Hacker News new | comments | ask | show | jobs | submit login
The Hidden Truths about Calories (scientificamerican.com)
239 points by jamesbritt on Aug 27, 2012 | hide | past | web | favorite | 182 comments

If this is at all interesting to you, I highly recommend Gary Taubes' Why We Get Fat and What to Do About it (http://www.amazon.com/Why-We-Get-Fat-About/dp/0307272702)

which is a more accessible version of his more rigorous book Good Calories, Bad calories.

I've been on a bit of a health binge since the beginning of the year and have been doing lots of research into these things. In a nutshell, not all calories are the same the some can wreak havoc on your system (grains, it turns out, aren't that good for us). The Paleo Solution, by Robb Wolf, is another great book - he goes to great lengths to discuss the science and biochemistry behind the points he is making. Highly recommend.

I've read GCBC and initially found it quite persuasive. But it seems that Taubes isn't really correct. The best refutation I've seen is from obesity researcher Stephan Guyenet.

From http://wholehealthsource.blogspot.com/2011/08/carbohydrate-h...

I hope you can see by now that the carbohydrate hypothesis of obesity is not only incorrect on a number of levels, but it may even be backward.  The reason why obesity and metabolism researchers tend not to take this idea seriously is that it is contradicted by a large body of evidence from multiple fields.  I understand that people like ideas that "challenge conventional wisdom", but the fact is that obesity is a complex state and it will not be shoehorned into simplistic hypotheses.

Carbohydrate consumption per se is not behind the obesity epidemic.  However, once overweight or obesity is established, carbohydrate restriction can aid fat loss in some people.  The mechanism by which this occurs is not totally clear, but there is no evidence that insulin plays a causal role in this process.  Carbohydrate restriction spontaneously reduces calorie intake (as does fat restriction), suggesting the possibility that it alters body fat homeostasis, but this alteration likely occurs in the brain, not in the fat tissue itself.  The brain is the primary homeostatic regulator of fat mass, just as it homeostatically regulates blood pressure, breathing rate, and body temperature.  This has been suspected since the early brain lesion studies of the 1940s (47) and even before, and the discovery of leptin in 1994 cemented leptin's role as the main player in body fat homeostasis.  In some cases, the setpoint around which the body defends these variables can be changed (e.g., hypertension, fever, and obesity).  Research is ongoing to understand how this process works.

Many obesity researchers find Guyenet to be quite the joke.

For example, Peter over at HyperLipid recently ripped him to utter shreds:



Not to mention, Taubes countered Guyenet with a 6-part series refuting Guyenet's widely mocked food-reward/palatability hypothesis:


The thing is, there's a lot of infighting; These guys are always going back and forth. They both have their camps. Those both have their arsenal of studies. It's almost a sport...

The simple fact of the matter is that we're not really going to actually know conclusively and incontrovertibly one way or the other for at least another decade.

I'd just like to say that my athletic performance, body composition and mental clarity have been far, far better on paleo than any other diet. I'd also recommend the two books gxs mentioned.

Of course, not all digestive systems are created equal and YMMV. For instance, wasn't until after eating paleo for a while I discovered I had a mild gluten and dairy intolerance (most people develop the latter).

Also, Zooko and Amber Wilcox-O'Hearn are reviewing the science related to ketosis and diet here:


Another book recommendation:-

Girth Control: The Science of Fat Loss & Muscle Gain - Alan Aragon

This is the bible, it has absolutely no fluff/filler content.

Taubes isn't taken seriously by most dieticians and nutrition researchers we talk to (I work at a nutrition startup).

I'd say that if there is any value to Taubes, it is this:

Different foods, given the same calories, will fill you up or nourish you at varying degrees.

300 calories of oreo cookies isn't the same as eating a banana and some chicken.

The main practical benefit people get from low carb, paleo, etc is that they're eating less sugar and junk, and thus eating healthy food with better overall satiety. As a result, they lose weight because they're eating fewer calories.

You can lose weight eating twinkies, it's just hard to do because it's bad for your health, not nutritious, and leaves you feeling hungry because it's all carbs.

A more scientific and less paleo-obsessed way to think about carbohydrates is this:

Carbs are fuel, if you're an athlete, you have a use for them. If you live a sedentary lifestyle, excess fuel isn't going to do you any favors.

Adjust accordingly.

In the end: calories in, calories out. Humanity hasn't defeated the laws of thermodynamics. It's just a question of how easy and healthy it is to maintain a sub-maintenance caloric intake.


Okay people, I just got done saying low carb made sense if you weren't an athlete, what exactly are you arguing against? I agree with you guys, I just think the reasoning and rationale should be taken in the context of current nutrition research and not some demagogue's book that is obsessed with a dietary aesthetic rather than data.

The reason people are arguing against you is that even in 2012 on HN in the face of currently incompletely understood complexity people instantly start forming micro-religions complete with prophets and various 'higher-powers', and of course heretics and infidels.

I mean seriously this article pretty much says "wow, this stuff is complex and really hard to make sense of", and the top comment is "but I have found the truth in this book!"

I fully agree with you, I think the paleo approach has some pretty sound ideas. But when I hear people talk about how "evolution" commands we shall eat no grains I get really worried about how quickly educated people can turn science into magic.

>>but I have found the truth in this book!

Really? My comment pointed towards two books, and a follow up argument suggested Feinmans old well known article for kicks.

What exactly is the point you are trying to make? By contrast, your comment lacks any real substance other than accusing me inciting religious war and telling us about your burden having to worry that educated people turn science into magic.

Bullshit. Where are your articles? Please, point to the articles that show that a diet abundant in grain from carbohydrates is ok, let alone good for you.

You seem to be responding to claims the poster didn't make. Calm down and read the post again, without assuming an attack.

I am not sure how you can read accusation of religiously motivated blindness/idolatry as anything but an attack really, it isn't a well-reasoned argument against a position, just ad-hominem.

Coincidently I am currently reading the aforementioned book myself (oh no, I am a religous loon!!), and if anything I get the impression of an exacerbated person basically pointing out that:

1: The current status quo in nutritional science was established before there was large scale data to confirm/refute it. It was heavily influenced by very few (I believe he mentions 1 or 2) specific scientists and became orthodoxy almost through force of will not necessarily because the data (which they didn't have) actually supported the hypothesis.

2: There are now numerous studies and quite a bit of data that seem to at least point to alternative hypothesis that should be explored. I don't recall yet having seen the author claim even the carbohydrate hypothesis he talks about is correct, just that it should be getting more study/attention than it is in main-stream nutritional science, and that it is better supported by the study data we do have than the currently accepted hypothesis which the studies we undertaken to test.

3: The idea that sedentary lifestyles coupled with abundantly available food just doesn’t track with the actual data as obesity seems to track much more closely with availability highly refined carbohydrates and massively increased sugar consumption. That said both could well play into it, it isn’t necessarily an either-or but much like the famous global warming vs. number of pirates graph it also doesn’t show causation, just correlation, and perhaps meaningless/misleading correlation at that.

4: Explaining obesity by overeating is as useful as explaining alcoholism by overdrinking (okay, I just threw that in because it is a funny quote :))

His primary push that I have taken away thus far is ‘we should be exploring this hypothesis, which isn’t actually terribly new nor is it terribly radical’. He claims people generally aren’t for unclear reasons (though I haven't picked up any sort of conspiracy vibes yet, which I was expecting). I have looked around for someone to refute his book but all I found was a weak refutation in Reason[1] which he responded to, and in my opinion, destroyed the critique quite handily [2]. I would be very interested in any responses that actually take to task the quoted studies / science from his book instead of attacking him by-proxy by calling Atkins a kook.

[1] http://reason.com/archives/2003/03/01/big-fat-fake/print

[2] http://reason.com/archives/2003/03/01/an-exercise-in-vitriol...

1. You're misusing the term ad hominem:

>The mere presence of a personal attack does not indicate ad hominem: the attack must be used for the purpose of undermining the argument


2. I'm not going to respond to (or even read) any of your points about nutrition because you seem to be arguing with someone else, not me! I have no horse in this race, because I've never looked into it. I think you and gxs rather misunderstood Homonculiheaded's words -- which were about the general way folk were arguing, and not really about specific points at all.

More importantly, there's a subset of diets which cater to the "I'm a lard-ass, how can I lose 200 pounds quick" crowd.

Other people are more worried about general health than getting their calorie consumption under 3000 a day.

