Accuracy of Workout Calories Burned

mschurmann

I just started tracking my gym workouts with MFP. Previously I just let my Apple Watch log calories based on time of a Function Strength workout. The MFP calorie estimate is about twice the Apple Watch. I would expect MFP to be more accurate given the specifics of weight and reps, however I am finding it very hard to believe that I am burning 700 calories in a 1 hour workout. Curious if others have found tracking workouts to be effective.

tomcustombuilder

One of the biggest reasons people don’t lose weight with this app is because they eat back their exercise calories and use those calories that they’re telling you were burned and those are going to be potentially inaccurate. You can eat back calories do whatever you want then after about 4 to 6 weeks of doing that you’re gonna have to look at what your weights doing and make adjustments if necessary.

Retroguy2000

Wait, how are you using MFP to track your workout?

I use Cardiovascular->Strength training and enter session time and the result is what I'd reasonably expect from MET estimates for lifting. I don't use a fitness tracker watch but they're going to be inaccurate for that sort of thing anyway, they're better suited for cardio.

Something else you need to consider is the watch may be telling you that you burned 700 in an hour, and depending what you were doing maybe you did actually do that, but:

a) Let's assume that's correct, it might not be.
b) It's not 700 "extra", because had you been sitting on the sofa watching Netflix you'd still have burned 70-150 doing that.
c) Don't eat back all of those regardless of how MFP is designed wrt workout calories. The body doesn't work like that. Eat back about 50% and monitor your progress over weeks.

PAV8888

@Retroguy2000 I have to admit you know a *kitten*load more about strength workouts than I do.

But how on earth are you coming up with (c) above. I don't disagree with your actual remedy. Just with the stated reason: "the body doesn't work like that"

In actual fact the 50% is more probably than not correct but caused because the body does work with actual additional net calories as opposed to hopeful calories!😛

The actual arithmetic of net calories less reduction in neat subsequent to a workout (plus inaccuracies plus random departures from the mean) end up making the 50% workable.

The MET values in MFP are gross calories. MFP has already assigned Calories to the time frame of the exercise that would have to be subtracted to get net calories. The extra arguably applies to a time frame of doing actual "work" not "breaks". And there may exist some compensatory reduction to subsequent NEAT. Or, conversely, some afterburn increase 🤷‍♂️ or logging inaccuracies or individual departures from the mean.

In any case, for most people, the actual net will be between (and NOT) 0 and the estimated value. So 50% and reevaluate tends to work fairly well.

Now specifically for watches and specifically while integration is actually working correctly.

Integration (with all activity entered on the watch side) results in the watch estimate replacing the MFP activity value one has selected.

So you've off loaded the estimate of caloric expenditure to your watch.

It will either do a good job or a terrible job. But it will tend to do a relatively consistent job (or mess) for most people as the case may be.

If you also log your food consistently in a few weeks you will have a relatively consistent view of what your logging tools are expecting your weight to do and what your weight actually does.

You can then figure out how big of a gap you have and come up with better eat back percentages

And last but not least you have to apply some sanity checking. Heart rate is detectable and standing in for the amount of oxygen you're consuming to fuel your muscles. Which is a bit harder to detect on a watch 😛

So if you're using up oxygen because you're moving you're burning calories. But if your heart rate is up because you're ready to explode because of what you heard on the phone or on TV, well then, you're less likely to be legitimately burning calories.

There's a number of exercise calculators online that give out net calories

Most of them are an OVER estimate of net calories burned when it comes to MFP.

They tend to net out 1 * BMR Calories per minute of activity to arrive at their net values. But, MFP (as of 2025) appears to assign 1.4 (1.6, 1.8 and 2.0) * BMR Calories per minute based on the activity setting one selects during setup. (Old timers may still be using 1.25, 1.4, 1.6, 1.8 which were the previous values)

If one is trying to lose weight, a deficit is necessary.

