tl;dr
Adding exercise barely increases your total cArOliEs oUt.
If it does at all, less than expected, and the effect diminishes over time.
The body cannot magically conjure up more cArOliEs if you go jogging, it just takes the energy from somewhere else. Just like spending money doesn’t increase your income, it just re-routes existing expenditures.
This is what actual measurements show, everything prior was total speculation.
This explains why the “move more” part of “eat less, move more” is garbage.
Unfortunately, the rest of the 300-page book is fluff or useless mainstream cAroLieS & ulTRa prOcesSed fOoD nonsense.
Pontzer spends a whopping 5 pages (1.6% of the book) regaling us with a story about how, this one time in grad school, he got totally wasted and had a hangover and puked. Many such cases.
Disproves “exercise → fat loss” but doesn’t add anything else.
Pontzer still thinks you should exercise for health; just don’t expect any fat loss.
Doubly-labeled water, the Gold Standard
Herman Pontzer is a metabolic researcher. He’s made quite the splash in the last few years. One of his specialties is using “doubly-labeled water” to measure the ACTUAL caloric burn of people (and animals) in their daily lives.
It’s a pretty elegant method, which works by giving you special “labeled” water to drink and measuring how slow or fast it is secreted out of your body.
“Labeled” here just means it’s a special version of a common molecule, usually deuterium (2H) for hydrogen and an oxygen molecule called 18O for oxygen.
Remember, water is H2O, and both the H and O molecules can be “labeled” this way. Hence “doubly-labeled” water.
The whole trick of DLW relies on the fact that hydrogen and oxygen leave your body at different speeds, depending on how active your metabolism is.
Here’s how it’s done:
You measure the baseline ratio of these special “labeled” molecules in your urine.
You drink some 2H/18O “labeled” water and let it spread throughout your body a few hours.
You measure the new ratio of the 2 labeled molecules, which will now be higher in 2H and 18O.
Your body continues burning cArOliEs which uses oxygen and produces CO2
Over time, the ratio will normalize back to your baseline as more and more labeled oxygen and hydrogen make their way through your body, being replaced with normal molecules from the (unlabeled) water you take in.
A few days later, you measure the ratio again. The labeled oxygen will have been excreted via both CO2 (breath) and H2O (water), notice how they both have Os. The labeled hydrogen will only have been excreted through H2O (urine/sweat/..) since CO2 has no H in it. Therefore the ratio will have changed depending on how many cArOliEs you “burned” in that period.
What’s so great about doubly-labeled water?
Pontzer goes into the history of metabolic research in the book. Prior to using DLW, researchers couldn’t ACTUALLY measure the caloric burn of free, wild, living animals or humans.
Indirect CO2 Calorimetry
They could make you sit down on a couch, measure your breath CO2, and calculate your “resting metabolic rate” or “basal metabolic rate.” This is what I’ve previously measured and my RMR came out to be around 2,300kcal/day.
It’s a pretty old and cheap technology, I pay less than $100 at a local fitness place to have it done.
The limitation is, of course, that you can only measure what a person will burn sitting on a couch. You can’t exactly run a marathon, or go to the gym, or anything else while breathing into a tube on a couch.
Metabolic Chamber
This is essentially the same thing scaled up: they make people (or animals) live in a air-tight chamber that measures the CO2 going in and out. It’s therefore much easier to do for longer periods, as you just have to stay in a room vs. being confined to a couch, breathing through a tube.
You can do way more activities this way: run on a treadmill, do burpees or pull-ups, do your (desk) job…
But you still can’t do all jobs (welding? climbing telephone poles?) and it’s also INSANELY expensive. I’ve read it’s several hundred thousand dollars per person to do this, since these metabolic chambers are very expensive equipment and extremely rare.
And you can’t measure the metabolic rate of a hunter-gatherer living in the wild while locking him in a small room.
DLW to the rescue!
DLW solves this problem, since you don’t need to measure the subject’s breath at all. You just need 2 or 3 urine samples, and you need to give them a drink of “labeled” water.
Therefore they don’t need to be constrained to a couch, or a metabolic chamber, or at all. They just need to drink some water and give you urine samples.
That means you can do it in the wild, in Africa, in living real-world hunter-gatherers.