I really, really appreciate your response.

The micro-religion fanatics were causing me to question why I even bother.

It sounds like you've probably got a lot of insight to offer the community here, but you're doing yourself and us a disservice to resort to name calling like that. Just let your knowledge speak for itself and let's all remain civil. At the end of the day we're all in a pretty tight knot community here. Or maybe I'm just a crazy hippy :)

You're right, but it's frustrating to have people dismiss an entire field of research out-of-hand just because they read a book. Or two.

>>Taubes isn't taken seriously by most dieticians and nutrition researchers we talk to (I work at a nutrition startup).

That's because most dietitians and nutrition researchers follow conventional wisdom, which was established back in the 50s and 60s. Taubes himself included an amazing commentary on how conventional wisdom is incredibly difficult to dislodge in medicine and related fields. He had no illusions about his book being enough to do it, especially since he's a science writer rather than a scientist.

But his work is very valuable, and his dissection of why we believe what we believe is spot on. He goes a little overboard in blaming carbs, but does an excellent job vindicating fats.

>>That's because most dietitians and nutrition researchers follow conventional wisdom

Sadly, you hit the nail on the head, 100%. That is exactly the problem, most dietitians will still try to sell you on the CDCs protein intake recommendation of 50g and and a diet that consists of 80% carbohydrates - which just plain contradicts mounting scientific evidence.

Furthermore, most people will just say you need a "balanced" diet, having taken absolutely no thought to consider why they consider certain diets balanced to begin with.

If you move past your comfort zone, to the point where you can draw your own conclusions, without needing to refer to government guidelines to validate your every belief, you will see that in a few years we will look back on our current recommended diets and laugh.

Taubes does a more than adequate job of trouncing the old thermodynamics, calories in, calories out argument. The guy does have a masters in Physics from Stanford, after all. I leave it to you to look it up.

Edit: Feinman on a calorie is a calorie and why it violates the second law of thermodynamics


"The guy does have a masters in Physics from Stanford"


I've got a Ph.D. in CS from UTAustin, but I don't think you would want to take medical advice from me.

The point was to show that Taubes has the education and training to critically review scientific literature. Not that he is an expert on medicine.

I disagree with Taubes reasoning for other reasons, but the point the poster made was that someone who studied physics at Stanford will understand how the Laws of Thermodynamics work.

>> Sadly, you hit the nail on the head, 100%. That is exactly the problem, most dietitians will still try to sell you on the CDCs protein intake recommendation of 50g and and a diet that consists of 80% carbohydrates - which just plain contradicts mounting scientific evidence.

In nutrition there is never ever (never ever) such a thing as a "scientific evidence".

All the nutrition research around is a pile of conflicting information. Just as an example, a short time ago in HN was pointed a research where "a calorie is a calorie" was proven, in the context of a diet.

I still remain convinced that having a basic clean nutrition and a decent amount and consistent of sport would turn any healthy individual (9x% of the population) into a fit one. People keeps feeling attracted to nutrition religions just because they think they can get in shape without changing their unhealthy habits.

Yes, the advice of all these old guard dietitians is working fantastically for us.

Looks around at all the diabetic, morbidly obese people.

Yep, working great.

How many of those people do you think follow the advice of a dietician even 25% of the time?

It's weird. Often times you'll hear people saying they ate an apple for desert instead of that chocolate souffle the waiter recommended, when their main meal was a 2000-calorie monster full of breaded chicken, onion rings and french fries. And then they will tell you their dietitian told them to eat more fruits, so they are following his/her advice!

First, it is my experience that most people do, religiously so, for a short amount of time (a month or so seems common). And then, the 20% for which this advice produces the expected result continues practicing it, and the 80% for which it doesn't work start following less and less religiously, until they don't at all.

That's mostly evidence that dieticians are solving the wrong problem. To borrow from another poster in this thread, if the composition of what you eat was enough to solve obesity, it is likely that saying "don't drink alcohol" would be enough to solve alcoholism. And yet, hundreds of years of experience show that it is wrong for the latter.

A diet that is easy to follow (such as Roberts' Shangri-La and Asprey's RFLP) is what dieticians should be striving for, but instead most of them recommend a regime that their clients are unlikely to follow (and then blame the clients).

I'm not sure I believe that most overweight people have even had the advice of a dietician.

I suppose things like the food pyramid sort of count, but even the old one didn't endorse the empty calories evident in shopping behaviors (i.e, chips and pop).

Sorry, I assumed you were in referring to people who have dietetic advice in general, but I now see you referred specifically to the morbidly obese.

> but even the old one didn't endorse the empty calories evident in shopping behaviors (i.e, chips and pop).

But it made the (now) classic "calorie is a calorie" mistake, so it implicitly okayed them.

Portion control is actually one of the primary issues, less so than the actual food being eaten.

Many people will eat relatively healthy, but due to poor satiety or portion control, still gain weight.

Lets roll in the science-mobile and discuss real, concrete things.

1. Not eating carbs has nothing to do with our hunter-gatherer origins, there are no over-arching truisms that can be made about the macronutrient distribution of how hunter-gatherers used to or currently eat. Concordantly, paleo "proper" insofar as it resembles utilizing our understanding of hunter-gatherer diets to eat healthier than a Standard American Diet is not necessarily low carb.

2. Eating carbs doesn't cause you to turn into the Stay-Puft man like magical fairy dust, even if they're relatively unhelpful to sedentary people as a nutritive calorie source. Sedentary people should get more protein and fat than carbs relative to what an athlete would.

Moving onto 'keto', which is the more GI/ketosis focused variety of low carb/paleo:

High-fiber and whole-grain calorie sources should be preferred to the opposite. The only thing that makes an apple not "junk food" is the fiber. Fiber has been shown to blunt the glycemic impact of carb-bearing foods. Apples are otherwise useless unless you really need the carbs to survive/fuel your activities.


>Determining which aspects of the Western lifestyle are truly aberrant for our species and pose the greatest risk of obesity is complicated by the conflicting and limited data on diet and metabolism in non-Western populations. For example, while Western diets are certainly more sugar-rich and energy-dense than more “traditional” diets and wild foods [4], [8], [9], many hunter-gatherers seasonally consume a large portion of their daily calories as honey [10], [11] (Fig. S2), which has high concentrations of glucose and fructose [12].

>As is typical among traditional-living Hadza, over 95% of their calories during this study came from wild foods, including tubers, berries, small- and large-game, baobab fruit, and honey [17] (Fig. S2).

>The only thing that makes an apple not "junk food" is the fiber.

This is an oft-repeated mantra among proponents of low-carb paleo-style diets, but it largely ignores the role of micronutrients and particularly phytochemicals in human health. An intelligent choice of fruits can provide a bevy of phytochemicals that lower the risk of a wide variety of ailments. Looking at the macros when talking about the role of fruit in diet is largely missing the point.

Also, hunter-gatherers may get a lot more calories gathering than hunting. Women often gather more food than male hunters, collecting things like tubers, fruits, honey, etc.

IIRC, Homo erectus evolved to eat a lot more carbs than homo habilis. It's only if you pick specific points in human evolution that you can say we didn't eat carbs.

Bit of a strawman, nobody takes the pritikin diet seriously either. The place and purpose of healthy fats is well understood.

It sounds like you haven't actually read Taubes' book. He makes the case that no fat is actually "unhealthy." Saturated fats, which have been demonized, are harmful if and only if they are consumed with a large amount of carbs. That happens to resemble the average American's diet, which is why saturated fat consumption is correlated with heart disease - but it does not cause it.

calories in, calories out

After a lifetime of fighting with my weight, including surgery to give me a fight change, I have to say that statement is the single most useless statement in a segment filled with misinformation. It is used by people who haven't struggled to (perhaps not intentionally) insult people who do. "Gee, can't you count. It's just a calorie."

The problem with that statement is it assumes our bodies are like a gasoline engine, where, yes, a calorie in is a calorie out. Our bodies are not like that. Instead, we take in many different kinds of fuels. Everybody's body is different in how it processes each of those kinds of fuels. More importantly, everybody's brain is different in how it reacts to inputs coming from the gas tank and engine.

Until you understand how the various fuels interact with your body, a calorie is a calorie is completely meaningless.

Carbs are fuel, if you're an athlete, you have a use for them.

Yes, yes, yes. I completely agree with this. The brain needs some carbohydrates to run. The body can, if absolutely necessary, convert other foods into carbs to make that happen.