But the goal is not/should not be to create the largest deficit possible to get this done as fast🐇 as one can. That path, more often than not, ends up in another go around a bit later. A reasonable, sustainable, fast turtle 🐢 is what I would want!😎

Hence the 50% and evaluate progress tends to be good advice.

springlering62

I’ve found my Apple Watch to be fairly accurate. Sometimes, I even count steps, too, just to see if it agrees with me, and it’s generally pretty close. The only time I’ve found it to go buck wild is if I’m doing something with a very specific, very repeated hand motion, like crochet. Once in a while it will give me thousands of “steps” if I’m happily immersed in crocheting something.

Why in the world have you invested in a tracker with a great reputation, and decided not to take advantage of it?

Not saying that to be a smartarse, just genuinely curious. It’s an amazing piece of kit, and I credit much of my own loss and maintenance to the data my Apple Watch has provided.

If you use it as intended, it will literally rock your world. It did mine.

@Retroguy2000 that’s exactly what the Apple Watch calorie adjustment is for.

springlering62

PS I don’t log weight room workouts on Apple
Watch because they are so stop and go. I do, however, record my training sessions as “functional strength” workouts.

Most the others are pretty straightforward. I use yoga, walking, running, swimming, cardio fitness etc.

It still utterly blows my mind that not only can it track swim laps, but knows if I’m doing back stroke versus freestyle. How does it know?!!! 🤨

PAV8888

Siri is on to you @springlering62 . Just don't tell her about the Milka in your suitcase!!!🤯🤭

springlering62

They’re all listening, Pav. I can’t tell you how many times in the past week or so I’ve gotten ads for things discussed casually at home or with a friend.

Little granddaughter facetimed, entranced by a rainbow, and we were simply admiring with her about how beautiful it was and how much we loved rainbows. Now I’m getting ad after ad for rainbow Pride merch, lol.

I bet after using the term “Milka MaXXX” which is their XL size chocolate bar, I’ll start getting condom ads.

In fact, ya’ll all might get them now. 😂😂😂

😂😂😊

springlering62

Sorry, totally inappropriate thread hijack.

AnnPT77

mschurmann wrote: »

I just started tracking my gym workouts with MFP. Previously I just let my Apple Watch log calories based on time of a Function Strength workout. The MFP calorie estimate is about twice the Apple Watch. I would expect MFP to be more accurate given the specifics of weight and reps, however I am finding it very hard to believe that I am burning 700 calories in a 1 hour workout. Curious if others have found tracking workouts to be effective.

TL;DR: As a generality, yes, I've found tracking workouts to be effective. I've estimated exercise calories carefully through just under a year of weight loss, and 9+ years of successful maintenance. During those years, there have been periods where my exercise load varied significantly, sometimes very little exercise at all (injury, illness, surgical recovery) for a month or more, sometimes quite a lot (because I'm a fairly active person). My body weight trends/magnitude based on calorie intake plus those exercise estimates have been quite predictable.

Two important things behind that:

1. Estimating exercise calories semi-accurately is important, if a person does much exercise. In that context, it's not necessarily as simple as unquestioningly believing MFP or a fitness tracker. IMO, it's useful to have at least a general understanding of how the estimates are derived.

2. When starting out calorie counting, IMO it's vital to use the first 4-6 weeks to reality test MFP's base calorie estimate, (or a whole menstrual cycle for women of that age/stage). Most people will be close to MFP's estimate, but for those few who aren't, that will confuse the whole issue further once exercise is added. This issue applies even for people who don't log exercise, though: The initial estimate may be inaccurate for people who are somehow not matching their demographic average calorie needs. It also applies to calorie needs estimates from other calorie calculators and even fitness trackers.

The MFP method can work fine; so can averaging planned exercise into a TDEE estimate and subtracting calories from TDEE estimate to get a calorie goal. If a person understands what they're doing, it's just a difference in accounting. I'd give a slight edge to MFP method for people with fairly high or variable amounts of exercise, or to TDEE method from people who want the same calorie goal every day and either don't do much exercise, or it's a consistent amount over time.

