Soooo...why does MFP overestimate calories burned? (if that's even true)

myszka0611
myszka0611 Posts: 17 Member
edited December 1 in Health and Weight Loss
I've read multiple threads where a delighted So-and-So would say "I exercised an extra X number of minutes, and was under my calorie goal, so I treated myself to <insert yummy thing here>, woo-hoo!" , only to see later in the thread some Debbie Downer piping in with, "Well, you KNOW that MFP overestimates calories burned, don't you, so I want you to know that you just totally ruined the three years you've been dieting because of that <insert yummy thing here> you ate. You only should have eaten 1/2 of it, you fool, or maybe just inhaled the aroma. Goodness, EVERYONE on here knows that but you, apparently". And then storms off in a huff...

So my question is....why doesn't MFP tweak their "exercise calories burned algorithm" if that's the case? Maybe even go to the other extreme by UNDERestimating exercise calories burned, so us chub-chubs might burn even more calories without even knowing it?
«1

Replies

  • MissusMoon
    MissusMoon Posts: 1,900 Member
    I haven't found that they are over-estimating with the exercise I'm getting. I double check the actual machines I'm using when I enter my weight and look them up on other sites...pretty close in my case.

    I do see people saying they over-estimate everywhere here, though.
  • panda4153
    panda4153 Posts: 418 Member
    At the end of the day everything is an estimate. For a very long time I lost just fine at the rate I put into MFP without food scales, using the MFP or machine estimate etc. I ate back every single calorie. I think what can happen though is that as you get closer to your goal weight, your margin of error becomes smaller, so we start to see threads asking how come I am not losing and then the "you may be overestimating" comes up.
  • AnnPT77
    AnnPT77 Posts: 34,600 Member
    Estimating is estimating. They do the best they can. Because the database estimate may be high, some people choose to eat back only part of it, and take the rest of the calories (if there are any in real life) as an add-on to their deficit. But the database estimates could be low, too. I think most people don't feel the need to take precautions because of that.

    Think about it: Mostly, the problem is that the estimates can't be exact.

    If the data base has a good estimate for 1 hour on the elliptical for a 200-pound person, it's an average over many observations, i.e. many different people.

    But when you get down to one person (which is what we care about in our diaries), a 200-pound person sprinting on the elliptical burns more calories than a 200-person swinging along lackadaisically. Probably a fit 200-pound person burns fewer calories than an unfit one at the same speed (because of efficiency and adaptation). The person may be using their arms a lot, or not much. It varies.

    The database estimates are science-based. But they're basic (they don't ask you how many hills were on your walk, they have a range of walking speeds in one entry, etc.) - though you can find specialized calculators on the web that use more inputs specific to particular activities, so can estimate more closely (still an estimate). Or you can use a device (Fitbit, HRM, etc.) to get a more personalized estimate (still an estimate).

    If you care a lot, you can get a better and more personalized estimate. If you don't want to be bothered, you can just use the database, and adjust for the risk you care most about - the risk that the burn is over-estimated - by not eating all the calories back.

    Personally, I work a little harder on making better estimates, then eat all the calories back. If my weight loss rate suffered as a consequence, I'd change my approach. But it worked.
  • myszka0611
    myszka0611 Posts: 17 Member
    3dogsrunning wrote: »
    You may find this blog helpful
    http://www.myfitnesspal.com/blog/Azdak/view/estimating-calories-activity-databases-198041

    MFP uses the Compendium of Physical Activities as many sources do.

    Peronally, I hate the MFP database bashing. Is it perfect? No. It is useful if used with common sense. Some entries will be better than others.



    There is not perfect way (outside of a lab) to know calories burned. Every method has it's strengths and weaknesses.

    Thank you! That explains it nicely!
  • Orphia
    Orphia Posts: 7,097 Member
    LazSommer wrote: »
    If they under estimate and some moron ends up malnourished because they can't figure it out "MUH calorie intake!", they will likely get a lawsuit because God bless America.