The things paleo and the various low-carb diets do is give a structure around how to reduce those carbs, especially in a culture that worships the ultra-refined, processed-thrice, almost-straight-sugar carbohydrate.

My family has adopted a single rule. Never buy anything with sugar on the ingredients list. I am of course aware that a little fructose would probably not kill me not even long term, but the biggest problem in dieting IMO is keeping track of your inputs. This rule alone rules out enough crap to make us eat reasonably healthy as the correlation of added sugar to crap food is pretty high. So I understand why simple rules are important I just believe they don't have to be, never eat any carbs.

That's a great rule. I'm going to have to think about and talk with my family about how to implement that in a way that isn't me being a tyrant and them wanting to rebel (yay for the early teens :).

Oh, and I agree on the "don't have to be, never eat any carbs". I've posted elsewhere that I'm a literal food addict. I have to put a lot of structure in place or I go off the deep end. For me, I have best control when I say "no carbs, period", but that is my personal diet (err, that I try to stick to, anyway :). I think a more reasonable world might be "try very hard to avoid simple carbs".

If you aren't lucky enough to not have to care (whatever combination of genes and lifestyle causes that, I curse thee ;), then finding the combination that works for you is critical. I found it though a lot of experimentation (including a major experiment of re-working my body's plumbing because so many previous experiments had failed).

>The things paleo and the various low-carb diets do is give a structure around how to reduce those carbs, especially in a culture that worships the ultra-refined, processed-thrice, almost-straight-sugar carbohydrate.

So we replace one religion with another? The structure is good, but yeesh.

It's replacing a wholly-incorrect model with a potentially flawed, but partially-validated one. I don't think we have the body of knowledge right now to precisely model how a particular diet works vs. another, but we have observed that low-carb diets improve health and wellness.

This isn't to say we should all ditch our bread and pray at the Church of Paleo. But with our current state of knowledge, I think it's more intellectually dishonest to cling to provably false "scientific" models of how different diets affect the body, than to just go with something that works (but we don't really know why). The scientific method dictates that theories must be made to fit observation, not the other way around.

Consider this: a theory that the world was round prior to any evidence of such has no more validity than the flat-world hypothesis. Post hoc, you may find out that the Earth is in fact round, but the knowledge was gained from the observation, not the theory. So call it religion if you want, but it's actually more scientific than the alternative.

This particular "religion" is backed by studies that say it's the most effective way to eat to lose weight. Granted, the only studies I know about are the ones Taubes cites, but is there evidence to the contrary?

It seems to work for a lot of people and the science behind it makes sense to me. I don't see why it should be discouraged.

What part of the laws of thermodynamics imply calories in -> calories out? This belies a total ignorance of their actual meaning. The first law of thermodynamics would be satisfied in any of the following scenarios:

1. Your stomach explodes, releasing the caloric content of the food as heat over a brief duration

2. Your digestive tract decides to stop digesting, and passes the food through unmodified

3. Normal digestion

As you can see, conservation of energy is completely meaningless in the context of diet.

The second law of thermodynamics concerns itself with efficiency of processes. To apply it usefully, we'd take into consideration the different metabolic pathways of proteins, carbohydrates, and fats. We know for a fact that the pathway for sugars is thermodynamically more efficient than proteins [1] - the laws of thermodynamics directly contradict "a calorie is a calorie"! Not to mention the fact that consumed protein is metabolized and stored differently than consumed sugars, which are in turn different from fats.

[1] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC506782/#__sec4ti...

> What part of the laws of thermodynamics imply calories in -> calories out?

The part where you restrict your assessment to more or less normal digestive function? Obviously if your stomach explodes or your intestine just plain stop working, caloric intake/deficit is now one of the least of your worries.

I think you actually hit the nail on the head here though, in using the word "imply". Calories in - Calories out doesn't give you an exact picture, but it comes reasonably close for the purposes of most people.

It's by no means a perfect understanding, but someone who is obese needs simple rules that can successfully get them started more than they need a perfect knowledge of the science involved.

"more or less normal digestive function?"

what is that exactly? I have twin girls, one is a full foot taller than the other. One is noticeably pudgier than the other.

They both eat differing amounts, and clearly pretty much the same kinds of foods is used in differing ways by different digestive systems.

what makes you think there is an establushed norm?

"Calories in - Calories out doesn't give you an exact picture, but it comes reasonably close for the purposes of most people."

Im not at all convinced that this is true. It sounds true, and it naturally appeals as a simple truth, but I haven't seen much evidence to support it.

From what Ive seen across a wide range of people, differing bodies process foods in wildly differing ways - some people fart or burp a lot, some people hardly ever, some people prefer eating large amounts once a day, some people prefer eating small amounts regularly. As a teenager I could consume mcdonalds like a horse, now a single burger sits like a lead stone in my stomach for hours.

Convince me that it is true?

Are your twin girls adults yet? Nutrition and diet are difficult while people are growing. Your body is actively building itself in a way that never happens again. You need larger quantities, of almost everything, than body mass would suggest. You're right, for children, it isn't simple calorie arithmetic.

That said, how disparate are the "differing amounts" they eat? Are they identical twins or fraternal? Lifestyle differences? There are a lot of factors that could produce the effects you're seeing besides "calorie arithmetic doesn't work".

The essence of telling people "calories in - calories out" is to favor a simple (i.e. "easily applied") rule over an absolutely correct one. While people are growing, the equation changes. Get all of your calories in the form of twinkies and your body doesn't get the nutrients it needs to process food, so the equation changes. Assume "normal" ranges for diet and exercise and it more or less works.

This is a fantastically complicated topic, but it is not immune to human nature. We like simple rules that generally work, even if they're full of loopholes and caveats. For people that are overweight or even obese, calories in - calories out is good enough to get started. There are a lot of factors in managing that equation (satiety, nutrient balance, impulse management) but I think a clear goal is a better start than dumping people in a sea of (correct) information.

Taubes shows those simple rules don't work (so many people have difficulty loosing weight) because the causality is backwards.

Ergo, give the obese people the rules and shut. up. about Thermodynamics.

I can agree with that opinion, but I also see the value in "appealing" to science.

People will believe all kinds of stupid things if you can imply that science has made truth of them. For the purposes of getting someone started this is a rather innocuous lie.

> metabolic pathways

well, here you go!


I agree that there are several factors involved in the process of absorption for "Calories in, calories out" to be taken as a sufficient guideline for losing weight. However, the point I think people try to make when they use it is that the raw amount of calories you consume are an UPPER BOUND for the amount of energy you have available.

I know that the actual values are hard even to estimate, but we only get energy from food: if you need amount X of energy to keep functioning and the raw total of energy contained in the foods you eat is less than X, you simply can't gain weight, "asymptotically" speaking. That's why people invoke thermodynamics here.

  In the end: calories in, calories out. Humanity hasn't defeated the laws of thermodynamics.
I think is a simplistic and problematic view of it. We are complicated machines. We can't process all types of input efficiently and without side effects. Any type of machine works differently depending on the source of energy. I can't put diesel fuel in my car and expect it to run the same way that it does on gas. Diesel fuel has energy just like gas but there is more to it than that.

> In the end: calories in, calories out. Humanity hasn't defeated the laws of thermodynamics. It's just a question of how easy and healthy it is to maintain a sub-maintenance caloric intake.

It is all about "calories in, calories out" but what on earth makes you assume that the human digestive system approximates the behavior of a bomb calorimeter? The whole article is about how the body extracts different amounts of the potentially available calories in food, and uses different amounts of energy to extract those calories.

> Humanity hasn't defeated the laws of thermodynamics.

But that doesn't mean what you think it means. Link: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC506782/

Great article, thanks for the link. For those who don't want to wade through this technical and fairly lengthy article, here's a quick summary:

The people who say "a calorie is a calorie, and to deny that would violate the laws of thermodynamics" are arguing from the first law of thermodynamics (conservation), but misunderstand the second law (entropy). The article shows that if you take into account both laws of thermodynamics, then the statement "a calorie is a calorie" is actually a violation of the laws of thermodynamics.