OK, end of TL;DR. Here's the stupid-long essay:

If the watch (whatever brand) estimates strength training calories using primarily heart rate, it's likely to overestimate calories in many cases. Heart rate typically increases during strength training for reasons that have little to do with oxygen consumption, and it's oxygen consumption that correlates to calorie expenditure. Heart rate is just a proxy for oxygen consumption, and a very imperfect one. (I'm going to omit the explanation for that "imperfect" alone, here, mostly for length.)

I've read that some fitness trackers now use METS to estimate strength training calories, which conceptually is probably a better idea. METS is what MFP uses. Reportedly, the "Strength training (weight lifting, weight training)" entry in the MFP cardiovascular section is based on research that considered fairly standard reps/sets strength workouts, and the between-set rests are factored into the METS value.

(For clarity: In general, MFP uses METS for exercise estimates; I believe the MFP-created exercises in the cardiovascular section are based on research-estimated values, but may not be updated as better research is available.)

PAV is IMO correct about the potential BMR/activity factor overestimate from the way MFP implemented this. Personally, I think that doesn't matter very much arithmetically in most people's fairly short exercise sessions, but can matter in longer workouts and could be major in long, not-intense activity such as long, slow walks.

IMU, none of the stuff about how MFP estimates exercise matters if the exercise calories are coming from a tracker synced to MFP, nor does the gross vs. net problem. The sync algorithms are supposed to take care of that.

With respect to the OP here specifically: It blows my mind that MFP's estimate is higher than your watch's estimate for a strength training session. That makes me wonder if you and we have the same understanding of what you are asking, @mschurmann?

You say "function strength": Are you talking about one of those workouts with fast-paced stuff like battle ropes, tire flipping and whatnot? Those are usually a combination of strength and cardio, and maybe even harder to estimate than pure strength training or more common types of cardio.

MFP's "Strength training (weight lifting, weight training)" calorie estimate, if left at its default calories, is a pretty low METS value. MET estimates are based in part on body weight. Getting 700 calories for one hour of that default MET value would suggest a body weight of 511 pounds. You don't say anything about your weight, @mschurmann, and don't have to, but if you don't weigh 511 pounds, I'm thinking you're not manually logging this in MFP with that database entry, or if you are, it has been changed from the default value. (MFP estimates me at 182 calories for a full hour of strength training, which doesn't seem crazy high to me. That suggests they're using 3 METS as the basis.)

Also, partly because of what I said in that previous paragraph, I'm confused about what you're comparing: Usually, IMU the tracker sync gives a person an all-day overall calorie adjustment based on all movement the tracker sees, compared to the movement that MFP assumed based on "activity level" setting in the MFP profile. With at least some trackers, that adjustment can go up and down over the course of a day because it's making assumptions about the rest of the day, with the actual final reconciliation happening at end of day. At least some trackers will also list exercise sessions with vaguely correct labels, and show the exercise calories associated with those sessions. IMU, those are informational entries, not the thing that's adjusting calorie goal.

On top of that, MFP doesn't use reps and weight in estimating strength training calories. It lets us record reps/weight in the Strength side of the database, but those are just memo entries that don't affect the calorie estimate. If the exercise is manually logged by typing into MFP in the cardiovascular section, the calorie estimate uses the METS estimating method. The cardiovascular section is the only part that will add exercise calories to calorie goal when manually logging exercise into MFP.

Given the last four paragraphs, OP, I'm not sure what to advise, because I'm pretty sure I don't understand the scenario accurately/completely.

I'm going to comment on "Don't eat back all of those regardless of how MFP is designed wrt workout calories. The body doesn't work like that. Eat back about 50% and monitor your progress over weeks.", especially "the body doesn't work like that". That PP is from someone very knowledgeable and helpful, with strong knowledge and background in strength training, but I would infer from various posts less nuanced knowledge about cardiovascular exercise, cardiovascular conditioning, and exercise calorie estimation in that domain. That's not a personal criticism. It isn't bad advice, either, just IMO rather broad brush. It'll work fine for situations with consistent exercise levels, or low exercise levels. It's likely simpler than trying to make the exercise estimates more accurate, too.