    Quoting for the awesome on-targetness.
  • Colorscheme
    Colorscheme Posts: 1,179 Member
    MFP uses a formula. It's an estimate. I rely on my fitness tracker, but I understand not everyone has one or can afford one.
  • scolaris
    scolaris Posts: 2,145 Member
    It's just an estimate. That's why. When I plugged my activities into MFP it always gave me a generic 500-600 calories for Zumba which I knew had to be high because I had some orthopedic stuff resolving & I knew the intensity level I was dancing at was slightly lower. My Fitbit HR confirms that I burn anywhere between about 375 to 435 from week to week depending on how my knee feels, how well I know the routines, etc.
  • senennieves
    senennieves Posts: 106 Member
    I have my fit bit surge synced in to my MFP account , and i have not been able to check my weight, but my pants are feeling loser and i have tighten my belt a little, xd.
  • nuttynanners
    nuttynanners Posts: 249 Member
    It all depends on your height, weight, age, level of exertion, the moon's gravitational pull on earth, etc. Okay the last part was a joke...

    But I like using my own HRM to get a more personalized estimation. It's still an estimation, though! You really can't rely on ANY machine or computer program to give you a totally accurate number, because it's impossible to know for sure.
  • Nicklebee93
    Nicklebee93 Posts: 316 Member
    edited April 2016
    I tend to see people that are extremely helpful and nice on these forums. Most people do recommend to only eat about 50-75% of exercise calories back because MFP can only give you an estimate.

    How does MFP know your intensity? If you followed a modified version, vs the normal or the higher version? The only way to accurately log those calories is if you do the formula yourself.

    I have a workout video that does burpees. One version is a slow, one leg comes in at a time modify. The other is a regular but without the jump in the end. And the last one is a full burpee. Depending which one i do, well then i'm burning more or less. How many did i do in that minute? Too many variables and that's why MFP sometimes lacks in the calorie burns category..
  • Pinkvela
    Pinkvela Posts: 21 Member
    I always underestimate the exercise calories and overestimate the food amount. I remember seeing a show where they tested a professional athlete on a stationary bicycle exactly for the purpose of "numbers" accuracy and there was a discrepancy between what the machine recorded and the way the lab recorded it (complete with oxygen/type mask thing and electrodes and what have you). Lesson was: It takes a lot to even burn a few calories and it has to be sustained effort at a certain intensity. My rule: If I don't break a ( big) sweat then it probably hasn't amounted to much.
  • Psychgrrl
    Psychgrrl Posts: 3,177 Member
    I figured it was the same reasons cardio machines over-estimate for people--because the burns are often based on a lbf, muscular (I.e. In shape) 20-something male. Even when adjusted for sex, age, weight, that basis can still make them off. Especially if there's no hrm input to determine actual intensity.

    Whenever I compare machine calories burned to my hrm calories burned, it's about 200-250 less for an hour on a Pre-Cor for the hrm. If I input into MFP, I have to lower the time I worked out to get the hrm calories out number, because for me, MFP over-estimates. I'm a small female.

    If someone is bigger, 250-300, I'd guess they may be burning more than the machine norm.

    The programming required for automatic customization of calories burned for each member based on their stats might not be feasible for a free site.
  • MelodyandBarbells
    MelodyandBarbells Posts: 7,724 Member
    A lot of it is results based, where people eat back a certain percentage of their exercise calories and lose the expected amount of weight per week. Now could this discrepancy be due to inaccurate food logging? Absolutely. But adjusting down the exercise calories can be a great way to compensate for that and still get good results.
  • jandsstevenson887
    jandsstevenson887 Posts: 296 Member
    I actually think MFP underestimates the workouts I enter based on other data
  • sijomial
    sijomial Posts: 19,809 Member
    You don't make an estimate more accurate by reducing it by a random percentage.

    There certainly are some estimates which seem far too generous but there's also many that are entirely reasonable and use industry standard formulae.

