Supposed to lose 2.4 last week. Scale said 1.2 (NUMBERS EXPLAINED!)

TheFeelsWay
TheFeelsWay Posts: 21 Member
edited November 23 in Health and Weight Loss
Some numbers to look at. Would like to know your thoughts.

I use MyFitnessPal very diligently and keep track of every single thing I eat (even non-caloric foods that contain sodium). I then use a Polar H10 heart rate monitor, along with the Polar Beat app, to track calories burned. This week I decided to do an analysis since I feel like I have been very good with the diet and should have lost at least 2 lbs. I also worked out 7 days last week; a mix of cardio and weightlifting.

Here are the numbers:
Actual weight lost (11/13 - 11/19): 233.8 - 232.6 = 1.2 lbs.
RAW DATA
Calories in: 12,063
Calories Burned: 3,746
TDEE: 16,912 -- (using Mifflin-St. Jeor Formula with the sedentary lifestyle option so I can then add in Polar Beats calories burned)

According to the raw data above, I should have lost 2.46 lbs. I actually lost half of that. Knowing that the HR monitor calorie equation has some room for error, I then started playing around with some adjustments to see what would give me 1.2 lbs. Below are just some percentages I threw around.

ADJUSTED DATA
Calories in: 15078.75 (25% error adjustment)
Calories Burned: (35% error adjustment)
TDEE: 16,912
ADJUSTED WEIGHT LOST: 1.22 lbs

CONCLUSION
There is one variable that has been left out: muscle mass. I could have gained some muscle in the process. However, the numbers leave us with things to think about: How accurate is the Polar Beat app in its calculations? How accurate are some of MFP's food database calorie counts? How accurate is the TDEE formula?
IN THE END, we should always factor in some error % adjustments to our figures. No matter what I lost some weight within the healthy weekly zone. LET'S KEEP TRUCKING! #Consistency :)

Replies

  • MossiO
    MossiO Posts: 164 Member
    You might have lost the 2.4 pounds but the human body has many variables and you cannot scientifically calculate them all. Water, for instance.
  • Unknown
    Unknown
    edited November 2017
    This content has been removed.
  • mph323
    mph323 Posts: 3,563 Member
    TheFeelsWay wrote: »
    Some numbers to look at. Would like to know your thoughts.

    I use MyFitnessPal very diligently and keep track of every single thing I eat (even non-caloric foods that contain sodium). I then use a Polar H10 heart rate monitor, along with the Polar Beat app, to track calories burned. This week I decided to do an analysis since I feel like I have been very good with the diet and should have lost at least 2 lbs. I also worked out 7 days last week; a mix of cardio and weightlifting.

    Here are the numbers:
    Actual weight lost (11/13 - 11/19): 233.8 - 232.6 = 1.2 lbs.
    RAW DATA
    Calories in: 12,063
    Calories Burned: 3,746
    TDEE: 16,912 -- (using Mifflin-St. Jeor Formula with the sedentary lifestyle option so I can then add in Polar Beats calories burned)

    According to the raw data above, I should have lost 2.46 lbs. I actually lost half of that. Knowing that the HR monitor calorie equation has some room for error, I then started playing around with some adjustments to see what would give me 1.2 lbs. Below are just some percentages I threw around.

    ADJUSTED DATA
    Calories in: 15078.75 (25% error adjustment)
    Calories Burned: (35% error adjustment)
    TDEE: 16,912
    ADJUSTED WEIGHT LOST: 1.22 lbs

    CONCLUSION
    There is one variable that has been left out: muscle mass. I could have gained some muscle in the process. However, the numbers leave us with things to think about: How accurate is the Polar Beat app in its calculations? How accurate are some of MFP's food database calorie counts? How accurate is the TDEE formula?
    IN THE END, we should always factor in some error % adjustments to our figures. No matter what I lost some weight within the healthy weekly zone. LET'S KEEP TRUCKING! #Consistency :)

