Why Calorie Counts Are All Wrong | Hacker News Reader

Why Calorie Counts Are All Wrong

scientificamerican.com

49 pointsDanBC96 comments

For the purpose of losing weight it really doesn't matter if the calorie counts are accurate or not: consistency is what is important.

If you want to lose weight reliably, track your weight and your caloric intake consistently. If you aren't losing weight, adjust your caloric intake until you are losing weight at a steady ~1-2lb per week.

Losing weight is simple...it's not easy, but it's very simple.

edit: You also should continually track this because as you lose weight, your body uses less energy and your maintenance level goes down meaning you may need to adjust your calories down a bit to compensate.

Right, I fear that this is another article that, while interesting, will be used to needlessly complicate discussions about weight loss. Paying attention to the (inexact) reported calorie information on the food you eat is still the only reliable way to lose weight.

Yup. The same goes for gaining weight of course. I've heard people say "I can't gain weight no matter how hard I try or how much I eat!". Wrong. Again, track your calories and increase the number of calories you are eating until you begin to gain weight.

> reported calorie information on the food you eat is still the only reliable way to lose weight.

Carbohydrate and sugar reductions seem like a very reliable way too. There are many anecdotes about successful weight loss on diets that allowed the person to "eat as much as they want" but had to limit their daily intake of carbohydrates.

This is just a way to reduce calorie intake, indirectly.

There's a fundamental difference between these approaches which is not highlighted enough in discussions about weight loss. There is a physiological factor: weight loss is attributable to calories in vs calories out (despite those figures being difficult to obtain). Then there's the psychological factor: some diets are easier to stick to than others, and some will naturally cause you to eat fewer calories than others.

Using reported calorie information is a physiological method, and appeals to those who prefer quantifying and tracking their everyday life. Carbohydrate reduction is (in this sense) a psychological method, and appeals to those who struggle with cravings and a lack of structure.

Limiting the daily intake of carbs is a proxy for limiting the daily intake of calories.

1 bowl (2 cups of rice, about 330g, more or less 1 serving in areas where white rice is a staple food) have a 331 kcal [1] Even if one would ate a whole package of Canadian Bacon (6 oz, 139g) it would only be 257 kcal [2].

Carbs are very easy to gorge. A medium sized fast food cup of soda is 180kcal. Same cup for raw orange juice is 223kcal. For comparison 150g of beef is 274kcal and that would be a big serving (IMO).

When people cut carbs from their diets they end up cutting the most densely caloric food: pasta, rice, starch and sugar and that's a good thing too, even eating as much of the rest as they want they will hardly be able to overconsume calories.

[1] http://ndb.nal.usda.gov/ndb/foods/show/6497?man=&lfacet=&cou...

[2] http://ndb.nal.usda.gov/ndb/foods/show/2636?fgcd=&manu=&lfac...

"For the purpose of losing weight it really doesn't matter if the calorie counts are accurate or not"

Pretty basic math tells you even a very small percent wrong (1-2) would affect weight gain/loss over the course of a year :)

I'm also not sure what you mean by "consistency". You say " If you aren't losing weight, adjust your caloric intake until you are losing weight at a steady ~1-2lb per week."

How can you adjust your caloric intake if you don't know the real numbers? I guess you are assuming you eat literally the same thing every week, and drop stuff on the floor until you have a set of foods you know make you lose weight, and then only eat that for the rest of your life? That seems a bad strategy to keep weight off. Most people would probably be willing to eat less of something (which they sanely could if they knew the real calorie counts), but probably not change their entire diet and keep it steady forever.

Even if you just reduce the number of calories to try to account for all the inaccuracies, you are still in a situation where you actually have no idea what you can or can't eat. IE i you say "well, 1500 calories a day didn't work, i'll try 1000", in practice, some days you will have eaten 1500, some days, 900. Your claim this will just average out magically is patently false (it only averages out over time if the numbers are less than a certain percentage away from right, and that percent is very small).

If you are eating different combinations of things, the likelihood you will ever find which are wrong/right, so you know to eat less/more of it, seems really slim. You still have zero idea what you can eat for real, and you'll still find if you swap what foods you eat, you may start gaining/losing weight. So "consistency" or whatever you want to call it, doesn't make any sense.

Now, i'll grant there may be no other way to accomplish your goal right now, but suggesting this is a good method is wrong. It's a horrible method. This is like trying to optimize a multivariable constraint problem through random guessing.

as long as the calories on the labels are correlated with the real numbers it works. plus, its not like you're going to adjust your calorie consumption by 1% to gain/lose weight. if you're altering your consumption by 20-30%, you can tolerate small amounts of errors.

also, counting calories works. anyone that has actually done it can tell you that.

> also, counting calories works. anyone that has actually done it can tell you that.

We have years of experience to show that across a population simple calorie counting does not work.

It depends on how you measure whether or not it works. In the short term, it is absolutely successful in causing weight loss. I've done it, I know others who have done it, there are people who make their livelihood helping others do it.

On the other hand, the problem with calorie counting is that most people only do it for a very short time, because it sucks. It's totally unsustainable for most people because they don't have the time or aren't willing to change their lifestyle to include hours per day spent focusing on nutrition. So they lose some weight over the course of several months, then let their nutrition lapse and gain it all back in a few weeks. So in the long term you're right, across a population calorie counting doesn't work. But only because across a population people are lazy.

We know that other approaches work better, and many of those approaches avoid counting calories. They work better because they focus on life-long behaviour changes; and on being maintainable.

You don't need to know the real numbers...the estimates on foods are "good enough". You don't have to eat the same thing everyday, just be consistent about logging every calorie you eat and stop worrying if the estimates on the package are 5-10% off or your body is 2% less efficient than mine...that all gets lost in the noise. If you track your calories for a month and find your weight stays exactly the same while you are eating 2500 calories per day of whatever foods you normally eat, then reduce your calories to 2000 without changing your diet you will lose weight at roughly 1 pound a week. Simple, not easy.

I don't know, if losing weight is so simple, I'd assume obesity rates won't be as high as 80% in USA during the "diet/health" craze phase we have going on. Theres gotta be other factors that we just don't know about, like metabolization, which will be ultimately what drives obesity down.

obesity rates are high because people don't try hard and consistently to lose weight. people lead busy lives, diets are not fun, and its very easy to eat a high calorie diet. diets don't work because people don't stick to them, not because diets don't work.

"Simple, not easy".

You track your calories, you track your weight. If you're not losing weight you reduce your calories until you are. This method compensates for peoples differing metabolism and activity levels.

It's key in how he summarizes it at the end: "it's not easy, but it's very simple."

Losing weight isn't complicated. Eat less, move more. That's it.

Losing weight is extremely difficult. The drive to eat is a fundamental instinct. That instinct isn't calibrated for constant easy availability of food. We inherently want to eat more than is good for us in an environment of abundance. Fighting that requires a great deal of willpower, which is a limited resource.

Successful weight loss comes down to figuring out how to reduce your intake and increase your movement without requiring more willpower than you can actually muster. This is really hard. But losing weight is still simple.

I think "other factors" could well include the reluctance of the average human mind to undergo consistent discipline, and the reluctance of health food corporations to concede that it pays off if successful.

Exercise is good for you and it seems to be pretty damn simple, but we are way too sedentary. Would you similarly suggest that exercise therefore must in fact be a lot more complicate than it appears?

I would suggest people are overweight because the steps to lose weight are unpleasant, or require willpower, rather than being too complicated and mysterious.

One of the most insightful studies I've read on dieting is that people trying to lose weight underestimate how many calories are in the food they're eating. It's not surprising once you start logging everything you eat -- you'll see that a lot of common foods are extremely calorific.

I can't find the exact study but here are two on the matter

http://www.ncbi.nlm.nih.gov/pubmed/16954358

http://www.bmj.com/content/346/bmj.f2907

To give it a car analogy, it feels like the weight loss discussion is asking: "Why are there so many automobile accidents?" Then come a lot of replies in that say "When you hit the brake, the car will slow down!" I feel like there are multiple things going on at different levels that so many people don't/can't hit the brake at the right time. That doesn't make the calorie-in-out statement false, and any solution will certainly involve reducing calories in. But, it suggests to actually get the weight of the overall population down, we need to look a bit more broadly.

I hear this a lot. I doubt it's sustainable. A lot of people who calorie count seem to regain eventually.

None of the long term fit people I know count calories. They tend to do a couple things:

1. Have well functioning hunger, which comes from 2. Eating whole foods and 3. Eating high satiety foods

I'm pretty steadily at my lean weight. But when I'm not, I just switch to higher satiety foods and the weight comes off.

This method is both easy and simple, but I rarely see it recommended. Are there people it doesn't work for?

(Of course, the hard part of this method is having a personal dietary culture that includes buying, cooking and eating whole foods.

Note that I'm not including cost, because in areas that aren't good deserts whole foods tend to cost less than processed foods)

Whether or not it's sustainable is a different topic though.

One of the reasons people regain is that they've gone on diets with a calorie intake that is too low (or lacking in protein) and experienced muscle atrophy rather than fat loss. A decrease in muscle mass also decreases their metabolism, which means their BMR is now lower, but because they don't know, when they return to their seemingly normal calorie intake, they'll quickly put the weight they lost back on.

That's a factor, but even if you just lose fat, you calorie expenditure from the same activity pattern, ceteris paribus, will be reduced (because you are carrying less weight), so if you return to the eating pattern which at which you were stable at the higher weight (assuming no other systemic/metabolic changes, just the fat loss) you will gain weight asymptotically approaching the the weight at which that "normal" calorie intake had you stable before.

Muscle loss from calorie restriction is much less than most believe (marginal until you get around sub 10% bodyfat). Same goes for metabolism slowdown. You will lose strength on a calorie deficit, but the muscle tissue remains.

Google "starvation mode myth"

See also this military study where men on extreme calorie deficits and high activity levels did not lose lean muscle tissue until hitting extremely low (sub 5%) bodyfat levels.

http://jap.physiology.org/content/88/5/1820

What about people who don't have much resistance activity? That's the more typical case.

> I'm pretty steadily at my lean weight. But when I'm not, I just switch to higher satiety foods and the weight comes off. > > This method is both easy and simple, but I rarely see it recommended. Are there people it doesn't work for?

I would hazard a guess at: most people with weight issues. The problem is not how simple the method is, but how psychologically easy it is to adhere to.

4. Willpower and discipline.

Less than you'd think. On the plan I described you pretty much just eat when hungry, until you're full.

You do need to say "potato chips and candy bars are not things I eat", or have some other mechanism for limiting those without willpower (dessert used to fill this function).

But I hardly need willpower to eat this way, because the body's system handle telling me when to eat.

Well, anecdotally, as someone who lost 100 pounds and kept it off for several years, the mechanisms in place don't work for everyone. I can easily eat a whole "American-sized" restaurant meal without thinking, whereas lean people tend to feel completely stuffed after eating, say, half a sandwich and a few fries.

Also, I get hunger sensations pretty much all the time, no matter how much I eat or how satiating the food is. The worst is that I actually get hunger sensations for quite a while after I eat. I suspect this can be the case for many obese people. For me, I need to use my eyes to regulate my eating rather than my hormones.

The amount of willpower necessary can vary considerably from one person to another, and for people like me just 'eating when hungry until I'm full' is not an option.

But you're definitely right about focusing on better food quality rather than simply eating less. It's far too easy to go overboard eating plates of spaghetti and meatballs than it is grilled-chicken atop salad. After I lost weight and cleaned up my diet, I now perfer the latter to the former. When my diet is consistently clean, it's easier to maintain the cleanliness. There's a lot of inertia when it comes to eating and controlling the velocity is far more valuable than just changing the speed.

Very interesting, thank you for your comment. I'd like to know more about hunger disregulation: does it cause initial weight gain, or is it the result of carrying extra adipose tissue, and does it reverse under any conditions.

Also I agree with your point about consistent cleanliness. I find it very easy to avoid eating something if it's a habit. I find it very hard to "limit" a food except by hunger. And hunger doesn't work well with junk, even when hunger is functioning normally.

"This method is both easy and simple, but I rarely see it recommended."

It's hard to make money marketing this method. I'm not being cynical it's just fact. If you look at weight loss "solutions" they are based around products that interact well with a supply chain. They are highly processed, loaded with preservatives, and individually packaged. As a result, these solutions tend to be based on methods of calorie restriction that don't provide satiety. Additionally companies are incentivized to create a solution that helps people maintain lost weight _only_ when they are on the company plan.

That doesn't make any sense, unless you're consistently eating the same foods prepared the same way.

As I've said elsewhere, calorie estimates on packages are close enough that it doesn't really matter what you eat as long as you estimate consistently.

That still doesn't make sense. One example cited in the article says that for almonds, 129 calories were absorbed out of the 170 listed on the label. That's a 25% difference. The article unfortunately doesn't give a whole lot more info on how this tracks across various foods under various methods of preparation, but it would seem to me that if this difference is highly variable, it basically defeats the idea of fungibility of "label calories", which seems to me to be the entire point of listing them in the first place. Granted, if you're basically eating the same stuff, but just increasing or decreasing helpings, that's not such an issue.