The only problem, historically, was the cost: creating these special “labeled” water drinks was enormously expensive. It used to cost hundreds of thousands to create enough 2H and 18O for one person.
But the original technology is from the 70s, and has come down in price steadily since then. It’s recently become cheap enough that it’s being offered to consumers, at home, for sub $1,000. For example there’s a company called Calorify which offers this service (spoiler alert).
The dramatic decrease in price made it possible for people like John Speakman and Herman Pontzer to do hundreds, if not thousands of these tests.
They went nuts. They tested the Hadza hunter-gatherer tribe in Tanzania. They tested elite ultra-marathoners and endurance bicyclists and rowers. They tested military troops on arctic expeditions. They tested pregnant women. They tested every animal they could find that would give them a urine sample.
And they put all of this “actual caloric burn” data into a giant database to compare.
That’s basically what this book, Burn, is about.
Exercise doesn’t burn carolies
The main finding, and the real contribution Pontzer has made with his research, is this:
Increasing the amount you exercise typically does not significantly increase how many calories your body burns.
Or, restated:
Exercise does not burn extra calories.
Or, as in my provocative headline:
Burning calories does not burn any calories.
Additive Model (also called Factorial Model)
Pontzer explains the history once more. Previously, people were only able to measure the Basal Metabolic Rate (the couch thing). Let’s take mine at 2,300kcal/day. This is completely at rest, what I’d burn if I was so sedentary I never even walked to the fridge to grab a pint of delicious heavy cream.
But almost nobody lives like that, except coma patients. So what did we do? Before the advent of our hero doubly-labeled water, we simply added the calculated “caloric burn” of any given amount of exercise on top of the “basal” metabolic rate.
Turns out, that’s not how it works.
Constrained Model
What Pontzer discovered, his main point in the book: when you decide to spend some extra energy on exercise, it doesn’t typically increase your total burn (additive model). It is taken from somewhere else in your energy budget, like the immune system or reproductive function.
Hence, the total energy you can burn in any given day is “constrained.” You can shift more energy from one area of your metabolism to exercise, but that won’t increase overall energy expenditure much.
Here’s Pontzer’s paper describing the Constrained Energy Model.
He likens the body to a business managing its finances. The business cannot just decide to go out and invest in more stuff or hire more workers. If it doesn’t have enough “money in” to pay for an increase in its “money out,” then it needs to cut back on something else, like firing some existing workers, or selling off some old equipment.
We now have 3-5 decades of doubly-labeled water under our belts, and the data is in, says Pontzer: exercise doesn’t burn any extra calories in almost anyone, and almost nobody loses fat doing it. Especially not over the long run.
He is quite adamant about this, so I’ll just throw some quotes from the book at you.
This is about the Hadza hunter-gatherers, whom much of the book is about. They walk for 5+ hours every day, including hunting with bows & arrows and climbing trees for honey:
Everyone knew that the Hadza had exceptionally high energy expenditures because they were so physically active.
Except they didn’t. [..] Hadza men and women were burning the same amount of energy each day as men and women in the United States, England, the Netherlands, Japan, Russia. Somehow, the Hadza, who get more physical activity in a day than the typical American gets in a week, were nonetheless burning the same number of calories as everyone else.
And here, a study from the Netherlands, in which totally untrained people were put on a more and more intense running schedule for 40 weeks, ending up running 25 miles per week:
In the beginning, subjects were running 20 minutes per day, 4 days per week. By the end, sessions were 60 minutes long, and subjects were running roughly 25 miles per week. [..]
If the factorial model were correct, we’d expect their daily energy expenditures to be at least 360kcal/day higher by the end of the study [..].
Instead, at week 40, their daily energy expenditure was only about 120kcal higher. These women went from never exercising to running 25 miles per week, [..] and their daily energy expenditure was essentially the same as when they started. Men in the study showed similar results.
Animals are the same:
And it’s not just humans. Constrained daily energy expenditure seems to be the rule among warm-blooded animals. [..] Again and again we see the same result: daily energy expenditure doesn’t change even as the animals work harder and harder. [..]
Just as the constrained model would predict, we’ve found that zoo-living primates have the same daily energy expenditures as those in the wild. Same goes for kangaroos and pandas. Each species maintains its evolved metabolic rate whether it’s struggling for survival in the jungle or chilling out at the zoo; lifestyle has little effect.
