Burning calories question
Options
ladavis88
Posts: 2 Member
When I log in my exercises it automatically gives me a number for calories burned. However, this number is very different from the amount of calories burned according to my Very Fit Pro program (think generic fit bit, and is basically a step counter). My question is, are the calories burned on myfitnesspal based on my height and weight that I put in to the program, and therefore more accurate than the ones shown on my other program?
0
Replies
-
I found them to be pretty spot on but others here have not. You have to track your intake for a while, adjust up or down accordingly.
How much of a difference are you speaking of?0 -
When I log in my exercises it automatically gives me a number for calories burned. However, this number is very different from the amount of calories burned according to my Very Fit Pro program (think generic fit bit, and is basically a step counter). My question is, are the calories burned on myfitnesspal based on my height and weight that I put in to the program, and therefore more accurate than the ones shown on my other program?
In my experience, some of them are pretty ok and others vastly over estimate. What kind of difference are you seeing? Either way, it's an estimate...I'd stay conservative.
I'm a cyclist, and back in the day I just used the estimate from my HR monitor for my rides. I use a power meter now, which is very accurate...my HR monitor was good enough, though slightly inflated from the numbers I get with a power meter.1 -
In my case, MFP overestimates my calorie burn from swimming by as much as 30%. I've worn a temporary chest band heart monitor, and swimming laps for 45 minutes burns about 300-325 calories, but MFP tells me I've burned 420-450. I guess I'm just a better swimmer than MFP thinks I am :-)
I just adjust the time I record in MFP. Instead of logging 45 minutes, I log 32 minutes, which comes out about right based on what the heart monitor indicated1 -
Yes the database is based on METS which take your personal stats into account.
Not that is actually suitable for all exercises, for example a when cycling it's your power and not your weight that estimates should be based on.
With a database of hundreds and hundred of different entries it's not surprising those entries vary enormously in accuracy. Some are fine, some are very high, some are probably low for some people and their particular selection of a few exercises from a vast selection.
If you told people what your exercise is and gave examples you can get some different opinions and possibly better methods than either your gadget or the MDP database.0 -
When I log in my exercises it automatically gives me a number for calories burned. However, this number is very different from the amount of calories burned according to my Very Fit Pro program (think generic fit bit, and is basically a step counter). My question is, are the calories burned on myfitnesspal based on my height and weight that I put in to the program, and therefore more accurate than the ones shown on my other program?
Calorie burning app are an estimate. They take into account weight and age and create an average.
There is no way to give an exact number of calories burned. They are based on many factors.0 -
One issue with the watches is they use HR data to determine effort. However your HR can be raised because you aren't fit or because you're hot or because you didn't sleep well the night before or because you are fighting a virus. When I run, my watch gives me more calories for the later miles of a run than the early ones because my HR drifts upward.3
-
@cwolfman13 - You hit the nail on the head! Do a workout and track it three ways. First estimate the calories in MFP. Let’s choose cycling for 60 minutes at 18-20 mph. You get like 1000 calories. Now look at your heart rate monitor. It will say you burned 850 calories. Last, upload data from a powermeter. It will tell you that you < 700 calories.
The first is an estimate based on body weight, age, estimated fitness, RPE, moon phases, whatever else. This is not very accurate.
The second is a measurement of response to physical activity. This is how your heart reacted to effort. The trick here is that your heart also reacted to a bunch of other subjective inputs that may not contribute to work.
The third measurement is repeatable and accurate (scientific). It is also based on physics. Work = Force x displacement (distance). One joule is equivalent to the work done when 1Newton of force is used to move anything 1meter. Using these irrefutable laws, you can very accurately measure the work you did with a powermeter during an exercise. The body’s actual calorie burn can vary from person to person, but you are at least defining the work done in scientific terms. Everything else is simply guessing. (Some guesses being more educated than others!)
0 -
I'd say it's a bit of trial and error. I have found some of the amounts to be spot on --- and some to be wildly high/low.
I basically use a combination of: what MFP says, what my HRM says, and what my RunKeeper app says and pick a number somewhere in the middle of all that and then if I notice that I'm gaining or losing I can adjust that (I'm maintaining my weight)..... Good Luck!1 -
@cwolfman13 - You hit the nail on the head! Do a workout and track it three ways. First estimate the calories in MFP. Let’s choose cycling for 60 minutes at 18-20 mph. You get like 1000 calories. Now look at your heart rate monitor. It will say you burned 850 calories. Last, upload data from a powermeter. It will tell you that you < 700 calories.
The first is an estimate based on body weight, age, estimated fitness, RPE, moon phases, whatever else. This is not very accurate.
The second is a measurement of response to physical activity. This is how your heart reacted to effort. The trick here is that your heart also reacted to a bunch of other subjective inputs that may not contribute to work.