People might disagree with my reasoning here, but I'm not really seeing how I deserve downvotes.

That 25% difference is for a single serving of food and only translates into a ~2.5% error in your daily food total. People eat a varied enough diet that a 25% difference in a single food item really won't make a huge difference.

Lets take a man who needs ~2500kCal per day to maintain his current weight. A nice steady loss of 1lb per week would require this guy to eat at roughly a 500kCal/day deficit (a total deficit of 3500kCal translates to a fat loss of 1 lb). Even underestimating his total caloric intake by 10% (he eats 2200kCal instead of 2000kCal) every single day will still see him losing weight at .6lb per week.

After 4 weeks of this he can see that he's only lost 2.4 lbs and can therefore reduce his intake by another 200kCal/day to achieve the desired 1lb/week loss. Tracking your weight/calories like this makes it fairly trivial to lose weight. It really doesn't matter if all the individual food items have perfectly assessed caloric content...as I say, they're close enough.

They say "all wrong" but they appear to mean "not perfectly accurate." Given that calorie counting is one of the most successful methods for gaining control of weight, that's not just semantics; that's unethical.

Additionally, as far as I know they typically overestimate the caloric value. Which means the error will almost always work IN FAVOR of someone calorie counting to lose weight.

The author is to blame for this unfortunate headline. "Humans engage in a kind of tug-of-war with the food we eat, a battle in which we are measuring the spoils--calories--all wrong."

Interesting, but still mostly fiddling with details and niche cases. I found that simply "counting calories" (more like roughly keeping track of them in the back of my head) as written on the labels and then reducing my daily consumption by 25-30% to be the only reliable way to lose weight. No hassle, you can eat what you want and you aren't even that hungry. Those labels might be off by 10% or even 30+% at times but, ultimately, that should barely matter.

For weight, it's always calories in vs calories out.

Scientifically sound food recommendations [1] accept that different types of calories make you feel more or less full.

If you supplied your entire day's calories in sugar, you'd be insanely hungry.

While short-term weight loss studies from the 90s focus on calorie counts, modern studies take a longer term view and examine how well weight is maintained based on diet.

This is where the type of calorie becomes important for most people in weight regulation. (vitamins etc are another topic)

[1] http://www.hsph.harvard.edu/nutritionsource/healthy-eating-p...

"At the end of the day, it's always calories in vs calories out.

Scientifically sound food recommendations [1] accept that different types of calories make you feel more or less full."

That's a contradiction. The entire idea of "calories in, calories out" is precisely that there is no difference and all calories are exactly the same. If some make you feel less full than others, then you're invoking the existence of the feedback loop that "calories in, calories out" explicitly denies.

If you, as this article does, point out that many things affect whether a calorie can be absorbed, that is, again, another difference between calories.

If modern science is moving on... and it is, at an inexcusably sluggish pace... it's moving on from "calories in, calories out". There's really no way to salvage it... even if you want to go with "calories digested vs. expended" you've still got a complicated feedback loop even there between the pancreas, insulin, blood stream, various bits of the brain, the metabolic consumers of blood sugar (significant differences here too between muscles vs. organs), etc, and even two people who hypothetically start with the exact same blood sugar levels and digestive system state will rapidly diverge as this feedback loop does different things between them. "Calories in, calories out" is only true as a limit on the system of the bodies, it isn't even remotely suitable for use as "the only parameter in a model, which gets set to a constant value". (What other bodily function works so simply? What makes digestion so bizzarely unique that it can be modeled so simply? None and nothing.)

(And a rather pointless limit in some ways, too... nobody is running around claiming that the human digestive system is an over-unity perpetual motion machine, so making such a big deal about thermodynamic limits is rather silly.)

That's not true. If some calories are not absorbed, then they're not really "in", are they?

This partially expresses what I wanted to reply.

The other part is that food calories are the amount of energy needed to raise the temperature of one kilogram of water by one degree celsius. By that definition, calories in vs calories out seems pretty darn close to reality, and if it's not perfect, it's likely so close you're not going to notice a significant difference.

> That's a contradiction.

No, its not.

> The entire idea of "calories in, calories out" is precisely that there is no difference and all calories are exactly the same.

The idea is that they are the same to the extent that net calorie consumption, irrespective of source, determines weight loss. Effects on satiety and other effects which influence the ability of the dieter to adhere that come with different calorie sources are not ruled out by "calories in vs. calories out" approach, which doesn't hold that all calories are equal in there influence on your ability to maintain a diet regime, but only that they are equivalent in their effect on your weight when you consume them.