    You're not accounting for water weight, for one thing - that's a large variable and needs to be tracked over time, a single data point isn't valid.
  • TheFeelsWay
    TheFeelsWay Posts: 21 Member
    edited November 2017
    I follow you guys. One data point (i.e. a week) would not be enough for a full analysis. Although I'm on day 86 of my journey, I've only been using MFP for only 3 weeks. I'm going to keep tracking this over time with the goal of ultimately trying to better estimate an error % for the HR caloric burn. It would be nice to develop a range to go by. Like you said though, the body is variable.... the week prior I lost 3.2 lbs and the prior one to that 0.8 lbs. (which - for 3 weeks - comes out to average 1.73 lbs per week). I would think that on a consistent caloric deficit the weekly numbers would be a bit more consistent as well.
  • mph323
    mph323 Posts: 3,563 Member
    TheFeelsWay wrote: »
    I follow you guys. One data point (i.e. a week) would not be enough for a full analysis. Although I'm on day 86 of my journey, I've only been using MFP for only 3 weeks. I'm going to keep tracking this over time with the goal of ultimately trying to better estimate an error % for the HR caloric burn. It would be nice to develop a range to go by. Like you said though, the body is variable.... the week prior I lost 3.2 lbs and the prior one to that 0.8 lbs. (which - for 3 weeks - comes out to average 1.73 lbs per week). I would think that on a consistent caloric deficit the weekly numbers would be a bit more consistent as well.

    I think that's one of the hardest obstacles for many people to overcome while they're in weight loss mode. If you're tracking your food carefully and have a good handle on the number of calories burned through exercise, it's so frustrating to see the scale bouncing all around on a daily or weekly basis.
  • toxikon
    toxikon Posts: 2,383 Member
    https://www.washingtonpost.com/news/morning-mix/wp/2017/05/25/fitness-trackers-are-largely-inaccurate-when-counting-calories-stanford-researchers-say/
  • TheFeelsWay
    TheFeelsWay Posts: 21 Member
    mph323 wrote: »
    I think that's one of the hardest obstacles for many people to overcome while they're in weight loss mode. If you're tracking your food carefully and have a good handle on the number of calories burned through exercise, it's so frustrating to see the scale bouncing all around on a daily or weekly basis.

    Well said MPH
  • VintageFeline
    VintageFeline Posts: 6,771 Member
    TheFeelsWay wrote: »
    I follow you guys. One data point (i.e. a week) would not be enough for a full analysis. Although I'm on day 86 of my journey, I've only been using MFP for only 3 weeks. I'm going to keep tracking this over time with the goal of ultimately trying to better estimate an error % for the HR caloric burn. It would be nice to develop a range to go by. Like you said though, the body is variable.... the week prior I lost 3.2 lbs and the prior one to that 0.8 lbs. (which - for 3 weeks - comes out to average 1.73 lbs per week). I would think that on a consistent caloric deficit the weekly numbers would be a bit more consistent as well.

    Your numbers are consistent as you've just displayed by working out your three week averages. Your body isn't a perfect input/output machine. I mean even a car on the same route will use a different amount of fuel of different days giving you different mpg. The human body is far more complex and variable.

    Use a weight trending app if you really like data analysis. Libra, Happyscale, Trendweight.
  • PAV8888
    PAV8888 Posts: 14,303 Member
    VintageFeline wrote: »
    Your numbers are consistent as you've just displayed by working out your three week averages. Your body isn't a perfect input/output machine. I mean even a car on the same route will use a different amount of fuel of different days giving you different mpg. The human body is far more complex and variable.

    Use a weight trending app if you really like data analysis. Libra, Happyscale, Trendweight.