The next two paragraphs are the conclusions of the article, quoted verbatim:

A review of simple thermodynamic principles shows that weight change on isocaloric diets is not expected to be independent of path (metabolism of macronutrients) and indeed such a general principle would be a violation of the second law. Homeostatic mechanisms are able to insure that, a good deal of the time, weight does not fluctuate much with changes in diet – this might be said to be the true "miraculous metabolic effect" – but it is subject to many exceptions. The idea that this is theoretically required in all cases is mistakenly based on equilibrium, reversible conditions that do not hold for living organisms and an insufficient appreciation of the second law. The second law of thermodynamics says that variation of efficiency for different metabolic pathways is to be expected. Thus, ironically the dictum that a "calorie is a calorie" violates the second law of thermodynamics, as a matter of principle.

The analysis above might be said to be over-kill although it is important, intellectually, not to invoke the laws of thermodynamics inappropriately. There are also, however, practical consequences. The seriousness of the obesity epidemic suggests that we attack it with all the means at our disposal. Metabolic advantage with low carbohydrate diets is well established in the literature. It does not always occur but the important point is that it can occur. To ignore its possibilities and to not investigate the precise conditions under which it appears would be cutting ourselves off from potential benefit. The extent to which metabolic advantage will have significant impact in treating obesity is unknown and it is widely said in studies of low carbohydrate diets that "more work needs to be done." However, if the misconception is perpetuated that there is a violation of physical laws, that work will not be done, and if done, will go unpublished due to editorial resistance. Attacking the obesity epidemic will involve giving up many old ideas that have not been productive. "A calorie is a calorie" might be a good place to start.

That article underestimates the level of muddled thinking that exists.

There are people convinced that they can gain weight on a hypo-caloric diet. Of course this is easy to do, drink a glass of water, but that isn't what the muddled thinkers mean.

>In the end: calories in, calories out.

If you're just talking about "weight loss" then yea, that's right, but you don't address body composition at all here.

If you just wanted to "lose weight", then you could chop off an arm and BOOM, you just lost weight. What MOST people want to do is to burn and lose fat, NOT weight - they just express it differently.

If it was just calories in vs calories out, I could drink 3000 calories of vodka a day and still be okay as long as I burned off 3100 calories, and we all know that's not the case.

Yikes, that'd be nearly two fifths of vodka. I'm could be naive here, but I'm not sure the hardiest of alcoholics can put that away in one day.

I get the point, mind you, just shocked by how much alcohol that is...

Downvoted for opening with a (double) argument from authority.

For your "more scientific" view, you offer assertions without citing any supporting evidence. So, as is, your statement is an opinion, as is so much nutritional advice.

Hope your startup is based on more logic and less conventional wisdom than shown here.

I go with what the PhDs doing the on-the-ground weight-loss studies say, not the anecdote of some person on the internet who's read a book or two.

Programmers hate it when people question them, you'd think they'd extend the same professional courtesy and benefit of the doubt to other academic and professional disciplines.

Please provide citations if available.

It is entirely possible there is a well-understood theory with masses of supporting evidence that for whatever reason the public don't think is sexy enough to be true. This is the problem with evolution for example, where there is the data behind it but the US public still think its controversial.

Note: If Person A provides a book reference (secondary evidence) and Person B provides no specific reference (no evidence) then I am biased towards Person A. If Person B could provide a reference to some research (primary evidence) then I am biased towards Person B. If Person B ends with some general slur against my profession after someone asks for a reference, Person B loses the bias.


My comment here lays out some ground facts/truths.

Respond there for questions.

Calories are a measure of heat. Not Energy. You measure calories buy burning the food in question and measuring how hot it raises the temperature of a given amount of water.

Magnesium burns a lot hotter than corn therefore it has far more calories than corn yet eating a pound of magnesium a day is not going to make you as fat as a pound of corn a day.

So please stop with the "calories in, calories out" simplification.

> Calories are a measure of heat. Not Energy.

Calories ARE a measure of energy. One calorie = 4.2 joul = 4.2 wattsecond.

A calorie is defined* by the amount of energy needed to heat 1 gram of water by 1 degree celsius under a pressure of 1 temperature. But heat is energy, and a calorie (as well as a joul) are a measure of both.

> So please stop with the "calories in, calories out" simplification.

That part is true: Calories in, calories out is wrong, as has been shown repeatedly over the last 50 years. But somehow it is still the religious belief of many.

You're correct.

I should been clearer. I meant "energy" in terms of energy the body can use. E=MC^2 means everything can be considered energy. But can something be used as energy for a specific process, like making a body fatter, is another issue.

We agree, calories do not measure how much energy the body can extract or use from a piece of food. They only measure how much energy is produced burning the food. Bodies don't burn food, they digest it. So measuring calories doesn't say anything about whether a particular substance can be used by the body.

> They only measure how much energy is produced burning the food. Bodies don't burn food, they digest it.

You need to review your biology. The most common energy extraction process in the body actually IS, at its essence, oxidation - a.k.a "burning". And that's why people did the calorimeter burning measurements. The other common process is essentially fermentation, but in a healthy human, the vast majority of energy is obtained from food through oxidation.

> So measuring calories doesn't say anything about whether a particular substance can be used by the body.

It puts an upper limit because (as far as we know) oxidation is the most efficient process the body uses, but not a lower limit.

The body may flush out anything without any form of digestion, fat conversion, or oxidation (thus no energy extraction), even if it is pure sugar. Furthermore, the enzymes used by the body for oxidation are very specific. The most abundant carbohydrate around us, cellulose (wood and paper are examples), cannot be digested / oxidized / burned by the human body.

     Taubes isn't taken seriously by most dieticians and 
     nutrition researchers we talk to
That's an appeal to authority that dismisses the many studies given as reference in the Taubes book, studies which have results that are contrary to popular wisdom.

One of those studies involved a comparison between a group of people that were placed on a low-carb diet and the weight loss was more efficient than in the other group placed on a high-carb diet that ate less calories.

Of course, when it comes to nutrition, most studies are flawed because you can't do double-blind studies and because controlling the variables involved is hard. However, that's one reason why I go by my gut feeling and completely ignore "most dieticians and nutrition researchers", at least for now.

And I'm doing that because those same dieticians and nutrition researchers have recommended and are still recommending high-carb diets, with complete disregard to the health issues associated with consuming a diet rich in sugar, starch and deficient in proteins, fats and certain vitamins (the fact that we are omnivores that need a balanced diet should have rang a bell, but no).

Not to mention, counting calories is just painful. Have you ever really tried to do this? Frankly, it just sucks.

Has any other culture in history EVER counted calories? I'm betting "no" considering the calorie is a relatively new concept (1842). And yes, somehow they managed not to have the obesity problems we have today.


Counting calories is a great way to lose weight actually. Every time you go to eat something you think "Do I really want to record this in my food log?" Then you say no and don't eat it. I ate drastically less as a direct side effect of trying to count calories.

I wonder if can affect variety of the diet though? "I have no idea how many calories this salad is, let me eat that well measured donut instead and be done with the paperwork" would totally be my train of thought...

This never addresses the mentality of "eating better" instead of simply "eating less."

> "counting calories is just painful. Have you ever really tried to do this?"

Yes. It is (or at least was) the core mechanic of Weight Watchers (converted to a "points" scale), and it worked pretty damn well.

For me, it was a pain for about 2 weeks. By then, I'd figured out all of my common food items, and how to cook dishes with reduced calories but that were still plenty filling.

As a result, I now know how to structure my diet to get the same benefits without actually having to count.

I considered calorie counting too hard to do. Then smartphones and apps came to be. I've been using the myfitness app for several months, and it's definitely bearable.

In the past, most of the people didn't spend time watching TV while eating junk food.

And yes, I do count calories. I'm committed to staying fit, and counting is not an option.

I'm philosophically opposed to the phrase "calories in, calories out". True, we can't violate thermodynamics. However, the real important thing is calories absorbed, and how those calories affect metabolism, calories out.

Not as poetic, but you said it yourself - 300 calories of oreos are not equal to the same calories of chicken. And wood sure has a lot of calories (right?) but chew it up and eat it and you're not gonna get many of them absorbed. And some food triggers more of an insulin response which affects what your body does with the calories. It conveys the wrong point to the person who doesn't learn more to say "calories in, calories out" even though at some level of course it's thermodynamically sound.

There is definitely a scientifically provable correlation between caloric intake and weight gain. It's somewhat causal in that if you suppress calories weight gain (growth) is reduced.

The tricky part is that causality runs both ways. When a child grows, they take in a caloric surplus. If you surpress caloric intake, the growth will be suppressed. It's the same for plants too. Block the sun & they will grow slower.