About the Midwest Exercise Trial I:
Young adults who were sedentary and overweight were randomly assigned to either an exercise group or a control group. Exercisers worked their way up to doing about 2,000kcal of exercise every week for sixteen months. At 2,000kcal/wk for sixteen months, the exercisers should have lost 40lbs. Instead, the men lost 10lbs [..]. If that sounds grim, consider the women in the exercise group: they lost nothing.
They tried again in the Midwest Exercise Trial II, making them work out more:
Men and women were assigned to either 2,000 or 3,000kcal of supervised exercise per week. That’s an incredible amount of exercise, equivalent to running 20 or 30 miles per week for a 150lbs person. [..]
Average weight loss was around ten pounds, [..] far less than we’d expect from so much exercise. And there was no difference in average weight loss between the 2,000 and 3,000kcal/wk exercise groups [..].
Even more surprising, for 34 of the 74 men and women who completed the study, average weight loss was zero. These poor souls, labeled “non-responders,” exercised like mad, even managed to push their daily energy expenditure up a bit, and still lost nothing.
His prediction for exercise to lose fat is dire:
If you start a new exercise program tomorrow and stick to it religiously, you will most likely weigh nearly the same in two years as you do right now.
You should still do it! You’ll be happier, healthier, and live longer. Just don’t expect any meaningful weight change in the long term from exercise alone.
What caused diabesityheimer? Not sloth.
The global obesity epidemic cannot be a problem of energy expenditure. For one thing, as we see with the Hadza, daily energy expenditures are the same today in the industrialized world as they were in our hunter-gatherer past.
Just to hammer the point home:
Hadza men and women were far more physically active than people in the United States and Europe, yet they burned the same number of calories.
About a study done in Indian factory workers back in 1956:
Pencil-pushing clerks weighed the same as hardworking coalmen.
Jesus, we get it:
For the vast majority of people, physical activity and the energy it burns each day has no effect on weight.
You’re always such a Debbie Downer, Herman:
The danger, though, in selling exercise as a way to lose weight is that it doesn’t work.
The impact of the Constrained Energy Model
In short, science has finally proven what almost every fat person ever alive has always known: exercise doesn’t make you lose fat.
It almost never does, it almost never has, and with the exception of some extreme outliers, there’s no recorded evidence of it doing so.
When people tout CICO, the unfalsifiable fIrSt LaW oF tHermOdyNamIcs, or whatever else, they are simply wrong about your ability to influence the CO part.
You can somewhat influence your calories out, but you cannot simply “do 500kcal worth of exercise” to increase it by 500kcal. This is nonsense and science has now conclusively disproven it.
Exceptions: when exercise DOES burn carolies
Pontzer spends a whole chapter outlining the exceptions to this rule, which are on both extremes on the spectrum:
Going from completely sedentary to even a “walking around the office” level of activity seems to make a huge difference. So if you’re not even walking at all and just driving around Wal-Mart with your little scooter, going from 0 exercise to even a little bit might help tremendously.
Elite athletes like ultra-marathon runners or Olympic swimmers like Michael Phelps are able to push their metabolisms significantly higher than normal people doing normal exercise. But these people often train 5h/day as a full time job, and it’s questionable if everyone could achieve this, or if these sports are simply selecting for genetic outliers with a phenomenally strong metabolism.
Anecdotally, a good friend of mine has lost a tremendous amount of weight mostly by running. But when I asked him about this while reading the book, he merely laughed at the running regiment outlined above (20-25 miles per week). He says it never helped him lose weight below 30 miles a week when he was near 300lbs, and at times, he was running 100mi per week. That’s 14mi of running EVERY DAY, with no days off. At an average pace of 8-9mi per hour, that’s 1.5-2 hours every day, 7 days a week.
Clearly, this is far beyond most “just move more!” recommendations, and not exactly practical for many people.
Plus, the weight started coming back on as soon as he stopped running these crazy distances every week.
Say the line, Herman!
Unfortunately, despite having proven that the CO path of CICO is useless as a lever for fat loss, Pontzer is still a CICOphant:
It is absolutely and inescapably correct that obesity results from eating more calories than we burn. There is no other way to gain weight.
Of course, this is dumb as rocks. Calories are “causing” weight gain like miles cause travel. It’s an accounting tautology, and if you think those are causal, you need to go back to science school and learn what “causality” means.