In particular, while I agree that calorie compensation from exercise can happen, it's often oversold (as is EPOC). Exercise calorie compensation is fairly clear at a population level, so can look more general than it is. There's quite a range of variability among individuals in the studies I've seen. Looking at it as n=1
(and not just my personal n=1), that variability can matter.

I've seen posts here where people are advised not to do cardio above zone 2, or major exercise calorie compensation will kick in. I've seen other posts here saying that it's essential to switch up exercise routine because our bodies "get used to it" and don't burn many calories from that exercise. Strictly speaking, those ideas involve serious misunderstandings of exercise physiology, IMO and IMU. There are simplistic situations in which they can have some guidance value, but they're not strictly accurate. In some situations, that will matter. (Again, I'm going to skip the explanation of why that's so, because this is already too long.)

There are various ways of estimating exercise calories: METS, which we've mentioned. Heart rate as proxy for oxygen consumption. Formulas for some specific activities, like watts-based estimates for cycling when a good measurement of average watts is available from a power meter or stationary bike. There are maybe others; I've only looked into ones for things I do regularly. Because of the underlying assumptions in each estimating model, some methods will be likely to be more accurate than others for any given exercise type. But it's always an estimate.

Since MFP uses METS estimates, I'm going to include this link, which has a vast list of research-based METS estimates for various activities . . . of varying research quality in individual cases, of course.

https://pacompendium.com/

A person who wants to create personal custom exercises in MFP can use those METS estimates, plus a METS calculator like this one:

https://ergo.human.cornell.edu/MetsCaloriesCalculator/MetsCaloriesCalculator.htm

. . . to get an initial calorie estimate. Once the custom exercise is in My Exercises, the next time it's logged, MFP will use its underlying METS arithmetic to use then-current bodyweight in MFP plus the new session length to estimate exercise calories at the same METS value. If someone wants to, they can use arithmetic to discount the starting METS value to effectively eliminate BMR and even activity level from the estimate. Whether that difference is arithmetically significant in an individual case will vary depending on the demographic determinants of BMR estimate, size of the MFP activity level adjustment, and exercise METS value plus duration.

Retroguy2000

PAV8888 wrote: »

But how on earth are you coming up with (c) above. I don't disagree with your actual remedy. Just with the stated reason: "the body doesn't work like that"

First of all, 50% is just an estimate, let me make that clear. One fundamental issue with eating back 100% is if your calorie estimate for the workout is inflated, then you may be eating counter to the likely goal of losing weight. That number could be inflated by algorithmic error, user error in estimating exercise intensity, more rest times than average during the session, etc. etc. There may be bias in presenting an inflated number to the user too, like all those social media videos, "Do this whatever for 20 minutes and burn this [implausibly high amount] of calories!"

The main point here is it's not 0% and assuming 100% would be foolish if only due to potential tracking error, so 50% is a reasonable middle ground. Then track and adjust over time. After reading all the following, let me ask a different way, where is the science to say that 100% should be eaten back?

1. Jonni Shreve, a former pro bodybuilder I follow on YT, gave a 50% estimate. In his career, he and people like him meticulously track their calories and weight, and add a lot of cardio before shows, and that's his advice to assume 50%.

2. Exercise scientists like Dr. Mike at RP will say it's great to get yourself up to 8K-10K daily steps, but there are diminishing returns above that. According to the MFP model, there are no diminishing returns!

3. Common sense observation, especially after intense exercise. If you aren't used to such things, then go spend an hour playing basketball, or a three hour hill hike, you might spend the next several hours resting more than normal with reduced NEAT. You may have burned 1,000 calories doing that, but is your body even giving you signals that it needs an extra 1,000 calories of food? Probably not.