So (according to good/bad calorie theories) it's not that calorie accounting is incorrect. Take in fewer calories and you'll lose weight. It's that it misses the point. The correct question is why is your body growing (and pursuing the calories necessary to fuel it).

*My own pet theory is that foods aren't inherently bad or unnatural for us to eat. We're just wired to build different bodies depending on available food sources. Mono-croppers need to survive bad harvests & hungry winters. Sugary fruits usually aren't available for very long, it might be good to build up a reserve while they are.

> Humanity hasn't defeated the laws of thermodynamics.

Human body does not extract 100% of energy from food, and waste that comes out is not calorie free. So it is not simple 'calories in, calories out'

It's ok, Taubes doesn't take most dietitian and nutrition researchers seriously either :) (thankfully, imo). -demagogue

What's the startup? Do you have a URL? Thanks.

I think that's more academically interesting than it is practically useful. Some of the foods I eat may have calorie counts that are off by 20 or 30 percent? That's noise.

Being on an 1100 diet that you think is 1400 (or vice versa) doesn't matter that much. What matters is that it isn't 3500--something that can easily happen by accident if you don't pay attention.

If this is true, it makes counting calories incredibly hard for anyone who has a varied diet.

If each ingredient in the food you eat is digested at a different efficiency depending on the bacteria in your gut and other specifics of your digestive system and the food itself varies by upwards of 20% because of natural variation and how much you cook it, then it seems it would be downright impossible.

I've tried in the past and given up (I would gain weight on a 1200 calorie diet of processed diet food). I finally just went to an extreme nearly vegan diet where I get over 90% of my calories from plants and exclude grains, oil, sugar and salt. It makes it virtually impossible to overeat even if I have to eat 4x the mass of food other people.

I too have used that approach, and the only problem I've found is because I'm still used to taking in a large mass of low calorie raw foods, if I'm in a situation where the type of food I'd normally eat isn't available, then I have the capacity to really pig out.

It would be very interesting to see an entropy factor for common foods, and in particukar things that don't digest (nuts, etc). Like a reference table.

It's crazy to me that 'counting calories' is something people routinely take for granted. It is a uniquely human trait. No other animal in this world has a need to do this. Humans up until a hundred years ago (or so) had no need to do it. What the hell has happened?

My opinion - food processing and the agricultural revolution. A metabolic derangement caused by overly processed agricultural products.

What about Occam's Razor?

> "Humans up until a hundred years ago (or so) had a need to do it."

Actually, the aristocracy and nobility have all suffered from obesity since ancient times. Historical texts have many, many records of... robustly sized... monarchs, nobles, and the rich. Indeed, the image of the rich fat-cat is a lot older than merely the last century.

> "My opinion - food processing and the agricultural revolution. A metabolic derangement caused by overly processed agricultural products."

Or perhaps more simply, the majority of the population stopped doing hard physical labor day in day out. We started sitting behind desks for 8 hours a day instead of toiling in a field or factory for 12.

I'm also willing to believe that we are simply eating more than we used to. In industrialized nations the ratio of food cost to income is almost laughably low - compared to agrarian nations where the cost of food can sometimes be up to half of a family's income. If you look at countries with smaller average serving sizes you will see a corresponding drop in obesity rates.

Thankfully I think the country is starting to come around to the notion that no amount of sodium reduction, fat-reduction, or carb-reduction will work if you continue to shovel it into your mouth by the gobful.

It's become vogue to blame every ill of society on industrialization, when instead the answer seems to be simpler and more obvious.

"Thankfully I think the country is starting to come around to the notion that no amount of sodium reduction, fat-reduction, or carb-reduction will work if you continue to shovel it into your mouth by the gobful."

The things I usually want to shovel in my mouth are carbohydrates, and insulin spikes explain at least part of that. I eat far less when I'm trying to keep my insulin low.

Everyone knows you have to eat less to lose weight. The question is how. Some techniques are more successful than others.

I tend to take this outlook. Almost all forms of calorie restriction works. When I personally point out that a calorie is not a calorie it has everything to do with satiation, motivation and the psychology of what makes people over-eat.

Limiting calories works. Keeping people doing that long enough to lose weight, and the rest of their lives to some degree (so no yo-you dieting) is the true challenge.


We are learning that certain things in the diet (sugar, grains) promote metabolic issues that fuel overconsumption.

The ideal state of being, which everybody would like to enjoy, is being able to eat until you are full and then stop.

That is not such a simple process from a hormonal standpoint.

My point was that it was not a pandemic level problem up until recently. Of course rich people with access to an abundance of highly palatable foods can become fat. Those aristocrats of old are suffering the same thing many people suffer today.

Edit: I suppose people don't see the connection between over consumption, availability, palatability, and processing.

You just did an about-face. In your first comment, you stress "processed foods"; here you put the emphasis on greater availability.

I think potatolicious makes an excellent point- "processing" (the popular demon today) cannot be pointed to as the sole cause. It is probably part of the problem, but probably not the root.

No I didn't. You missed or ignored the 'highly palatable' part.

Fruit tastes good, and rich people could get as much as they want, but they generally don't get fat on fruit. Confectionery items on the other hand...

What I'm saying is that over-consumption and processing drive each other.

So when you say "processed", you're actually referring to plain old food preparation?

Well, I suppose confectioneries in particular with confectioner's sugar could be considered "refined"...

Processing takes many forms.

Removing of fiber, adding of sugar, etc.

The end result of many types of processing is a food which is both more nutrient dense and less filling.

Fruit comes with fiber, which has suppressing effect on appetite, whereas confectionery items are usually made with pure white flour, which due to insulin spikes actually leads to an increased appetite.

> No other animal in this world has a need to do this.

Have you never seen an obese dog or cat? Feeding to much of the wrong food will make any animal fat.

Exactly. Most animals exist in a world of scarcity, near starvation for significant periods - often every year. Mammals that live in conditions of abundance usually get fat. We've evolved to do that and hedge against scarcity.

I don't buy it. When I go out into nature I don't see many emaciated animals.

They may exist in a world of scarce resources, especially predators, but they are not constantly living on the edge of starvation.

"Historically, in Michigan the number of species diagnosed at the Laboratory as dying from malnutrition and starvation are second only to those dying from traumatic injuries."


Sure, I've seen an obese dog or cat, and I've also seen what these animals are fed.

I guess my point was not clear enough.

Feeding any animal an improper diet can lead to a metabolic problem where satiety and fat deposition do not work as intended.

Animals in the wild eating their natural diet do not have a need to limit their consumption. They self regulate effectively.

It's not so much self-regulation as the lack of resources in the wild.

I suppose what was implied here was "undomesticated animal out in the wild and not held in captivity".

One case to consider, in the modern era, might be performance. Like NASA expeditions, mountaineering, long-distance hiking. Anytime you need to carry the weight of your food (for 5+ days), under strenuous exercise. Calorie counting, weight/volume measurement, and nutrition labeling become essential.

> My opinion - food processing and the agricultural revolution. A metabolic derangement caused by overly processed agricultural products.

It's simpler than that: dense food and lack of exercise. More people have access to more calories than at any other time in history. At the same time, we spend more time at work and at play sitting and expending minimal energy than at any other time in history.

Seems like human beings have stopped moving their feet :-)

This may be of interest to you


People researching modern hunter gatherer societies are finding that they expend roughly the same amount of calories as westerners.

That does pose some interesting puzzles.

Perhaps the fact of moving changes how the body partitions nutrients. Or handles hunger. Or some other factor. Or the specific modalities of exercise (brief, high-intensity vs. long, low-intensity) matter.

The study appears to rule out one possible mechanism (total calories consumed). There's still the fact of bodyweight / body composition differences.

If your test fails to account for observed differences in phenomena, then your test failed to identify the cause(s).

Just as the sick man who goes to the doctor for a battery of diagnostics and is told "we can't find anything wrong". The problem isn't in the state of the man's health, it's in the state of the doctor's diagnostics.

"... they expend on average no more calories in a day than the adult population of the industrialized world." Did they take weight difference into account? is it per kg or per person? I find the article lacking in a few important details...

Our diets changed. If we revert to eating what humans ate 100,000 years ago we do NOT need to count calories and we do not get fat. What happened is we started cooking and eating things that could not be eaten raw, like grains, potatoes and so on. If you eat only stuff that can be eaten raw your body will do a pretty good job of maintaining your ideal weight without counting calories (assuming you don't have a thyroid problem or something of course)

We will also end up being a lot shorter by eating the diet of ages past, which suggests that that is not the optimal diet either.