The third measurement is repeatable and accurate (scientific). It is also based on physics. Work = Force x displacement (distance). One joule is equivalent to the work done when 1Newton of force is used to move anything 1meter. Using these irrefutable laws, you can very accurately measure the work you did with a powermeter during an exercise. The body’s actual calorie burn can vary from person to person, but you are at least defining the work done in scientific terms. Everything else is simply guessing. (Some guesses being more educated than others!)
MFP does METS estimates. They're research based. In theory, they're better for some exercises than others. There's more information about the methodology here: https://sites.google.com/site/compendiumofphysicalactivities/home
As I understand it, MFP's implementation is flawed. It uses gross METS (includes BMR) when it should use net METS (without BMR). In practice, the arithmetic impact of that flaw is pretty small, except for long-duration, low intensity exercise, such as hours of slow walking. (My estimated BMR per hour is 50 calories. Being off by 50 calories on a full hour of most exercise, for most people my size, is not a huge error, in context of all the other estimates that go into calorie counting).
In practice, any *consistent* but non-crazy-high exercise estimating method can work out OK, in a context where we have so many estimates, normal amounts of exercise (small fraction of daily calories), and a practice of adjusting intake if needed based on multi-week monitoring of scale-weight results.
It's not worth stressing over. Think about the math. How far apart are the calorie estimates from your device, vs. MFP, for your usual length exercise? If you're losing a pound a week, your deficit is 500 calories daily. If the exercise estimates differ by 100 calories or so, what's the impact? Maybe pick the low one, use that method consistently, monitor the scale, and adjust eating if you need to after 4-6 weeks. It'll be fine.
If you have a well-power-metered exercise (cycling with a power meter, some types of machine rowing, maybe *some* other machines that measure watts), use that. If you do steady-state mid-band intensity cardio, a heart rate based estimate is probably non-crazy. For something like strength training (where heart rate goes up for reasons unrelated to oxygen consumption), the METS-based estimate can be your best among bad bets. For walking/running, there are formulas that can give you conservative net calorie estimates, based on distance and bodyweight. And so forth.
If I don't have a reason to know which estimate is likely to be most accurate, then I usually pick the low one, of those that seem plausible. It's been working for 5+ years now.
Like I said, what's most important is that you use a consistent method, and try to make it one that's not crazy-high. When you think about what "crazy high" means, compare the numbers to your calorie deficit. If it can't wipe out a big chunk of your deficit, or have it drive you into dramatically undereating if wrong in the other direction, watch your results and sort it out that way. That'll work fine.2 -
the only way to know how accurate it is for YOU (because it is different for everyone) is to stick to one method for several weeks and track it and see what your weight does.
for ME, the fitbit is pretty accurate. I still rarely eat back all my exercise calories, BUT my calories are set at a level (1500) that even with exercise, I am still full and satiated. So I may eat back some, but not much, typically, at least. it also helps buffer water weight fluctuations, TOM, a higher calorie day here and there, etc.1 -
@AnnPT77 - I have to disagree with the accuracy of the MFP estimate for intense exercise. Instead of simply putting in an arbitrary number, let’s use hard facts and see if the data supports the argument that MFP is even in the ballpark.
I just entered in a cycling ride at 16-20 mph. MFP has told me that I burned 1070 calories in that 60 minute endeavor. (Actual number that I just looked up on the app). To reference a real effort, I recently did a (very cold) 1:17:28 ride that was captured with a Quarq by a Wahoo ELEMNT Roam. The data from this ride shows a moderate pace at 17.2 (right within the MFP range), and my calories burned are an underwhelming 626 KJ.
The second exercise was a good bit longer and FAR less caloric expenditure was required. I truly believe that this could give a person, inclined to lose weight, a false perception of their caloric balance. Imagine eating back those extra 444 calories. MFP is off by 42% in its estimate.0 -
@AnnPT77 - I have to disagree with the accuracy of the MFP estimate for intense exercise. Instead of simply putting in an arbitrary number, let’s use hard facts and see if the data supports the argument that MFP is even in the ballpark.
I just entered in a cycling ride at 16-20 mph. MFP has told me that I burned 1070 calories in that 60 minute endeavor. (Actual number that I just looked up on the app). To reference a real effort, I recently did a (very cold) 1:17:28 ride that was captured with a Quarq by a Wahoo ELEMNT Roam. The data from this ride shows a moderate pace at 17.2 (right within the MFP range), and my calories burned are an underwhelming 626 KJ.
The second exercise was a good bit longer and FAR less caloric expenditure was required. I truly believe that this could give a person, inclined to lose weight, a false perception of their caloric balance. Imagine eating back those extra 444 calories. MFP is off by 42% in its estimate.
I feel like you're disagreeing with something I didn't say . . . or at least something I didn't intentionally say, since of course I'm more than capable of unclear writing.