This is not to say that "calories in vs. calories out" is right or useful, just that this particular effect isn't inconsistent with it (though it is one of the less important reasons its of limited utility.)

That doesn't make sense. If calories are all that matter, then simple math shows that being off by 2% over the course of the years easily leads to weight loss or gains. You've seen this math numerous times in the context of "if you just didn't eat one donut hole a day you'd weight X pounds less a year". If the labels are off by 10% or 30%, then it can't be just "calories are all that matter"; you'd be losing or gaining a dozen pounds a year, unexpectedly vs. what your projections expected.

If, of course, the body is a dynamic process that reacts to what it encounters in the world, then of course the dynamic reaction is what dominates the situation, not percentages of error in calorie counting.

(The way the nutritionists managed to turn "The body reacts dynamically to incoming input" into unthinkable heresy in favor of a model in which the body is essentially modeled as a hapless passive recipient of "calories" with no feedback loop will someday be seen right up there with pholostigon theory, except less excusable. Any time you speak of a bodily process and aren't thinking about the feedback loops, you've bounded your best-case outcome to "utter failure" before taking the first step.)

This x10. I lost 43 lbs in 6 months upping my calorie intake by 500-1000 calories a day. But I drastically changed what types of calories I ate. I also halved my triglycerides, and brought my HDL and LDL levels far into the healthy ranges.

All by cutting out grains and sugar and replacing them with lots and lots of vegetables and meat (I'm lying, it was mostly bacon).

To a person who vehemently believes in the thermodynamic weight loss hypothesis (calories in minus calories out!), it would be more convenient if your anecdote, and others like it, did not exist.

I think the point is that if you reduce your possibly inaccurately calculated calorie intake until you start losing weight and then keep your intake steady it doesn't matter if that value is 1000 or 100,000. Over time the method will definitely work.

To calculate weight loss or weight gain (by counting calories) you need to know how many calories you eat and how many calories you burn. Since the latter is nearly impossible to guess, it's usually estimated with the first one and weight loss during the first month.

This automatically adjusts to any error in the calorie estimation.

Lets say someone eats 2000 kcal a day, but the "real" number is actually 1900 kcal. Lets also assume that in fact his weight maintenance level is 1900 kcal. After a month the person has neither lost nor gained any weight. This person will then incorrectly assume that his true maintenance level is actually 2000 kcal, and reduce his estimated calories to 1900 kcal, thus starting the weight loss process.

Yes, it would be nice to get perfect estimates, but it doesn't matter much in the weight loss or weight gain process. In addition, most people don't have a deficit or surplus of a mere 100 kcal, it can be 500 to 1000 kcal, which makes the error in "true" calories even less important.

It works exactly the same for me. I approximately know how much calories I should eat during the day and if I keep below it I slowly but steadily lose weight. It does not even matter what I eat, as long as I stay below the threshold I will lose weight.

At the moment though I am gaining weight. My life is stressful at the moment and one way of dealing with this is eating tons of food. Because of this I gained about 10 kg during the last year.

I still don't see this as an argument against calorie counting. When I over eat I know that I'm getting more calories than I need. It doesn't matter if the stem of the plant I eat is harder or easier to digest when the average calories consumed is higher than the average calories needed to lose or maintain weight. I really wish people would just admit as much and not fight in the margins for a completely wrong theory.

> Proteins may require as much as five times more energy to digest as fats because our enzymes must unravel the tightly wound strings of amino acids from which proteins are built. Yet food labels do not account for this expenditure.

That's not quite true. Protein has 5 calories per gram, but labels only count 4 because of this fact.

> That's not quite true. Protein has 5 calories per gram, but labels only count 4 because of this fact.

Can you give me more information about this? I've always wondered whether or not calories listed are "net" calories (as in the number of calories you get after you extract calories from the food) or "total" (as in this amount minus the amount it takes to extract these calories). I'm just not sure what to google to read more :)

I found a Wikipedia article with the actual amounts from combustion:

https://en.wikipedia.org/wiki/Atwater_system

Cool, I'll take a look, thanks

Isn't this sorta obvious? I always assumed the calorie counts in foods were old data, calculated a century ago, and the foods labeled with those counts have been so modified since then that they are little more than relative guesses useful for moderate comparison only. Add in the variances of how different people digest, as well as how food combining impacts digestion, and those counts are marginally useful at best. "Calorie" in foods have little to nothing to do with "calorie" in chemistry and physics, other than they were originally intended to be the same, but any thinking individual directing attention to them sees they are not.