I've not heard about this, you got a source for me?

Looks like I am mistaken. We're taller than we have been for the past few hundred years, but there is evidence of some earlier humans having similar heights as modern humans. There is no mention of the cooked/raw status of those particular ancestors' diets, though.

I don't have a particular reference, just piecing together the non-trash results of googling 'diet height' and 'history of human height'.

We got the 'food pyramid' which told everyone that eating loads of carbs is good for you and eating fat in any form is bad. It's gonna take decades to reverse this mistake.

Fascinating article. Well worth the read if you have an interest in diets and nutrition.

Why is it scientists can't get an accurate grasp on the calorie content of foods, and how our bodies process them?

Why is it scientists can't get an accurate grasp on the calorie content of foods, and how our bodies process them?

I just posted another article that talks about the difficulty of doing human nutrition studies. Way back in the 1970s, I saw a television news magazine report about a group of nutrition researchers who found human volunteer subjects who were willing to be shut away from their normal life for weeks (months?) at a time (being paid for their time, as I recall) in conditions under which EVERYTHING the subjects ate was weighed to the nearest gram. The TV report even showed researchers washing crumbs off plates with distilled water so that the subjects ate every morsel of their measured meals. That was part of a study designed to elicit new information about human nutrition. Alas, I've never heard of results of that study. (Probably, the sample size was too small to have adequate statistical power to reach any firm conclusions.) This will always be the problem with human nutrition studies: people think they remember how much they eat, but often enough they don't really remember in detail, and the usual study methodology for a nutrition study is to rely on subject self-reports. If the sample size is extremely large, sometimes there is enough statistical power to pick up a few tantalizing signals from masses of carefully gathered data, but then there is still the whole issue of how well actual human beings apply advice based on the studies that have been done to date.

I am amazed at how much day-by-day obesity in very young people I now see in my town. I never imagined that most Americas in my generation or the younger generation would be as bulky as they now are. And the United States isn't even at the top of the world league tables in obesity. People all around the world today have eating tendencies based on human beings evolving in conditions of food scarcity, but there are fewer and fewer places each decade that still suffer from food scarcity. (Ethiopia is still poor, but it is now a food-exporting country.) The challenge for humankind in the remaining years of my life will be learning how to do what is "unnatural" and eat less than what is readily available to eat.

"And the United States isn't even at the top of the world league tables in obesity."

True, but only just (according to the WHO). The only countries that beat it on overweight or obese percentages of population are Saudi Arabia, Panama, or have a population under one million.

Perhaps they can but until science can unlock the psychological interaction of people and food they're not going to change anyone's food habits.

If you look at the progress of medicine with respect to addictive behaviors, there seems to be very little progress. I believe food eating behaviors are very similar. In fact I think obesity is the hardest addiction to overcome simply because you can't eliminate food from your diet as you can eliminate all other addictive substances.

I know I've lost 30-40% of my weight several times and kept it off for a number of years but it always returns due to my inability to resist the psychological pressure to over eat and/or eat dense caloric foods.

It's immensely complicated.

and from the article, can vary greatly by individual

It seems to me that the effect of food on the body has been studied for at least 50 years.

No matter how complicated it is, shouldn't there be a few basic things everyone can agree on? It seems that each new diet (based on scientific principles) contradicts the last. Low fat, low carb, low calorie, vegan, no dairy, no wheat...

They can't even agree on whether a calorie matters. (i.e. is eating 100 calories of steak vs eating 100 calories of sugar the same?)

It seems that each new diet (based on scientific principles) contradicts the last.

Nobody gets paid for solving solved problems.

This is very interesting. "Calories in, calories out" has long been an established principle and topic of debate & discussion among those who follow nutrition, and while it largely still applies, this article points out the factors that bend it somewhat. I wonder about the compositional effects of diet on the way ingested calories are used, too: studies have shown that isocaloric high-protein diets result in increased lean/muscle mass and lower fat mass, compared to high-carb ones. It would be nice to have a survey article to read on recent studies in this area, if anyone would be so kind.

I liked this factoid in the article: "Back when it was the craze to measure such variety European scientists discovered that Russian intestines are about five feet longer than those of, say, Italians." Useful adaptation there (I wonder if it tends to make Russian guts any bigger?)

Here's the core problem. At a fundamental level the thermodynamic argument with regard to body weight has to work, of course it does, there's no way to slip past the laws of physics. If you eat little enough and exercise enough you will lose weight, period, full stop.

But by the same token, it's a useless fact, utterly.

Look at the same argument as applied to finances. If you make more money than you spend you will build up savings until eventually you are rich.

The issue is that these are trivial facts. They don't tackle the core difficulties of the problem they are just bland tautologies. Ultimately long-term weight loss is a complex matter of psychology, motivation, habits, and indeed personal temperament and genetic baggage. For some folks it will come naturally and easily, for others it will always be an uphill climb, and for others it will be almost impossible. And the same thing is true with regards to being wealthy. Some folks quickly and intuitively acquire the skills to step into a business, tweak the knobs until its profitable and then start building up their personal finances from those profits until they have a desire to move on to something else where they do the same thing, continuing to get richer and richer over time. Whereas for others it can be much more of a struggle, or maybe even effectively impossible because they don't have the same opportunities.

> Here's the core problem. At a fundamental level the thermodynamic argument with regard to body weight has to work, of course it does, there's no way to slip past the laws of physics. If you eat little enough and exercise enough you will lose weight, period, full stop.

This is just not true. Our bodies follow the laws of thermodynamics, but the laws of thermodynamics do not dictate how or whether we process calories. We can poop out fat. If you drink a bottle of olive oil, you will probably not digest it all. If you eat fiber, you won't process those calories. Our bodies are not bunsen burners.

It's almost like you read the first sentence and decided to immediately rebut it without reading the rest of the comment.

He does end this section with "period, full stop".

However the rest of the comment is tangental does nothing to mitigate my response. Weight maintencance is not about calorie count, it's about understanding how our body processes food.

But all that means is you will process less than your intake, which does not disprove his point:

If you eat little enough and exercise enough you will lose weight, period, full stop.

Yes, precisely. But I'm talking about the limits. If you ingest fewer calories than you burn you will lose weight, period. That's the argument that people have been making about weight loss for ages. But the difficulty there is that in reality the range of "thermodynamic calories" that a person can ingest relative to the exercise they do while maintaining a healthy weight can vary across a huge margin.

I recall reading about an experiment where obese mice were injected with insulin regularly after being denied food. The mice starved to death despite still having fat stores. Of course, mice are not humans, but this implies that diets that stimulate a lot of insulin production (or normal amounts in insulin-resistant populations) would cause a person to not lose fat.

Sounds like a great study consistent with my experience and the prevailing view of the anti-diet.

I'm not sure how you looked at the sentence "if you eat little enough and exercise enough you will lose weight" and managed to read "if you eat enough and exercise little enough you will gain weight".

We're not talking about helping sick people gain weight by shoving food down their throats, we're talking about losing weight.

p.s. Pooping out fat is 'calories out'. Several weight loss products even make use of this.

> p.s. Pooping out fat is 'calories out'. Several weight loss products even make use of this.

Not in the sense of what people mean when they say 'calories in equals calories out'. Usually they mean that most calories that are consumed are utilized, and hence anything not burned gets added to your weight. This is of course an incorrect assumption based on both laws of thermodynamics, and also, as you point out above, ignoring the calories that aren't absorbed during digestion.

> "Ultimately long-term weight loss is a complex matter of psychology, motivation, habits, and indeed personal temperament and genetic baggage."

As someone who has suffered and struggled with obesity in the past, I hear ya loud and clear, but IMO the notion that "calories in < calories out -> weight gain" is still very valuable.

One of the fundamental problems that lies between our society and healthy weight is that we continue to believe in miracle diets, or that somehow we can cheat the rule. We talk about being carb-free, being fat-free, rarely acknowledging that maybe we should just stop eating so damned much.

When I was in high school I was quite obese, and the (Canadian) government paid to send me to a dietician in a group program. Let me tell ya, the notion that "calories in must be less than calories out" was news to a lot of people there.