First though, I'd like to go back to recap my specific advice to the OP, since theoretical stuff about exercise estimating methodologies is a bit of a tangent (not completely irrelevant). OP is comparing MFP's estimate to Very Fit Pro's estimates, for an exercise type(s) that remained unstated, but with an implication that it might have a step-based component, and asking which to use.
My advice was to compare the two, use the lower, and be consistent about which one to use. I implied but didn't state that she should back out estimated BMR from the MFP estimate. (I'd note that I have no idea whether Very Fit Pro provides a gross estimate or net estimate, whether it uses heart rate or step-movement or GPS distance to estimate walking, whether she can or has validated step-length in the device, etc.) I also suggested not worrying too much about differences that were arithmetically small compared to the overall calorie picture. (Sometimes people here totally overstress about truly minor differences in such estimates.) I suggested watching real-world weight results, and adjusting intake if necessary (that's useful for reasons beyond exercise estimates). I should've further suggested comparing the ExRx walk/run calculator's estimate, or another similar one, if the exercise was step-based. (https://exrx.net/Calculators/WalkRunMETs)
IMO, sijomial was correct to ask what the exercise is. That would let us give better specific advice. IMO, Wolfman was correct to say that MFP's estimates vary widely in accuracy by exercise type.
Overall, I think what you said about various estimating modes was reasonable. I am not trying to say that MFP's estimates are "accurate" in any general sense. Far from it. I'm trying to say that it's helpful to understand the assumptions behind the estimates.
What I disagreed with was your statement that MFP's "is an estimate based on body weight, age, estimated fitness, RPE, moon phases, whatever else." Even ignoring the jokey bit, that's not helpful, and not entirely accurate besides, IMO. MFP uses an estimating methodology, METS, that has some flaws in the underlying theory, and it implements the methodology in a way that's flawed, besides. I linked the Compendium because if someone wants to go down that rabbit-hole, that's a place to start.
In brief - possibly stating things you know but didn't reflect in your comment - METS estimates multiply BMR by a factor that was determined by research about particular activities. That means it depends on the accuracy of BMR estimates (which do use some of the factors you mention, but not all), on how closely work (in the physics sense) of the exercise mode varies with changes in (mostly) body size with little efficiency variation between humans, and the quality of the research into a particular METS value (and probably other things I'm not thinking of at the moment).
Some types of activity play *fairly* nicely with the METS system's assumptions. Others are howlingly far off from those assumptions. In some cases, even in a scenario where body size is very important to work (in the physics sense), and efficiency between humans not very variable, the particular research to derive a METS value makes the approach less than useful. If a person understands a little about how these estimates work in theory, they can make some common-sense inferences about which cases will be relatively more reasonable to estimate using METS, and which will be way, way stupid. (Many people don't care to be that technical. That would make it more important to get activity-specific advice from people who may have technical insights, such as yourself.)
Sadly, there are not good physics-based methods for estimating all activities, nor is heart rate good for estimating all activities, nor is METS good for estimating all activities. You're lucky to have a core activity, cycling, that can take good advantage of power metering (for those who have the $$ and inclination), because that is the gold standard, I agree.
The MFP cycling estimate is heading toward howlingly awful, IMO, because cycling energy expenditure is less directly correlated with body size, than (say) walking or running, even though it does IMU tend to have narrow differences in efficiency between one person and another. The implementation has further flaws, in that it uses speeds to categorize cycling intensity, which seems to suggest that riding a heavy mountain bike at X mph/kph on a steep gnarly mountain trail is equivalent in energy requirements to riding a racing bike on a purpose specific track at the same mph/kph, which is clearly just nuts. (FWIW, I don't have a power meter, because I'm not cycling-focused enough to want one, but as best I'm able to tell via comparison to other metrics, the MFP cycling estimates aren't crazy-high for my semi-heavy hybrid bike on mixed rolling terrain, mostly paved or firm-surface soil/gravel.)
It would be nice if every exercise could be power metered, but they can't. It would be nice if heart rate were an accurate energy-expenditure correlate for all exercises, but it isn't. It would be nice if the METS values were better thought out in some cases**, but it's literally not possible to make them precise in others (aerobic dance? gotta just hope people are close to average in enthusiasm, I guess - which hasn't been true in any situation I've participated in.)
** The rowing one makes me laugh, every time: The MFP entries are METS for stuff like "Canoeing, rowing, moderate effort" or "Canoeing, rowing, >6mph, vigorous effort" (the latter only a slightly better bad than the former). Sure, lump together activities with completely different biomechanics, and add in the same absurdity as in cycling, that the equipment makes no difference to the workload/speed relationship, plus make it an activity that isn't heavily dependent work-wise on the person's body weight, and that I suspect has quite a range of individual variation in efficiency. Just no. But rowing (the water kind) can't be power meter like cycling can. You can put strain gauges on oars, IMU, but the same good correlation with workload is just not there, as I understand it.