4. Academic studies exist about the concept of 'energy compensation' and 'constrained energy'. Here's a random one I Google'd:

"this suggests that only 72% of the extra calories we burn from additional activity translates into extra calories burned that day".

https://www.sciencedirect.com/science/article/pii/S0960982221011209

Here's another random example:

"The energy constraint theory indicates that increases in physical activity cause reductions in other components of energy expenditure, for example, “Increasing levels of activity may bring diminishing returns in energy expenditure because of compensatory responses in non–activity energy expenditures""

https://www.sciencedirect.com/science/article/pii/S216183132300217X

I remember seeing another that included data about the Hadza tribe, who walk for tons of miles yet their TDEE is in the range of regular Western fairly sedentary people. According to MFP, they should be adding thousands of calories of food each day.

https://evolutionmedicine.com/2022/03/28/constrained-energy-model/

AnnPT77

I'm very aware of those research findings. Focusing on the averages can mislead; look at the scatter diagrams, look at the standard deviation. There are things that matter at the individual level that aren't included and relevant issues not factored out as confounders.

The model is correct, at the population level. Calorie compensation is a consideration for individuals, yes. The degree of it matters, and there is more that goes into that than body composition.

Even your second link says "The available evidence indicates that in many scenarios, the effect of increasing physical activity on TEE will be mostly additive although some energy appears to “go missing” and is currently unaccounted for."

The Hazda study is interesting, but focused on comparing population-level compensation.

Your point 4 says "If you aren't used to such things, then go spend an hour playing basketball, or a three hour hill hike, you might spend the next several hours resting more than normal with reduced NEAT." That's exactly the flip side of my point: If you are used to such things, you probably won't spend the next several hours resting more than normal. You'll just go on with normal life.

I support the idea that people focused on maximum calorie expenditure, especially exercise beginners, would do well to stay at zone 2 and below to start. But as a generality, with fitness progress, higher zones become less fatiguing than they would have been to start.

If I were to adopt the routine exercise schedule of an elite athlete in my sport, I'd be wiped out - devastated, even. They aren't. Virtually no one here is elite, but there's a continuum, not just endpoints. If I adopted the "nothing above zone 2" or "eat 50%" recommendations, I'd fail to make fitness progress, and under-fuel besides.

I'm probably more sensitive about this, having arrived on MFP as an obese athlete, used to training pretty hard, and competing in a cardiovascular sport. Giving people broad-brush advice about calorie compensation can risk under-fueling for people like that.

There are a lot of exercise beginners here, and they should ideally keep things moderate . . . not just because of calorie compensation. If someone like that's doing so much exercise that they have major, major compensation, there are bigger problems in the picture than the calorie compensation, IMO.

I respect the experts you've quoted, but believe they're focused on their context. That isn't everyone's context. Stating things in terms of diminishing return on steps underscores that, even. Generally, while they "do cardio", they don't train cardio. The situation looks different from the perspective of people for whom substantial cardiovascular exercise load is common and also necessary, in which there's interplay between duration, frequency, and intensity.

Ultimately, yeah, for someone treating exercise calories as an add-on, 100% of actual net exercise calories should be eaten back on top of actual base calories. That's also common sense. The issue is how to get the estimate close to the actual, and close to the net. "Eat 50%" may be good generic advice for people starting out, but it is broad brush.

Retroguy2000

AnnPT77 wrote: »

Calorie compensation is a consideration for individuals, yes. The degree of it matters, and there is more that goes into that than body composition.

Even your second link says "The available evidence indicates that in many scenarios, the effect of increasing physical activity on TEE will be mostly additive although some energy appears to “go missing” and is currently unaccounted for."

Ultimately, yeah, for someone treating exercise calories as an add-on, 100% of actual net exercise calories should be eaten back on top of actual base calories. That's also common sense. The issue is how to get the estimate close to the actual, and close to the net. "Eat 50%" may be good generic advice for people starting out, but it is broad brush.

Every time this conversation comes up it always goes the same way. I present scientific studies and comments from experienced professionals, and people try to poke holes in that but never present any evidence of their own to back up the theory that 100% makes sense.