You're right of course - this knowledge, in an of itself, is insufficient force to turn a fat man into a thin man, but in the current state of our society, simply acknowledging and owning up to this fact is pretty tremendous.

Why do you say that? Is the knowledge that heroin is bad for an addict any motivation for them to quit? I hardly think so.

Considering that you can eliminate heroin from a person's usage and that you can't eliminate food it seems that the relapse rate for both is reasonably close.

Until obesity and eating psychology gets more study in this area, I don't think anything's going to happen to improve the health of the typical US citizen (or Canadian, eh?).

For me, I'm mostly convinced of the paleo view. It's by no means the only valid view or the whole view but I'm convinced by numerous anecdotes from people I know personally and elsewhere.

Here's one quote from Anna Karenina which suggests that a similar approach was well known in Tolstoy's time:

    On the day of the races at Krasnoe Selo, Vronsky had come
    earlier than usual to eat beefsteak in the common messroom
    of the regiment.  He had no need to be strict with himself,
    as he had very quickly been brought down to the required
    light weight; but still he had to avoid gaining flesh, and
    so he eschewed farinaceous and sweet dishes

If you want to experiment with your own dietary in-out behavior, I recommend you institute a change for 3 months and then plot the diff using this


It mathematically modelled from data of many large, longitudinal studies (Nurses' Health, Framingham, etc).

As a physician who reduced caloric intake in January, I've found the bwsimulator to be very predictive and informative on how I can modify things further in the future. But you really do need a couple of data points on yourself to plug in first.

Unrelated to the topic, but all the "X is not a X" subtopics remind of this interesting paper I read not too long ago titled "A File is Not a File: Understanding the I/O Behavior of Apple Desktop Applications"(http://research.cs.wisc.edu/wind/Publications/ibench-sosp11....)

Good article on why counting calories precisely is pointless and misleading. Another factor is that food labels are often incorrect - the FDA allows them to be up to 20% off in the USA. Even if you could figure out how many calories you consume and absorb, estimating the number of calories you burn is even more difficult.

Whether or not the raw numbers correspond 1:1 to actual calories doesn't really matter. What matters is that you adjust your caloric intake based on whether you are gaining or losing weight, and which direction you need to be moving in. Even though the relative amounts aren't necessarily exactly correct, it gives you a basis for adjustment.

The same goes for counting calories burned while exercising. You may not be able to figure out an exact count, but you can get in the ballpark, and that means that you can then adjust your habits so that you are moving toward your goal.

Counting the calories precisely will not allow you to scientifically formulate an optimal diet and exercise plan on paper without experimenting on yourself. It will, however, anchor you and protect you against fudging the numbers in the direction of letting you eat more or exercise less.

You could, I suppose, use an unbiased rounding system to try and use only as many bits of calorie information as you believe are significant, but that's a lot more work for very little theoretical gain.

>>Whether or not the raw numbers correspond 1:1 to actual calories doesn't really matter. What matters is that you adjust your caloric intake based on whether you are gaining or losing weight, and which direction you need to be moving in. Even though the relative amounts aren't necessarily exactly correct, it gives you a basis for adjustment.

But that's the point: you cannot make accurate adjustments unless you also keep track of the type of food you eat (cooked, processed, raw, etc.). Because you have no way of knowing whether you gained weight due to taking too many calories, or because the food you ate last week was processed and cooked.

That's one of the reasons that people who make modulating their body composition a key priority (bodybuilders) often stick to very consistent diets. By not varying the type, quality, and preparation of the food, they can modulate quantity for desired effects (bulking, cutting).

Yep, my diet is ridiculously consistent. I cook my food in bulk (saves time and money), drink lots of protein shakes, and when I eat out I usually order the same foods. My typical weekly meal list probably contains fewer than ten items that I switch around. Most of the variety comes from experimenting with new zero-calorie ingredients like condiments (e.g. cinnamon) and salt/pepper.

It adds noise, but you compensate for that noise as you go. If the noise causes you to take in too many calories one week and gain weight, then you can make that up by reducing your calories the next week. If you lose weight too quickly, you can increase your intake. As long as the calorie estimates are within a reasonable range of the real value, it shouldn't matter much in the long run.

Counting calories might be misleading but it's definitely not pointless. You still want to target a range of calories during the day and not go over it. Even with the inaccuracies of the system, counting calories is still part of keeping healthy.

Absolutely. Counting calories is a fast feedback loop (count each day, compare to your target, increase or decrease your intake as appropriate); you need a slow feedback loop to set the intermediate target against your actual goal (in this case, say, scale weight). The accuracy of the fast feedback loop isn't critical since you have a pretty accurate slow feedback loop to correct it.

The general approach of partitioning problem solving into fast/inaccurate and slow/accurate is very useful, since it's generally hard to make something that's simultaneously fast and accurate.

Note one other interesting thing about this approach: you can do unit conversions! That is, I can measure something with different units than the thing I'm trying to control. Our reference is body weight, but what we measure is caloric intake: the function we use to slowly adjust the calorie target as a function of our weight has units of calories per body mass.

How do you know what range of calories to target? Or that the labels are correct? How many calories are in the burger from that food truck? That the calories you are eating are healthy? I agree with you that it's good to know roughly how many calories are in your food, but literally counting calories during a diet can be counterproductive. Most people won't count calories in the longterm, so it ends up being a bandaid solution that doesn't encourage lifestyle changes. Ultimately listening to your body (for me this includes semi-regularly weighing myself and checking my body fat %) gives a better idea of how much you should be eating and leads to a stable, longterm solution.

>How do you know what range of calories to target? Or that the labels are correct? How many calories are in the burger from that food truck?

A range of 2,500-3,000 seems to be a good fit for me. I'm a 5'9" adult male with a light workout schedule.

I don't expect labels to be 100% on target but I do expect them to be close enough for my rough estimates.

The food truck may not list the calorie count for the burger but I've probably eaten all the ingredients in the burger before. From experience of noticing nutritional content in the past, I can tell if it has more fat (lots of mayo, extra cheese), sugar (ketchup) and overall calories (heavy, overall size of the meat patty) than ones I've had in the past.

The caloric content is not the only tool, it's one of the many tools you should have when trying to keep a healthy diet.

Labels might be 20% off or even 50% off, but still give you ballpark figure, and is better than nothing. I haven't had patience to precisely track calories, but I find it enlightening to look at labels and compare. For example, my intuition tells peanuts are more calory-dense than carrots. But by how much? Oh, 14x! That's good to know--1kg of raw carrots might barely cover one meal, but 1kg of peanuts would power me for 3 days--careful!

>I haven't had patience to precisely track calories, but I find it enlightening to look at labels and compare

Yes, absolutely, everyone should do this (I certainly know how many calories are on the label of most things I eat). But you aren't counting calories.

I'm not really sure from this comment what your objection to counting calories is.

Is it that you can't do it accurately? That seems to be implied by the second two questions. I haven't found that accuracy actually matter that much; mostly what's important is nominal calories over time vs. weight/other health indicators over time. If the number's not technically accurate, I kind of don't care. All I know is I roughly ate so many calories, my weight changed by x amount, and I can walk x far without getting tired. Over time, I can have a pretty good guideline for the number of "back of the box" calories I should be eating to reach my targets (which addresses your first question of "what range to target")

You mention explicitly that most people won't do it over the long term, which is definitely relevant. But even doing it for a little while can give you a good baseline so you can at least make educated guesses as to what you're eating, which is especially helpful in the "food truck burger" scenario.

I guess my point is that we all agree knowing roughly how many calories you're eating is good, but the only way I know to figure that out is to, at least for a while, literally count them. There's still a huge margin for error, of course, but if I didn't keep tabs on it by looking at the back of the box, I'd have no idea at all.

I think you miss the point of calorie counting in the first place. You count calories so that you become aware of just how excessively you've been overeating without noticing it, and where those calories are coming from. It doesn't have to be that accurate.

I realize the point, I think it's the wrong way to lose weight (and I say this as someone who lost considerable weight and has kept it off for more than ten years).

Counting calories temporarily tells someone how much he is overeating, but the problem is the amount he eats is an ingrained habit. Habits die hard, eating less for a couple months doesn't change anything for most people. There is a lot of evidence that most people who diet put the weight back on because they took temporary measures to lose weight instead of making deep lifestyle changes. [Anecdotal evidence alert]: making deep lifestyle changes was one of the hardest things I've ever had to do, and I believe counting calories would have distracted me from that goal.