It would be super nice if every exercise had a truly good estimating methodology. Some methods are better in some situations, worse in others, except for power metering in narrow-efficiency range activities. (I actually suspect that rowing machine power metering has deficiencies for calorie estimating, because of the efficiency factor, but I can't prove it. Some of the folks I see at the gym are expending all kinds of energy, but remarkably little of it is going into the flywheel.🤷♀️)
Some people are going to say that if exercise estimating is so fraught, it's better to use TDEE methods than MFP's NEAT+exercise method. But TDEE just spitballs pretty much all exercise types as having the same calorie expenditure, and averages them over the week based on planned (not necessarily actual) exercise. That's really not lots better.
So, in practice, it's mostly going to be a crapshoot. The challenge is to understand which methods are relatively better in which situations, and use those. Nowadays, some of the better fitness trackers, IMU, are implementing different methods in different situations, which would be great (though it only goes so far . . . .). Often, MFP's estimate is not good. In a few cases, it's probably among the best of a bad lot (standard strength training, for example).
I *think* what I disagree with you on is fairly narrow: I wouldn't throw out every last MFP estimate automatically, because in a few cases the normal alternatives can be worse IMO; and I think it's useful to be somewhat clear about how the MFP estimating process actually works, rather than making it sound like some kind of Evil Dark Art, so that people can evaluate it for themselves.3 -
@AnnPT77 - I have to disagree with the accuracy of the MFP estimate for intense exercise. Instead of simply putting in an arbitrary number, let’s use hard facts and see if the data supports the argument that MFP is even in the ballpark.
I just entered in a cycling ride at 16-20 mph. MFP has told me that I burned 1070 calories in that 60 minute endeavor. (Actual number that I just looked up on the app). To reference a real effort, I recently did a (very cold) 1:17:28 ride that was captured with a Quarq by a Wahoo ELEMNT Roam. The data from this ride shows a moderate pace at 17.2 (right within the MFP range), and my calories burned are an underwhelming 626 KJ.
The second exercise was a good bit longer and FAR less caloric expenditure was required. I truly believe that this could give a person, inclined to lose weight, a false perception of their caloric balance. Imagine eating back those extra 444 calories. MFP is off by 42% in its estimate.
But you can't pull one example out of a database and extrapolate the results to the hundreds of others available.
I'm also a power meter using road cyclist and agree the speed rating estimates for cycling on here are rubbish - for me, my bike and my riding. If I was an off road cyclist on a heavy old bike the estimates would be a lot closer.
But that isn't MFP making up an arbitrary number as it's sourced from the compendium of physical activities. Blame the source but not the user.
Fortunately as we both know there are far better methods available for our sport and not just the expensive option of using a PM.
Same as for many other exercises there's a range of methods of getting estimates, some poor, some reasonable (which is good enough), some accurate.1 -
Thank you all for your input! I'll just keep doing what I'm doing and hope for the best, lol0
-
Thank you all for your input! I'll just keep doing what I'm doing and hope for the best, lol
Did I miss it or did you actually say what your exercise is?
Just a random example of the problem with step counters......
I was working at a decorating job with a partner the other day. The step counter on the phone in my pocket recorded 3,000 steps in a 3' x 6' room. Hmmm.
My partner had a Fitbit on their wrist and it recorded 6,000 "steps" - most of which were hand movements scraping off wallpaper and painting.0 -
Thank you all for your input! I'll just keep doing what I'm doing and hope for the best, lol
Did I miss it or did you actually say what your exercise is?
Just a random example of the problem with step counters......
I was working at a decorating job with a partner the other day. The step counter on the phone in my pocket recorded 3,000 steps in a 3' x 6' room. Hmmm.
My partner had a Fitbit on their wrist and it recorded 6,000 "steps" - most of which were hand movements scraping off wallpaper and painting.
Yup. I've had my Garmin register steps (cleared the step movement motivational counter) when I was hand-stirring something vigorously, while cooking. 🤣0
Categories
- All Categories
- 1.4M Health, Wellness and Goals
- 391.6K Introduce Yourself
- 43.5K Getting Started
- 259.7K Health and Weight Loss
- 175.6K Food and Nutrition
- 47.3K Recipes
- 232.3K Fitness and Exercise
- 393 Sleep, Mindfulness and Overall Wellness
- 6.4K Goal: Maintaining Weight
- 8.5K Goal: Gaining Weight and Body Building
- 152.7K Motivation and Support
- 7.8K Challenges
- 1.3K Debate Club
- 96.3K Chit-Chat
- 2.5K Fun and Games
- 3.3K MyFitnessPal Information
- 23 News and Announcements
- 936 Feature Suggestions and Ideas
- 2.3K MyFitnessPal Tech Support Questions