    QFT!
  • AliceDark
    AliceDark Posts: 3,886 Member
    All of the numbers in your equation are approximations. Even if you're using a food scale and tracking as accurately as you possibly can, you're not 100% accurate on your intake; that kind of accuracy isn't possible. Fitness trackers are all inaccurate, and any TDEE or BMR equation is an estimate. (For example, the result from the Mifflin-St. Jeor equation says that a person with your stats theoretically burns about that many calories, NOT that you personally burn exactly that many calories).

    You really can't get a high degree of predictability using just those tools over such a short period of time. You CAN get a higher degree of predictability using your actual observed results once you have 8-12 weeks of data recorded.
  • ladyhusker39
    ladyhusker39 Posts: 1,406 Member
    Boy, I'm so glad I don't sweat what the scale says day-to-day or week-to-week. It sounds exhausting.
  • AnnPT77
    AnnPT77 Posts: 34,600 Member
    edited November 2017
    Consider this:

    https://examine.com/nutrition/does-metabolism-vary-between-two-people/

    Just my opinion, but I think that rather than try to massage your data to fit a "calculator"* TDEE estimate, a more useful adult science fair project would be to use your carefully-estimated** eating and exercise data to estimate your actual, personal TDEE.

    * TDEE "calculators" don't calculate, they estimate . . . and pretty approximately, at that. The linked article explains how/why.

    ** You can't precisely measure calories eaten or burned, although you can estimate them closely enough to be very useful. Food companies' calorie estimates are permitted to be 20% off IIRC, one apple is sweeter than another, etc. Also, I hope you're not thinking a HRM gives you any kind of reasonable calorie estimate for weight training?

    P.S. The standard TDEE calculators underestimate my pre-exercise TDEE (a.k.a. NEAT) by 30% or more. Why? Don't know . . . and only care academically. I just happily eat all those calories, and stay at a healthy weight (I'm in maintenance, after losing 50+ pounds a couple of years ago).

    Edited: Typo
  • TheFeelsWay
    TheFeelsWay Posts: 21 Member
    Shellz31 wrote: »
    ladyhusker39 wrote: »
    Boy, I'm so glad I don't sweat what the scale says day-to-day or week-to-week. It sounds exhausting.

    Don't presume it's sweating the scale. Some of us love crunching numbers. I have an excel where I calculate my weekly tdee based on calories and weight, then average it all out to try to get an idea of what my personal maintenance intake would be. Obviously, It's a moving scale as I lose and not incredibly accurate, but it's a fun exercise for an accountant like me :)

    I heard that! I just love the numbers...keeps me more engaged in the process and more motivated.

    Thanks to all for the great feedback.
  • yirara
    yirara Posts: 9,985 Member
    One other thing to keep in mind: Heartrate does not equate calories burned. There's no correlation between both. Let me give you a few examples:
    - you're very unfit, there's an icecreme van stopping in front of your home and you run out to get it. Your heartrate shoots up.
    - you're very fit, there's an icecreme van stopping in front of your home and you run out to get it. Your heartrate does not shoot up.
    - Given the same bodyweight you're calorie burn is about the same.
    - I didn't pay attention this morning and took my thyroid meds twice. As a result my heartrate is elevated. I don't burn more calories though
    - I run and it's a lot hotter than normaler, resulting in a higher heartrate. Still I don't burn more calories.
    - The same for having a small cold, for not feeling like working out, being stressed, etc.

    If you are running then 0.63*weight in lbs*distance in miles is a good estimate for calorie burn. For walking I think the multiplyer is around 0.3
  • ssbbg
    ssbbg Posts: 153 Member
    @PAV8888 shared a lovely spreadsheet for calculating TDEE here: http://community.myfitnesspal.com/en/discussion/10534897/tdee-calc-vs-fitbit#latest

    (@PAV8888 Thanks again for sharing! I've used it daily since I've got it, and I've found it extremely useful.)

    I would certainly recommend more than a week's worth of data to get an accurate number. You can modify it to separate out NEAT vs. Exercise calories. If you are particularly meticulous, you may be able to figure out which methods to determine exercise calories values are most accurate. But that would require much, much more than a week's worth of data.