One can't address a problem until one is aware of it; counting calories is a technique useful for creating that awareness. It's not a diet plan, it's an awareness technique.

Counting calories is a short-term means to an end. It's intended to get you to a target goal. Once there, good lifestyle choices should take over.

Whatever works for you, go with that. But weight and body fat will fluctuate naturally and shouldn't be a huge factor in how you decide to eat. You could eat all kinds of crap that would keep you low-weight and low-fat but wreck your body.

Limiting yourself to reasonable portions of healthy foods 85% of the time is the easiest way to keep a long-term diet. Want pizza and beer? Save it for Friday night. It can't get simpler.

I rely on more of a feedback loop. I eat when I'm hungry, and I monitor simple metrics like my weight, sense of well being, endurance, and physique. I adjust my food intake (but still keying off when I'm hungry), and watch the results.

The problem with relying on hunger is that, for many people, that hungry signal doesn't shut off at the right time. The Hacker's Diet[1] goes into this in great detail, but the gist is that scale weight + calorie counting is a good proxy for replacing a broken hunger signal.

This is one of those things that's really hard to understand unless you experience it, so don't worry if you don't get it.

[1]: http://www.fourmilab.ch/hackdiet/

Nah, I can understand that. I'm actually mildly hungry much of the time; part of tuning that feedback loop is identifying what level of "hungry" to consider, well, hungry.

If you add up all the error margins in the article (and they don't even mention that the FDA allows published calorie counts to be off by 20%) you end up with an error margin that makes the resulting number pointless.

I know the hacker in you wants a number to measure but if you know anything about math at all you should concede that calculating numbers which such a high error margin is indeed pointless.

>Good article on why counting calories precisely is pointless and misleading.

It's really not, you just need to eat healthy relatively unprocessed food for which it is reasonable to estimate calories and start with a baseline calorie target and adjust according to your goals as you go.

This subject happens to be uh...my job...so I can answer questions if anyone has any.

The article seems to contradict that - unprocessed foods are the hardest to estimate. Pounding, grinding, cooking free available calories, bringing them closer to the published number.

I'd say portion control is a bigger problem than any of the things you just listed.

The amount of effort it takes your digestive system to process raw chicken vs. cooked chicken is obviously different, but the fact that they chose to focus on these issues doesn't make them more impactful than portion control.

I was thinking in terms of cooking at home versus going to eat at a restaurant. Restaurants are far more egregious in terms of being calorific as well as highly variable and prepackaged food usually isn't that good for you.

I really think the "unprocessed food" thing is a heuristic that has lost its power. The problem is that it is completely possible to make unprocessed food that is terribly unhealthy. And unfortunately, as with processed food, the cheapest ways to make unprocessed food tastier also make it less healthy. Keep in mind that honey, which many people would consider "unprocessed", is very nearly identical to HFCS-55 (and no, there is not good evidence that the slight differences between them significantly impact their health effects).

So early on, "avoid processed food" was a good way to guess at which foods were likelier to be healthy without having to actually examine them closely. But the heuristic is highly exploitable (i.e. you can trick people into thinking unhealthy food is healthy by making it from "unprocessed" ingredients), and in a relatively free market, it was inevitable that it would be exploited.

Now, if you advise people to either eat raw, unmodified ingredients, or make their own food out of raw unmodified ingredients, then that will still help them. But I think most people hear "avoid processed food" and think "Oh, I should buy these cookies that are made of whole grains and honey", and that is not helpful advice anymore.

OK, my take on this is that I need a framework to operate within. The frameworks I have chosen are Weight Watchers because it gives me a daily food budget which I can fine tune depending on whether I think I am losing enough weight (0.5-1Kg/week). The specific food choices are informed by Glycaemic Index principles which ranks foods based on the body's glucose response and hence provides a measure of how effective the food item is at keeping you feeling fed. I get the impression that GI is not highly regarded in the USA, if so why is this?

1. The comment above says "upto 20%" - that seems a bit high, no? Is it not possible to make better estimate of calories?

2. Any thoughts/experience with raw foods? (also juicing, water fasting etc? someone documented their water fast, I can't remember the URL now. It sounded dangerous, but also awesome to know the results)

3. For a moderately active person, which is better? Eating 2 huge meals a day, or 5-6 small meals?

Thank you.

> The comment above says "upto 20%" - that seems a bit high, no? Is it not possible to make better estimate of calories?

You could claim a different number but you'd be fooling yourself. Pick a number that seems reasonable relative to BMR and your goals and adjust as you go. That's kinda the whole point of my company is helping people make these adjustments without going through the hassle themselves.

Humans learned to cook because it allowed us to extract nutrients from food more efficiently and thus survive. If eating too much is your problem, tending towards raw food might help (but not to the exclusion of proper meals). If you're an athlete, not so much. I wouldn't take any one movement or fad all that seriously.

Juicing is a fad designed to bilk money from people, any diet/fast predicated on getting the majority of your calories from nutrition-less sugar is loony and bad for your health.

>For a moderately active person, which is better? Eating 2 huge meals a day, or 5-6 small meals?

It depends on what moderately active means. I think this is highly individual but I'll tell you what the "usual" is.

Strength trainers will generally eat 6 meals, but that's so that they can fit more calories in without getting stuffed.

Runner eating habits are highly variable and have a lot to do with glycogen replenishment. Carb-backloading before going to bed is a current fad for both runners and strength-trainers, but it seems pretty similar to the carb-loading stuff from the 80s, but with the benefit of more rigorous testing and study.

This is one of those things that require tinkering and experimentation. Some people eat 2 meals, skipping breakfast, some eat like hobbits. If you're trying to lose weight, optimize for eating habits that prevent snacking as well as over-eating during meals. Portion control is paramount.

Thank you for answering.

getting the majority of your calories from nutrition-less sugar is loony What did you mean by nutrition-less sugar? Vegetable/leafy juices (celery, spinach etc) have lots of nutrition, correct? (much more than just a fruit juice).

Sorry, there was a bit of sampling bias there. Most of the juicers I know will lean heavily on juicing fruits, not vegetables/leafy things, for the sake of flavor.

It's fine to juice things, especially if it helps you to get vegetables in your diet, but avoid fruits and excess sugar. I know too many juicers who end up getting like 70% of their calories from sugar because they're sucking down glorified fruit smoothies all the time.

> relatively unprocessed food for which it is reasonable to estimate calories

What I got out of the article is unprocessed foods are the least accurate when it comes to calorie counts because they aren't digested as easily.

What's your job BTW?

What would help me is if it were easy to track this stuff.

When I get the receipt at a restaurant, why isn't there a nutrition label printed on it, like the ones in grocery stores? Why aren't there QR encodings of nutrition labels for easy scanning by your choice of tracking app?

I see opportunity for improvement here :)

The article mentions that the potatoes and beef were organic (I assume as in organic food):

> Carmody fed adult, male mice organic sweet potatoes

What is the significance of this?

I'd assume making them organic helps to control for a pesticide or herbicide in them that could be changing the result. Not the best control but better than nothing.

I would assume the molecular makeup of food is modified by what you use to grow it. An organic potato might be smaller and more nutrient-dense than a chemically-grown one.

Has anyone here tried Seth Roberts' Shangri-La Diet?

Tried it. The main side effect is how silly people will think you are when you explain it. You do end up eating less, but I think a lot of the responsibility goes to the flavor-free periods you need in the day.

All any fad diet boils down to is "how can I eat like a pig, not exercise and stay thin?" and sell books about the diet.

>nine calories per gram

<imperial unit> per <metric unit>.

Must be an American article.

It took you that long into an article on 'Scientific American' to realize it was an American article?


calorie: the amount of heat required to raise the temperature of one kilogram (sometimes, one gram) of water one degree centigrade.

It's not an imperial unit, but it's not an SI unit either. One glaring example of it's inappropriateness is that the 'calorie' varies depending on what you're measuring (as you say, sometimes a gram, sometimes a kilogram). The SI unit is the joule.

I'll accept that. The calorie / Calorie / Kilocalorie confusion is pretty annoying.

Apparently EU food labeling may be in either kcal or kJ (at ~4.2 kJ/kcal).

Australian labelling must be in kJ (but it will often have the kcal amount listed as well)

Nah, mixing imperial and metric is a canadian thing. They even do it in aviation, which I find insane...

Everyone is an expert on diet.

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact