to slap it on their wrist and its pretty close to accurate, but mine is miles off and it would be great if I could somehow fix it to be more accurate.

skinny4me2be

I know you didn't post this to me, but I just have to chime in as I'm finding the same issue. I have a fitbit One, I have my height, weight, and age in my account plus my strides which I've tested over and over. I'm finding that m fitibit One was underestimating my activity as well. I just recently purchased the Fitbit Charge HR-the one that montiors your heart rate all day....on average...mine was under estimating by 2-500 calories!! I'm still in the testing phases of this but boy has it made a difference and for all i know the Charge HR could still be underestimating..... Just an FYI.

MandaLeigh123

Thanks for the input! I didn't actually post to anyone in particular, just in case someone else finds the information useful. Of course, I do want heybales to help me tweak my FitBit if he can. Someday I will probably get another tracker, but this one was gifted to me and so I am trying to make it work. As a student, I don't have a lot of extra cash to be buying fun fitness gadgets. I'm just glad I calculated my TDEE without a monitor or I would be under eating by a lot. I'm also glad I mis-entered my information on Scooby and said I was "male" because if I was going off of the Scooby calculator, the max cals there for my stats are around 2600 for a very active lady. I'm glad I just kept pushing my calories up and up to see what my TDEE was on my own, rather than using a gadget or an online calculator! So it's just a thought for you too, that even your Fitbit Charge could be underestimating. I just kept increasing my calories until I started to see a slow gain. Which wasn't until I was up around 3,000 calories/day so then I cut back down.

tigerblue

I've had two trackers now. The first was a Bodymedia (like on biggest loser), and it hugely overestimated my NEAT. I gained 10 lbs and never consistently ate anywhere near what it said. I finally realized how inaccurate it was when it counted typing as exercise for me. It was just hyper sensitive to my arm movements and counted even the slightest movement as steps. Incidentally, it seemed to track my running activity very well, as compared to my running app and my HRM.

I ditched it about 6 months ago and got a Misfit Flash. It was on sale for $24.99 at Christmas. Regularly it is $50.00. It is the cheaper version of their Shine product. They say the difference is in the aesthetics of the products, not in the actual working parts.

I love it! It gives me a number in the ballpark of most online calculators. But on run days, it shows me a calorie burn of about 400 more than on non run days. That sounds about right. If I sit around all day it gives me a low number. For lifting, I tag it as tennis to pick up the arm movements better. That also gives a pretty reasonable number for my workout per online calculator estimates. And the best part is that if I eat according to it, it seems to produce reasonable weight change results. So for me that verifies its accuracy.

Anyway, I'm posting because:

1. You should know that monitors are not the gold standard (it looks like you aRe figuring that out, but I just wanted to give you another real life example).

2. Misfit is available as a very affordable option that works well for me, anyway.

MandaLeigh123

Interesting- thx for more feedback. Online calculators also vastly under estimate my TDEE so if you say MisFit is close to online calcs I dont know if it would work for me... thats awesome that its cheap though! I def know they arent gold standards & wouldnt have one if it wasnt gifted to me. But now that I have it, I figure its worth trying to get it to work for me!

tigerblue

MandaLeigh123 wrote: »

Interesting- thx for more feedback. Online calculators also vastly under estimate my TDEE so if you say MisFit is close to online calcs I dont know if it would work for me... thats awesome that its cheap though! I def know they arent gold standards & wouldnt have one if it wasnt gifted to me. But now that I have it, I figure its worth trying to get it to work for me!

Just use your good common sense and don't let it lead you astray! I think that was my problem. Bodymedia was giving me numbers that were way off kilter. My own common sense should have told me to be skeptical. But I kept trying to use it for over 18 months! Slow learner, I guess!

heybales

Ok, BF% can be estimated close enough for this by several methods that are free or cheap.

And calipers in the hands of someone not trained nor experienced are no where near accurate enough. They can be upwards of 5% in the right hands, using the 3 or hopefully 7 site method.
http://www.exrx.net/Calculators/BodyComp.html

Average that with the results of some measurement calcs using different body parts which can also be upwards of 5%, and you got yourself a figure close enough for the intended purpose. This is also why the picture method can be difficult, just as we could have parts not in proportion to others, it is because we may carry fat last in 1 or 2 spots that throw off pictures, or even these measurement methods, which at least use different parts.
www.gymgoal.com/dtool_fat.html

My main spreadsheet has all these methods included the bodyfat % calculator for logging and doing the math on, though it requires the 7 site method of calipers. Also includes a place to include a BF% scale figure.

Getting a good stride length for average walking pace and running pace is also required, not just for general accuracy, but when you change the height, Fitbit will change default stride length. And just because metabolism is higher doesn't mean stride is. 2nd to last question here.
http://community.myfitnesspal.com/en/discussion/10098937/faq-syncing-logging-food-exercise-calorie-adjustments-activity-levels-accuracy

So when you get the BF%, will need general stats along with it - age, weight, height.

Then to confirm what Fitbit is using, looking at their graph for calorie burn per 5 minutes, the calorie burn for your non-moving blocks of time, should stand out easy enough.

And since you likely stand a fair bit at your job, but are only given sleeping BMR calorie burn - I'm not surprised it's underestimating.
Carrying the backpack with extra weight, well, that's just not going to get factored in nicely anyway you slice or dice it.
Other thing you are underestimated for, processing the food you eat is about 10% of what you eat, for you that means about 280 calories is unaccounted for.
Also, lifting at maintenance means getting a great workout in, which requires more repair. That repair means increased metabolism for 24-36 hrs post workout, also unaccounted for. If the workout really tore you up, you can get just as much extra calories during the repair, as during the workout, which while it is low, that does help.