No such thing as an honest CICOpath
This is a saying I have. No proponent of CICO I’ve ever met has been able to honestly talk about it. Typically, they will hold the impossible position that CICO is both axiomatic and unfalsifiable, an accounting tautology - but also scientific (which would mean falsifiable) and actionable, and instructive on how to lose weight (e.g. “eat less and move more.”)
But when you point this out, they begin attacking you or complaining about “semantics” (you know, what things mean. Seems important when.. talking).
That’s why I maintain there’s no such thing as an honest CICOpath. Everyone who’s intellectually curious & honest at least concedes that CICO allows for myriad complications as to what actually should and should not be done to lose fat.
That said, Pontzer is particularly annoying because he says a lot of dumb things very confidently and in a cocky way. A Stephan Guyenet is more “serious” and measured, so when he says nonsense, it’s typically couched and hedged better.
Here are some excerpts that really ground my gears:
In its purest form, the argument that calories don’t make you fat makes as much sense as the argument that money doesn’t make you rich.
Yes, of course. Money DOES NOT make you rich. Money is a measure of your riches. A man on a lonely island, in possession of $1B, is not rich, he is poor. Money is only coveted because we can exchange it for things. (In fact, that’s pretty much the definition of money.)
It’s also “answering” the question on an entirely wrong level. It reminds me of an old Steven Wright joke:
Steven gets pulled over. The cop asks: why were you going so fast? Steven points towards the gas pedal. “See this? I press it all the way to the floor. Makes more gas go into the engine. Engine runs faster.”
If you told me that any method of getting rich worked as long as it put dollars into your bank account, I would think you’re an idiot who misunderstood the question.
Duh, dollars entering your bank account is the definition of getting richer. But should I go to medical school? Should I start a dry cleaning business? Bro, I’m serious here. No dumb tautologies, please.
Pontzer continues to pontzificate on cAroLiEs:
Energy balance is the only thing that alters your weight. That’s the inescapable reality of physics.
Maybe the terms should give it away. Balance is an accounting term, not a science term. Does the balance in your checking account alter your wealth? No! It’s a measure of your wealth.
I simply cannot begin to understand how someone can earn a PhD and continue to believe such fundamentally nonsensical crap. It really is a cult.
Rather than believing that low-carb diets are magic and allow us to break the laws of nature [..some dumb shit..]
Rather than believing caRoliEs are magic and allow us to somehow reverse causality, turning accounting tautologies into actual mechanisms, nutrition scientists should stop drooling over Kevin Hall and other CICOphants. Makes you think the guy has dirt on all these young scientists. Maybe that 5-page diatribe on getting drunk was a cry for help? Blink twice if you’re under duress, Herman.
Here’s a tip: if you describe my position as “magic” then I’ll think you’re arguing in bad faith.
Instead of suggesting his own solutions to obesity, Pontzer takes a page (literally) out of Stephan Guyenet’s book:
In short, our modern diets are too delicious.
Aww, shucks. Don’t tell an old Italian granny that, or people who have been to France, or many of the other countries with food way more delicious than ours. If anything, the most obese countries don’t have an identifiable cuisine, and certainly not one most people would call “delicious.” You see many Australian or American or British restaurants in Paris? Nobody likes these foods except for nostalgic expats.
People say many things about McDonald’s and The Cheesecake Factory, but “too delicious?”
One obvious strategy to manage our weight and maintain good metabolic health is to build our diet around foods that are filling and nutrient rich without packing in a lot of calories.
One wonders, if this strategy is so obvious, why nobody’s ever tried it. FFS.
I could go on with these quotes, but I think you get the point. Despite having ruled out the CO half of the magical accounting tautology, Pontzer has no answers at all.
The infuriating thing: he even cites the Kevin Hall study on Biggest Loser contestants, where he showed that the crazy caloric restriction had ruined their metabolisms long-term, and almost all of them had gained the weight back, some overshooting their original weights.
So Pontzer just proved to us that “move more” is bullshit, and he knows from other research that “eating less” doesn’t work.
And yet he loudly proclaims these non-solutions. I think that’s pretty incoherent. If the mainstream nutrition science answer was “Bro, we have no clue how to do it, but we know exercise and eating less aren’t it” that’d be one thing. But they show us that neither of these work, and then still basically tell us to do them.
