By which I mean: this is much more likely to be effect than cause, e.g. overweight people might have more sugary diets, so sugar-loving bacteria would be favoured.

http://norvig.com/experiment-design.html

being fat is not even about overeating, as long as you put in the effort to workout to burn those calories, you can eat whatever you want and still remain in shape.

You won't get a sixpack, since that requires a proper diet, but you should still be in decent shape.

Do the math. It's not too difficult to snack along, have three meals a day, and end up with 6000-8000 calories. Assuming you're a big guy, you might have a 2500-calorie base metabolism, and burn maybe another (being generous) 1500 calories a day in normal movement. That means you have to exercise specifically to burn 1000-3000 calories a day. If you run for 800 calories an hour, you'll have to run for hours every day just to burn off what you're eating.

It's completely impractical to try to burn as much as you could eat.

Every obese person I've ever known ate crap and didn't exercise at all. And when they tried to get thin they would either half-ass it or do the latest fad diet for 2 weeks or something and get right back to where they were later.

Well, sure. That's what "ate healthily" means, and it's how you determine that they're doing everything right: they're not obese.

I'm not suggesting that there's anything involved but intake and exercise; I'm responding to the idea that you can eat as much as possible and just burn it off with a little extra exercise. That's totally not doable.

And just because a person eats a little amount doesn't mean much, in my experience it more depends on what it is you eat. You can eat about as many buckets of broccoli and chicken as you want and stay fit compared to one who nothing but chocolate in smaller amounts or whatever (choose some unhealthy food).

And also I'm sick and tired of people who rely on genetics as their problems. Go into the household of an obese family and there on their dinner table you will find tons on unhealthy foods and snacks. In each new generation of this family they grow up accustomed to eating in this manner and are obese more because of that than anything else.

And when you're young you're both active and have a metabolism that's burning calories like crazy--that's what's keeping you fit. Not genetics.

Also as for obese people who eat little--I knew such people, or at least I thought so. Then during each meal with one such person I noticed how little they ate. Then once during late at night I went upstairs to the kitchen to see the light on--I looked around the corner--there was this such person eating nothing but ice cream and pie; and a lot of it. Just because you don't see it doesn't mean it isn't happening. In the same way people can hide their drug and alcohol addictions so too can obese people hide their food addiction.

You burn roughly 10 x (your weight in pounds) calories at rest, plus an additional 10% (of that number) for each activity level (office/couch potato = 10%, moderate exercise 4 times a week/average (1 hour each session) = 20%, strenuous exercise 4 times a week/average = 30%, dedicated athlete = 40%. The percentages increase with muscle mass, but not by much. These numbers are for base calorie burn; calories actually expended through physical activity are in addition to these numbers.

Hell, you'd have to be an Olympic or professional athlete to come even close to needing more than 5000 calories in a single day.

Here is the short article about Michael Phelps and his diet. Supposedly he eats 12K calories per day. http://sports.yahoo.com/olympics/beijing/blog/fourth_place_m...

An ultra-maraphon runner Dean Karnazes provided a sample diet for a day when he runs for the whole 24hrs in his book. I think it was about 26K calories.

These are the extremes, of course, but not impossible.

Phelps proves a point, though: so long as one uses the calories, there is no weight gain from fat.

Well, we stipulated three meals and snacking, and you've accounted for it already with just the meals! Say you also like potato chips and dip, and want to eat a bag of chips with a can of dip between lunch and dinner, and then after dinner. If that's all the snacking you do, you've added another 4500-5000 calories just in snacks.

Could you sit down tomorrow and eat all that? Of course, you probably couldn't, if you don't eat nearly that much now. But it's not hard to get there if you start eating because you're bored, or because you want to make yourself feel better, or whatever. Over the weeks or months, "I'll just have a little more this time so that eating lasts longer and is more enjoyable" pushes the amounts up until the calorie counts here seem only excessive, rather than completely bizarre.