You've just agreed that compensation is a consideration, and quoted one of those sources saying "mostly additive" as if that's a rebuttal, but "mostly additive" is not the same as "fully additive".

Are the bodybuilders and exercise scientists saying those things wrong?

I understand your experience with adding 100%, but all that tells me is that the base activity level you gave MFP + the exercise you added amounted to the weight change you expected to find. That's not evidence of the latter being correct. It could be evidence of the former being wrong. You could be between two of the MFP categories (sedentary, lightly active, very active, etc.). You're probably not exactly at 1.4*BMR or 1.6*BMR or whatever. You might be 1.42*, or 1.53*, and so on.

I know I personally can't use MFP as intended because it's way off, for me. I'm definitely very sedentary aside from additional exercise, which is relatively easy to track being simply walking and lifting. Neither of which amount to a huge amount of additional calories. And yet MFP's estimate for me with "sedentary" and "maintenance" is about 750 below my actual estimated TDEE, arrived at by tracking daily calories and weight changes for a long time. There is no way that 45 mins walking and/or 60 mins lifting per day is an average 750 calories per day. Maybe if I was nearly double the weight I actually am.

So I maintain my own calorie target and track calories and weight change.

If someone is under-fueling they'll rapidly find themselves low on energy, workout performance suffering, faster weight loss than expected. They can then adjust as needed. It may be that they've estimated their base daily activity wrongly.

AnnPT77

Retroguy2000 wrote: »

AnnPT77 wrote: »

Calorie compensation is a consideration for individuals, yes. The degree of it matters, and there is more that goes into that than body composition.

Even your second link says "The available evidence indicates that in many scenarios, the effect of increasing physical activity on TEE will be mostly additive although some energy appears to “go missing” and is currently unaccounted for."

Ultimately, yeah, for someone treating exercise calories as an add-on, 100% of actual net exercise calories should be eaten back on top of actual base calories. That's also common sense. The issue is how to get the estimate close to the actual, and close to the net. "Eat 50%" may be good generic advice for people starting out, but it is broad brush.

Every time this conversation comes up it always goes the same way. I present scientific studies and comments from experienced professionals, and people try to poke holes in that but never present any evidence of their own to back up the theory that 100% makes sense.

Again, the studies are presenting trends, but clearly show meaningful variability among the individuals included in the studies. We are individuals. Knowing the trends is important and useful, but not complete.

I'm not going to quote experienced professionals, because my sources aren't internet-able. They're things like the education for coaching certification in my sport, my personal coaches some of whom are now friends (including coaches of high-achieving NCAA Div I and Div III programs, former national team members, Olympic medalists . . . because I've been extraordinarily lucky to get high-level coaching even as a fairly under-impressive regular duffer older athlete).

The focus is not usually losing weight, but rather performance, fitness improvement, overall nutrition and avoiding weight loss, but calorie issues are absolutely relevant there. Generically, under-eating is a common risk in the sport.

I know not naming people limits my credibility, but I don't mention friends' names online without their permission.

You've just agreed that compensation is a consideration, and quoted one of those sources saying "mostly additive" as if that's a rebuttal, but "mostly additive" is not the same as "fully additive".

It's not intended as a rebuttal. This isn't black and white, which "rebuttal" would suggest. It's underscoring my point that it's not necessarily as simple as applying a universal discount percent to all exercise calorie estimates.

There would also seem to be a gap between "mostly additive" and "50%" for all. (I do understand that the 50% is intended to adjust for more than calorie compensation. There are other ways to adjust for things other than calorie compensation, if a person wants to pursue that.)

Are the bodybuilders and exercise scientists saying those things wrong?

No, not wrong, I have repeatedly said that. I said "broad brush". Bodybuilders specifically look at calorie consumption differently than athletes/coaches in intensely cardiovascular sports. Fueling is only one dimension relevant to exercise load. It was weird to me when I first joined MFP to see calorie expenditure presented as pretty much the only relevant thing that would determine appropriate exercise load. I get why that's a bias here, but that doesn't make it not a bias.