    The application of common sense and then willingness to adjust based on results will get people where they need to be.

    For most focussing their energy (no pun intended) on their food logging accuracy will make a far bigger difference.
  • Francl27
    Francl27 Posts: 26,371 Member
    The only time I used a HRM on the treadmill, my HRM number was actually 20% higher than what the machine told me!

    So I don't really know what to think anymore.

    But I compared my stationary bike and my HRM and it was 100% on point. So that gives me about 250 calories an hour (and it's some pretty good effort for me)... so when I see people logging 1000 calories for one hour of bike... yeah, I don't quite believe it. Sorry!

    What makes me roll my eyes is 300 calories burned in one hour of yoga...

    Either way, this is why I switched to TDEE. My activity is pretty much the same from one week to another anyway.
    3dogsrunning wrote: »
    You may find this blog helpful
    http://www.myfitnesspal.com/blog/Azdak/view/estimating-calories-activity-databases-198041

    MFP uses the Compendium of Physical Activities as many sources do.

    Peronally, I hate the MFP database bashing. Is it perfect? No. It is useful if used with common sense. Some entries will be better than others.



    There is not perfect way (outside of a lab) to know calories burned. Every method has it's strengths and weaknesses.

    Super useful article. Thank you!
  • AnnPT77
    AnnPT77 Posts: 34,600 Member
    Psychgrrl wrote: »
    I figured it was the same reasons cardio machines over-estimate for people--because the burns are often based on a lbf, muscular (I.e. In shape) 20-something male. Even when adjusted for sex, age, weight, that basis can still make them off. Especially if there's no hrm input to determine actual intensity.

    Whenever I compare machine calories burned to my hrm calories burned, it's about 200-250 less for an hour on a Pre-Cor for the hrm. If I input into MFP, I have to lower the time I worked out to get the hrm calories out number, because for me, MFP over-estimates. I'm a small female.

    If someone is bigger, 250-300, I'd guess they may be burning more than the machine norm.

    The programming required for automatic customization of calories burned for each member based on their stats might not be feasible for a free site.

    Adding just for clarity: The free version of MFP does adjust the exercise calories based on your bodyweight. I'm only slightly joking when I say it's dispiriting how many fewer calories I get for a spin class as a l'il ol' lady vs. when I was a fat ol' lady - it's like 380 calories per hour now (120 pounds) vs. 580 per hour a year ago (183 pounds). Just doesn't seem fair! ;-)

    My heart rate monitor puts me in the same ballpark before and after, BTW.
  • 85Cardinals
    85Cardinals Posts: 733 Member
    edited April 2016
    I'm suspicious by nature, not quite paranoid but close. I'm not saying the calories burns are necessarily wrong, just color me skeptical.

    Also I've never witnessed the type of scenario that's described in the original post, where someone is mocked like that. Most everybody here is pretty nice imo.
  • Budjola
    Budjola Posts: 148 Member
    rule #1 when in doubt log larger portion in mfp
  • MsBuzzkillington
    MsBuzzkillington Posts: 171 Member
    Francl27 wrote: »
    What makes me roll my eyes is 300 calories burned in one hour of yoga...

    I have had a couple yoga classes where I ended up slightly out of breath and sweating/hot. Sometimes yoga can be intense. It's not just "la de da da deee, I am going to reach for my toes and breathe slowly."

    Not saying I burned 300 calories but it IS possible to get a really good work out from yoga.
  • ElizabethOakes2
    ElizabethOakes2 Posts: 1,038 Member
    Pretty much what everyone else said- It comes down to personal metabolism, fitness level, etc. What I burn on a 1 hour walk won't be the same as what someone else burns. I took to wearing my hrm for nearly everything for a couple weeks, and now I have a pretty good idea of how much I burn doing what. Some of MFPs estimates are crazy, sometimes they're spot on. I just eat back half of what MFP says and that generally works for me. :)
  • dutchandkiwi
    dutchandkiwi Posts: 1,389 Member
    Generally I find it does not overestimate much - I check against my fitbit and find them usually within 5-10% of eachother which considering both are estimates quite good. In fact my fitbit usually gives me more calories burned than MFP. But it is important that you then have your weight set to the correct levels.