I truly believe that calorie counting is just a big waste of time. sure you can pin point all the macros your eating, but unless your training for a competition or are a pro bodybuilder, than I believe just eating healthy and eating a lot each day would help you gain muscle. some good information on proper dieting is on this website I recently check out called http://bodyweightburnreviews.com/ See for yourself If your interested...

Despite the fights breaking out over accuracy I still found this article insightful. Eating 170 calories worth of broccoli vs 170 calories of honey does have a strong difference. I'm sure you can still loose weight by counting calories and adding a little exercise, even if you're adding honey to your diet. But this article at least explains why you will progress better by eating stuff that is a lot less processed.

I heard some of this already 30 years ago when I took classes in Nutritional Chemistry. It's a highly complex system which is hard to measure and often highly dependent on the individual. Today we know more about the gut organisms which helps explain some of the variability, but like the brain much of what we think we know is still speculation and often anecdotal.

Plus the yuck factor makes experimentation not much fun.

Of course, you have to keep in mind the form those calories are coming in. Fats, though energy dense, are pretty much slow release. Fructose, on the other hand, is just nasty stuff that leads to a build-up of visceral fat. Both of them contribute to your calorie count, but they both affect your body in very different ways.

Dietary fat contributes to visceral fat as well, and at a greater degree than fructose. There's always some energy lost during DNL.

I recommend watching That Sugar Film[0], It addresses this in part because as it turns out fat calories != sugar calories.

0.http://www.imdb.com/title/tt3892434/

How does the "eat less is simple" crowd rectify with the information seen in this same space that gut flora can be the true difference between obesity and airbrush candidacy?

This article is, in summary, a guy going "Well, actually, ..." about a tiny, but irrelevant detail.

For some time now i have wondered what happens if one only eats corn for a week.

Corn has to be pretty processed before it becomes edible -- cornmeal, corn mash, corn mush, corn cakes, corn flour -- all of these are highly processed versions of corn. But simply grilling corn on the cob? You'll see where most of your calories are going when you flush!

This article takes up a common position that calories are "wrong" because, due to our bodies' complex natures, such an arbitrary measure ignores a lot of what is going on. And yet, every time this view is expounded, someone comes up to calories' defense, saying that they are a reliable measure of nutritional energy. So which one is right?

While it is true that calories don't measure the nutritional energy absorbed in any single meal, that would be a trivial proposition. Calories are a statistic, that is, an estimate of an unknown but posited number, and partial representation of an underlying distribution. Saying that calories are "wrong" in this sense would be merely stating that they are possibly wrong. And they would be wrong, if the underlying processes are somehow consistently correlated with each other, which would produce inconsistent and biased estimates of caloric content.

Yet, I don't have any reason to believe that the probability that I buy some potato cultivar today is correlated with the composition of my gut flora three weeks in the future, much less that it would be consistently so. On the contrary, the factors supposedly affecting my nutrition according to biologists are so many, and so diverse, that I might as well appeal to the central limit theorem, in which case calories would be just as a good measure as we can have. If today some potato gives me 166 kcal, and tomorrow 158, and further on 163... I don't see what's the issue. The likelihood of a critical event, such as starvation, due to misreporting of calories or due to some feature of my constitution seems so low that I would only very likely trigger it if I ate nothing but one type of food, in the same quantities, for a long period of time.

Furthermore I will never understand those who prefer no measures to at least one rough measure. No things are created perfect, but only made better. I wonder would have happened it occurred to Eratosthenes that his measure would have been too rough, too imprecise, or too vague for his mind. If all ancient mathematicians were bothered that no "real" triangles were to be found anywhere, mathematics wouldn't have happened at all. By such strict understanding of measure and precision, of course measures fail. They couldn't possibly live up to it. An other example is personality theory, which is said to have been nearly abandoned at it's inception because some psychologists thought of it entirely in terms of archetypes, rather than a general description of probable behavior.

And lastly, if calories have been adequately defended here, then I'll also add that ogling at them in order to find what is wrong with nutrition theory seems completely wrong. We already know about calories, and they are a measure of what we know. Rather than looking at them, we should be looking at the measure of our ignorance, which is the error term. If some calories consistently fail to account for some observations, then it's because no one has looked inside the black box enough to figure out what is going on.

In short, it's a sensational headline. The only people who perhaps should be worrying if calories are "wrong" in the sense of not being accurate to the tenth decimal are Soylent-ers.

Reply on news.ycombinator.com