    I also use the 0.63/0.3 (run/walk)*weight* distance value for running and walking. GPS based methods, like Strava, give me values that are 3x higher and I really don't believe them (in part because I have a pretty good idea of what my NEAT is based on many months tracking in the spreadsheet, and my weight loss numbers make more sense with the lower 0.3*weight*distance numbers.)

    This is where those 0.63 and 0.3 multipliers come from: https://www.runnersworld.com/weight-loss/how-many-calories-are-you-really-burning

    There is an updated article that has similar numbers: https://www.runnersworld.com/peak-performance/running-v-walking-how-many-calories-will-you-burn

  • yirara
    yirara Posts: 9,985 Member
    I also get very high numbers from most running apps. In my case they seem to be closer to reality on short runs, and get more exaggerated the longer the run is. I have the feeling that duration is somehow taken into the equation as I'm a super slow runner and my average 10k takes 80 minutes. Oh well... I have a couple of custom entries in the cardio section and chose the one that comes closest to the data based on my own experience, and 0.63x seems to work for me.
  • Shellz31
    Shellz31 Posts: 214 Member
    ssbbg wrote: »
    @PAV8888 shared a lovely spreadsheet for calculating TDEE here: http://community.myfitnesspal.com/en/discussion/10534897/tdee-calc-vs-fitbit#latest

    (@PAV8888 Thanks again for sharing! I've used it daily since I've got it, and I've found it extremely useful.)

    I would certainly recommend more than a week's worth of data to get an accurate number. You can modify it to separate out NEAT vs. Exercise calories. If you are particularly meticulous, you may be able to figure out which methods to determine exercise calories values are most accurate. But that would require much, much more than a week's worth of data.

    I didn't mean to imply I only use a week's worth of data, just that I calculate it on a weekly basis then average all the weeks together. The weekly calculation varies by thousands week to week so it's obviously not accurate, but averaging over months of data is a bit more reasonable.

    The reason I wanted to is because I walk leisurely a decent amount (city dweller) and I don't rely on trackers. I wondered about how much on average I burned. So far it's working out to about 150 calories above mfp sedentary which seems realistic.
  • PAV8888
    PAV8888 Posts: 14,303 Member
    Thanks for the ping @ssbbg

    I am on the phone so I can't easily compare but I think that the latest version can be found at: https://docs.google.com/spreadsheets/d/1VDmqNpLPu7sbQSochUJNXdp2F7AN15AGgkvS3zLw1GU/edit?usp=drivesdk
  • PAV8888
    PAV8888 Posts: 14,303 Member
    edited November 2017
    @yirara most exercise burn values include the portion of burn that is assigned to your basal metabolic rate. If the app you're using does not discount the non-exercise activity calories that you are given for the time period of your exercise... the longer the exercise and the less intense it is, the bigger the cummulative and relative errors.

    For example if I go for a two hour walk and mfp gives me 600 calories (5 a minute) but does not remove the 240 (2 a minute) calories it was going to give me anyway there is 240 calorie error, more than a third of the total exercise.

    If I was running for 30 minutes at 10 Cal a minute the corresponding values would be 300 and 60, and the error would only be 60, just a fifth of the total exercise.

    Since most people do not exercise for that many hours these errors end up being relatively speaking small but they are the basis, I think, for the advice to only eat part of your exercise calories that is often heard

    However please note that this is **totally different** than an exercise adjustment that you receive from a device such as a Fitbit.

    Integration uses the exercise adjustment as the mechanism, but the base calories that have been assigned to you ARE taken into account before that adjustment.

    So the only reason not to eat device integration exercise adjustments is because you have somehow determined that for some reason your device / band does not correctly estimate your total daily energy expenditure. Or that you have to compensate for food logging errors. Or some combination. Which is what I try to capture with that spreadsheet.
This discussion has been closed.

Categories