I suspect, but can't prove, that the bodybuilding community probably looks at it as mostly about health, body composition, strength progress, and calorie expenditure. If so, that's not wrong either. It's just a different perspective.

For some purposes, "broad brush" is fine, and avoids complexity.

I understand your experience with adding 100%, but all that tells me is that the base activity level you gave MFP + the exercise you added amounted to the weight change you expected to find. That's not evidence of the latter being correct. It could be evidence of the former being wrong. You could be between two of the MFP categories (sedentary, lightly active, very active, etc.). You're probably not exactly at 1.4*BMR or 1.6*BMR or whatever. You might be 1.42*, or 1.53*, and so on.

I haven't used MFP's activity levels since the first few weeks of logging, back in 2015. I'm like you in that respect.

MFP's estimates, plus estimates from my good brand/model fitness tracker that estimates well for others, are way off for me, like 25-30%. It's more like 500 calories in my case, but I assume I'm quite a bit smaller than you are, and presumably with a smaller calorie budget.

All I'm saying is that there are ways to make exercise calorie estimates more careful, close to accurate, etc. Individual variability matters in that, but so does awareness of potential issues like calorie compensation.

I'd also observe that experience under significantly varying exercise load sheds some light on how likely it is whether the estimating issues are from base calories or exercise calories. We usually tell people to test a given calorie level for 4-6 weeks to get a reasonable reading on its accuracy. Over the nearly 10 years I've been logging, I've had periods that long with zero intentional exercise (one recent, though that one had complications for calorie estimating ).

If people want to broad-brush SWAG exercise estimates, that's fine, but I think it's legit to check that they know they're doing that. It'll matter to some people, not others.

I know I personally can't use MFP as intended because it's way off, for me. I'm definitely very sedentary aside from additional exercise, which is relatively easy to track being simply walking and lifting. Neither of which amount to a huge amount of additional calories. And yet MFP's estimate for me with "sedentary" and "maintenance" is about 750 below my actual estimated TDEE, arrived at by tracking daily calories and weight changes for a long time. There is no way that 45 mins walking and/or 60 mins lifting per day is an average 750 calories per day. Maybe if I was nearly double the weight I actually am.

Of course base calorie estimates can be inaccurate, too. Subjectively, it seems to me like people here are more likely to blame the exercise estimates than the base estimates, even though either or both can have issues. I'm acknowledging calorie compensation and things like gross vs. net as among the issues for exercise estimating. I'm acknowledging that some METS estimates are based on sub-ideal research. I'm acknowledging that METS is a wacky way to estimate certain exercise types, and heart rate is a wacky way to estimate others.

My main point is that it's possible to zone in on better accuracy - not perfect accuracy - and that there's more nuance to it. That may not matter to someone with very consistent exercise, or some basic exercise types, or who prefers not to go further than "log 50% then adjust". I think the "test then adjust" idea works better for base calories than exercise calories, at least in cases of quite variable exercise load.

So I maintain my own calorie target and track calories and weight change.

If someone is under-fueling they'll rapidly find themselves low on energy, workout performance suffering, faster weight loss than expected. They can then adjust as needed. It may be that they've estimated their base daily activity wrongly.

Sure. But better estimating can help avoid that, on both the base calories and exercise calories side of it, and individual factors are relevant in both cases.

Again, all I'm saying is that telling people things like "nothing over zone 2" or "eat back 50%" is leaving out some relevant factors, and that those will matter to some people . . . probably a minority, here, sure.

age_is_just_a_number

My two cents:
To lose weight you need to be at a caloric deficit. Focus on what and how much you are eating
Exercise is for maintaining and building muscle mass.

Both calories consumed and calories expended are estimates. No two people are the same. You need to track your food intake and results (such as weight loss) and evaluate whether your results are going in the direction you want. If not, you need to reevaluate and adjust your intake.

Take care