If anyone else wants a check, just need those stats.

MandaLeigh123

Okay, I am going to head to the store and get a better measuring tape. I realized about two months ago mine starts at 1.75 inches so my measurements have been off since the beginning. Will work on getting body fat % measured tonight the tape measure way. Thanks for all of the added information. Everything you are saying makes a lot of sense as to way the numbers aren't adding up for me.

MandaLeigh123

So in the spreadsheet with my numbers: Average body fat % of the two to four methods 30.15%
Meh. This is probably the first time since I've started EM2WL where I've had some feelings over seeing a number I didn't like. I don't know how accurate this calculation is, but I've been looking at all sorts of pictures of people who are 30% body fat and I don't look anything like them. Anyway, my height is 5'7" and I weigh 145lbs, 33 years old.

Thought it was interesting at the bottom where it says "need a suggested goal weight?"
I have a small frame and it recommends 122lbs! This is just not accurate for me personally. When I started getting below 135lbs, I start looking pretty sickly. At 129lbs my grandma started writing my mom letters saying she was worried I had an eating disorder. I'm guessing 135lbs would be as low as my weight would ever go and and still let me retain my muscle mass.

Here are my measurements for reference:
Bust (over nipple) 36
Chest (under bust) 31.5
Waist (natural) 29
Abdomen (at belly button) 37
Hip Dip (in the middle)
Widest Point (bottom of butt/top of pubic hair) 40.5
Thigh ( widest point) 24.25
Knee (directly above knee cap, 7 inch down from top of knee standing) 15
Calf (widest point) 14
Upper Arm (kimono) 11
Forearm (right below elbow) 9.25
Wrist: 6
Neck: 12.25

heybales

I'll mention the goal weight is based on several used formula's from Dr's research, merely there for those that have never seen a healthy weight since childhood so have no idea. Based on average body composition. But who wants to be average!
www.cordianet.com/calculator.htm

If the body fat calc said "you may not measure well" - it's because the 2 calc's are not within 5% of each other, and since they mainly use different body parts, usually indicates you may have some part out of proportion that is messing up the averages it was based on. This may be a part always out of proportion, or just now with some fat to lose.
After a month and you see one drop more than the other, it's usually the slow moving one that is more accurate side of the range, though the whole range will be dropping.

For this calculation though, 5% isn't that big deal, and you'll see that because I'll use 25 and 30%

Mifflin BMR used by MFP and close to it by Fitbit.
1395

Katch BMR - 30% - 25%
1364 - 1435

So only a 71 cal difference, obviously at your activity levels, that can translate in to much more between ranges.
Compared to Mifflin though, it's right in the middle of the Katch range.

Height adjustment to get Fitbit to use BMR close to Katch BMR and Cunningham RMR average.
30% - 25%
69.7 inches - 74.2 inches

So if you think you are closer to 25% - use the taller figure.

If you think right smack in the middle perhaps at 27.5% - 72 inches.

So try that, confirm you have a manually corrected stride length in there, if it is already, it won't be changed by changing the height.

That will increase daily calorie burn by some amount, let's see for the between figure 72 inches.
1764 currently / 1395 = 1.2645 avg METS daily
1.2645 x 1475 = 1865 at a minimum

But you can't really do the whole day math like that, because Fitbit literally takes your per minute calorie burn (was 0.96875, now 1.0243 or 5.7% increase) times whatever level of pace it seems your current steps are at.

The higher figure will help out with the non-moving time that it sees.

So have to see what a week gives you.

MandaLeigh123

Damn right who wants to be average- haha. I see what you mean too. There are definitely people who just don't know what to aim for. This is all so awesome. This is so helpful and I just stink at the math stuff so I really appreciate the time you took to help me out with this one. I calculated stride yesterday three times at the gym and took an average of the three since they were all slightly different, so I'm hoping this will all come together to help me a little!
I think I might put myself at 27.5% and go from there! The body calc did say, "you may not measure well". Hah. Whats new? That seems to be thing with all these numbers. I don't measure well I agree with what you say about proportions too. I have a pretty small waist compared to the widest portion of my hips so that might be something that makes the % skew a little. I do have fat to lose as well.

heybales

I should mention too, that average METS figure, which is a unit defined as burned divided by resting calorie burn.

You have estimated you burn say 2800 calories daily, based on eating levels.
Your estimated resting calorie burn is 1549 for Cunningham RMR based on BF%.
2800 / 1549 = 1.81

That is enormous, and possible. Sadly can't adjust BMR level to compensate for that, since it's the total day's burn.

But, just wanted to confirm one little possibility.

How do you log your foods, by weight?

And do you throw a safety factor in there so the eating calories will always be higher than they may really be - "just to be on the safe side" ...?

If that is the case, your eating level may not really be 2800.
Same problem for those that measure their foods instead of weigh, usually the opposite direction. They think they are eating say 1500, but in reality closer to 1800 say.

In the scheme of basing math on eating levels - not that big a deal. Eating a real or inflated 2800 calories, when time to take a deficit you'll eat say a real 2380 or inflated 2380, a real deficit either way, fat loss occurs.

And now, you are attempting to get a Fitbit adjusted well for you so it can keep track of upcoming changing activity levels. And then you'll potentially base eating levels on it's recommendation.
So at least the same non-issue in the scheme of things if it adjusts correctly.

Just a thought, depending on how you log food.

MandaLeigh123

Good thoughts! I measure & dont weigh foods. I dont add calories to be safe. So if anything, I could be eating more than 2800.
Havent adjusted the FitBit but will do that soon.