Pontzer & Paleo/Keto/Carnivore
Like so many mainstream nutrition bros, Pontzer has a bizarre fetish hating on paleo, keto, and carnivore. Confusingly, he is mixing them all up and burning (heh) down strawman after strawman.
For example, he claims Paleo is BS because it doesn’t allow for any carbs and is low-carb. This is, of course, nonsense. Paleo is not at all low-carb.
I have my own beef (heh) with Paleo, but when I did Paleo in the 2010 era, people like Robb Wolff were talking about yams & sweet potatoes & similar “ancestral tubers” all the time. That’s exactly what Pontzer brings up as counter argument to “Paleo.”
I suppose you can argue that keto isn’t paleo, but yea, no shit. I don’t think many keto people are claiming that paleo man was constantly in deep ketosis and never ate any plants.
Like most mainstream critique of the more enthusiast/out there nutrition paradigms, Pontzer’s is pretty sloppy.
The problem [..with Paleo..] is that it’s hard to tell what hominin ancestors ate with any real precision. The evidence is hard to come by and doesn’t usually tell you what you really want to know: What was on a typical week’s menu for Paleolithic humans?
I mean, ok, fair. But we do know certain things that were almost certainly not on the menu: all the things invented relatively recently. Which is pretty much what Paleo says.
Paleo diet evangelists have distinguished themselves by projecting a hard-nosed, steely-eyed view of human nature and evolution. Humans, they assure us, have evolved to eat meat, bro. They push high-fat, low-carb diets that send the body into ketogenesis, arguing that our ancestral diet was all bison and no berries.
Here he’s confusing 3 different diets: Paleo, keto, and carnivore. Paleo isn’t necessarily low-carb, carnivore’s not necessarily particularly ketogenic, and I don’t know if any keto advocate is claiming that Paleo man was drinking heavy cream and eating tallow with a spoon.
It’s easy to take down strawmen if you don’t understand how to build them well.
Paleo might be anti-grain, but it certainly isn’t anti all carbs. C’mon, Herman. Do better.
Still, it’s a common misconception among many in the Paleo crowd that our hunter-gatherer ancestors were somehow only hunting.
Let’s check how Robb Wolf, famous Paleo guru, defines the diet:
I suppose Robb thinks that we were hunting all those fruits, vegetables, nuts & seeds? A full 50% of the “Okay to eat” category is rooted to the ground and can hardly be “hunted.”
Let’s check in with Mark Sisson of “Primal” (=Paleo under his own brand name) fame.
Primal Blueprint Law 1: Eat Plants and Animals
Fill your plate with high-quality meat (fish, fowl, and all types of meat), colorful vegetables and fruit, and fats we love like avocado oil, olive oil, nuts and seeds, and dairy (if tolerated).
(From: https://www.primalkitchen.com/blogs/recipes/what-is-the-primal-blueprint)
Again: meat, fish & fowl make an appearance for sure. But also colorful vegetables, fruit, avocados, olives, nuts & seeds. Can dairy be “hunted?”
Pontzer has more:
Archaeological excavations in Jordan have recently uncovered an ancient oven and charred bread remnants dated to over 14,000 years ago, thousands of years before the emergence of agriculture.
Super, Herman. So maybe the first bread was made 14,000 years ago instead of 12,000 years ago or whatever Paleo estimated.
Need I remind you that the Paleolithic period started 3 million years ago? 14,000 years are .4% of that time period. And it’s the last .4%.
But he also says some things about Paleo that I agree with, and that maybe Paleo gurus would agree with:
All of the populations [..in the book..] are equally natural, and as far as we can tell, all of those populations were equally healthy, despite the fact that their diets ran the gamut from mostly plants to mostly meat. Humans can be healthy eating a broad range of diets, and have done so in the past. There is no single Paleo diet.
I would largely agree. Or maybe it’s better to say: while certain foods probably weren’t super common during the Paleolithic, there’s clearly a more recent 10,000 year period in which people didn’t have diabesityheimer. Maybe they weren’t all 6’5 chads with abs like our hunter-gatherer ancestors, but wouldn’t it be great to get back to no obesity, no diabetes, no stroke, no heart disease?
Aliens.gif
But something - or, more likely, several things- about our modern environment is causing the hypothalamus to misfire, leading us to consume more calories than we expend.