    Having said that some of the database entries state things like vigerous effort, or fast or rather subjective descriptions and that makes it harder to interpret. I tend to go for the lowest ones there. I am careful with eatig those calories back but not because I think it is overestimating, but also because that way I also catch underlogging on the food side a bit.

    Also I do think that if the food database has some varied and wildly inaccurate entries, that would also a possiblity in the excersise database. So I try to keep mindful of that. So far it has served me well and I fully intent to keep going the way I am
  • MelodyandBarbells
    MelodyandBarbells Posts: 7,724 Member
    MsBuzzkillington wrote: »
    Francl27 wrote: »
    What makes me roll my eyes is 300 calories burned in one hour of yoga...

    I have had a couple yoga classes where I ended up slightly out of breath and sweating/hot. Sometimes yoga can be intense. It's not just "la de da da deee, I am going to reach for my toes and breathe slowly."

    Not saying I burned 300 calories but it IS possible to get a really good work out from yoga.

    I've been in yoga classes that felt like being on a treadmill. In one she kept repeating this flow sequence and wouldn't let up. I wouldn't have trouble believing 300 calories in a class like that. And overall I've been to maybe two "easy stretching" yoga classes and they bored me to tears. For the most part, that stuff's hard!
  • Christine_72
    Christine_72 Posts: 16,049 Member
    I've never logged exercise on mfp. So i just did a little test and mfp gave me 20 calories less than my fitbit..
  • anonymouslyours
    anonymouslyours Posts: 1 Member
    If you TL;DR my post, here's a fantastic recommended website that shows you additional calories burned by any activity, vs. total calories burned (these include calories burned at rest ie. calories your body burns just being alive). I've linked to my example of walking:calorielab.com/burned/?mo=ac&ac=17190&ti=Walking%2C+3.0+mph%2C+level%2C+moderate+pace%2C+firm+surface&q=&wt=145&un=lb&kg=66

    Okay, so you are correct in saying that MFP exaggerates calories burned. I just realized this harsh reality last week, and I have been corroborated by my friend who is a trainer, my internet research, but mainly by the nutrition class I'm taking that made me understand this. What MFP is telling you when you enter an activity is the total number of calories burned during an activity, instead of the ADDITIONAL calories burned. For example, I weigh 145 pounds, and when I walk for thirty minutes at 3mph, I burn a total of 109 calories- this the same number that MFP gives me, give or take a few cals, and it is misleading. The real number of additional calories I've burnt is only 76. My total calories burnt INCLUDES my calories burnt at rest. Calories burnt at rest shouldn't count because they have already been factored into our total goal number of calories for the day.
    My friend told me that exercise machines and fitbits do the same thing- they all measure total calorie loss rather than the additional calories you've burnt doing the exercise. This is totally frustrating and disappointing to me. I've been informed that the best weight loss strategy counts only additional calories burned, but at the same time, the lowered numbers feel so discouraging to me. I'm not sure why MFP, other similar sites, and exercise machines are programmed this way, because it feels misleading- not only that, but people wouldn't get the results they wanted- wouldn't a more accurate model be better for business?
    But at least I finally understand why my weightloss has been slower than intended. I've lost 9 pounds while using this website for 1 year and 8 months, when I should have met my total goal of 15 months ago. I rarely go over my calorie limit by more than 50 calories, but I tend to meet my daily limit every day. It's very rare for me to be more than 50 calories under my limit. Now I realize that I need to be probably 100 calories under most days because I've been counting too many extra calories! This will make things more complicated when I'm filling out my day here. I really wish MFP had an option where you could choose to record only additional calories rather than total so I could get a better picture of how many fewer cals I shoud be eating.
This discussion has been closed.

Categories