So close, Herman. It’s almost like something entered our food supply relatively recently, maybe around the 1850s, slowly taking up more and more market share, displacing all those other cArOliEs we used to eat. And this hypothetical substance would cause our bodies to somehow “misfire” and “miscalibrate” and even all the exercise in the world can’t beat it for most people.
I suppose it could be anything.
As Kevin Hall’s work has shown, diets dominated by processed foods lead to overeating and weight gain.
Hm, I sure wonder what it is about “processed foods” that- shut up, you say? Ok, I’ll shut up.
Should you read Burn?
Part of me wants to say no. Besides the fascinating Constrained Energy Model he introduces, there’s nothing new in there. His overall views on how to lose fat are… not even wrong. Typical mainstream cAroliEs nonsense. Eat fibrous, low-carolie foods. Avoid evil ultra-processed foods. Fill up on protein. More of the same nonsense that hasn’t worked for anyone.
But part of me really likes how he hammers home the “exercise doesn’t increase calories out” part. I kind of knew that, having read his paper and witnessed discussions on the topic. But he really hits it at another level, spending about 1/3 of the book to set this topic up, and give example after example of what they tested, and how they confirmed the findings. If you read my summary and thought “Ok, but what about..” then he probably tested that 10 years ago and addresses it in the book.
Makes you realize this isn’t just 1 guy and 1 study, this is the entire field of metabolic research for the last 30 some years.
And I will say Burn makes for excellent tinder: I’ve started several fires with pages from it. Title checks out.
Sam Bankman Fried was a profoundly poor decision maker, but he said one thing that I keep seeing more and more evidence for:
"I don’t want to say no book is ever worth reading, but I actually do believe something pretty close to that. … If you wrote a book, you f---ed up, and it should have been a six-paragraph blog post."
The more I read, especially non-fiction, the more I realize most most books are filler and once you understand the thesis you should stop before wasting any more time. Books also make it very difficult to have a conversation because you really can't reference them, and can't simply say "read this book, it explains everything" because no one is going to do that.
So anyway I'd like to thank you for reading this so I don't have to, and pointing to the actual science and evidence.
The problem is that everything burns calories, including not burning calories. Being too big burns calories. Eating more burns calories. Drinking coffee burns calories. Drinking a frapp from starbies burns calories. Going in the sauna burns calories. Cold plunging burns calories. Shitting yourself and falling down the stairs burns calories.
The question isn't "does this burn calories?" - if it were, the meme version of Ray Peat would be right and no functional limit on daily orange juice intake would apply. Drink more, lose more.
The question, instead, is "what effects does this have on the rest of the system, which if you' re reading this probably has dozens of broken homeostatic mechanisms?"
So, for instance, an example beloved of this blog would be "eating whipped cream burns calories". This is true in the short term (in the same way that drinking orange juice is), but is also (apparently) true in the long term -- many good things happen downstream from the cream (that rhymes, a motto waiting to happen; say it in Macho Man Randy Savage's voice and see for yourself). Insulin sensitivity, uncoupling, satiety, lipolysis, whatever - - they're all downstream from the cream, brother.
So let's do exercise now. Pontzer in the book says that one of the things that westerners' bodies are doing with the energy they aren't burning walking around all day is 'inflammatory processes'. These are known to decrease insulin sensitivity. They're also known to be decreased by exercise (in mice and in men). Every single study showing the uselessness of exercise for fat people starts with folks packed fulla PUFA, and forces them to try to empty out their ocean of inflammation with a little pail. Now, if I said insulin sensitivity 'burns calories' when it's downstream from the cream, why not when it's wise with the 'cize? If inflammation (and the resulting diabetes) go down, what's going to happen to obesity? Well, nothing if a person keeps eating all the poison that made him fat in the first place. Can't outrun a bad diet. But in a low-carb healthy-fat context, the exercise is going to be involved in making changes that result in fat loss. That's calories no longer stored in the body, not because they were 'burned' while on the treadmill, but because of downstream effects. Unless, of course, we're claiming all these downstream effects are only good when caused by cream, not by exercise.
Run a crossover on ad-lib high-fat low-protein keto, with 'hours of zone two cardio daily' as the experimental arm, and something good will happen. Hell, it might break through your summer plateau with no other changes needed.