Anyway, while running 5 miles a day still, I immediately began gaining 15 pounds a month. Eventually I couldn't run anymore.

I'm sorry, but the argument from thermodynamics is as bankrupt at explaining fat gain as it is in explaining height gain for adolescents. For height gain, the fact that energy in equals energy out is a true statement, but the explanation is growth hormone. If you calorie limit a growing child, the child will be somewhat (not very though) stunted, but will make up the energy difference through lethargy and by consuming all sorts of internal tissues.

For obesity, the explanation is insulin. I cut carbohydrates almost completely from my diet in January. Otherwise I ate what I wanted when I wanted (I measure my consumption in pounds of red meat per day). I'm amazingly healthier. I've lost 50 pounds. I've gained so much muscle that I can do pull-ups. Fucking pull-ups at my weight. With low-calorie dieting, I couldn't do pull-ups until I had lost another 70 pounds past that point. I'm putting on muscle and losing fat, and I'm eating huge quantities at every meal. Plus I have a desk job and barely exercise.

The argument from thermodynamics assumes that the fat tissue is a trash can and that weight homeostasis doesn't exist. It's a bankrupt theory that can't explain diseases like progressive lipodystrophy (a disease which makes people obese and emaciated at the same time). It can't explain why people who starve the calories off are not the same at the end as normal thin people and almost inevitably gain the pounds back. It can't explain why 1950s treatment to get anorexics to gain weight was through injecting them with insulin and not feeding them any more than they had been (during treatment).

In summary, you're seeking a psychological explanation of a physiological problem. Kind of like the discredited quack Sigmund Freud often did, eh? A physiological explanation might lead to the suggestion that we eat the way we evolved to eat -- no refined carbohydrate.

People almost always underestimate just how incredibly - almost unbelievably - efficient the human body is. We're used to being impressed by cars that can do 60 mpg, but the human body can easily manage 300 mpg, even for an unfit hacker like me; fitter people will be slightly more efficient, maybe up to 400 mpg (http://health.howstuffworks.com/health-illness/wellness/phys...).

What this means is that you have to run an enormous distance to burn off even a tiny amount of food - about 1 mile for every 100 calories. So you can run a mile on one banana (http://www90.wolframalpha.com/input/?i=banana), or 1.5 average cookies (http://www90.wolframalpha.com/input/?i=1.5+cookies), or just over one rasher of bacon. You can run a entire marathon (26.2 miles) on 2,600 calories - or one big mac and a packet of cookies.

A scary number of people in the western world eat enough to run a marathon every single day and still have enough calories left over of keep themselves ticking over (~1500 calories) - without changing their diet at all. This wouldn't be physiologically supportable for wear and tear reasons, but calorie wise a lot of us could support it.

These two facts in combination: the abundance of high quality, high calorie food in the developed world, and the incredible efficiency of the human body are a virtually guaranteed recipe for obesity, whatever other factors come into play. The fact that almost all of the easily available junk calories people consume come from highly refined carbohydrates (because these are so cheap to produce), massively compounds the problem.

Obviously there isn't a 1 to 1 relationship between calories in and weight gain and everyone is slightly different and most food isn't just converted into glycogen and then to fat, etc, etc... but as far as I can see, most people will not reasonably be able to burn off the calories from an average western diet, no matter how hard they try - there just isn't enough time to run that far and hold down a job. Trying to do that much exercise, on top of trying to diet and do everything else is just too much mental strain. On the whole, I think people are better off just not eating the carbohydrate in the first place - cutting it off at source is much easier than trying to get rid of it later, and only requires you to do one thing, rather than two.

How does the thermodynamics theory explain weight homeostasis? The vast majority of people throughout history were able to maintain fairly constant weights throughout their lives. If it's just calories-in, calories-out, they would have to be regulating what they ate to within 10 calories a day to maintain their weight over a decade.

I'm sorry, but this is silly science. You can't explain obesity by simple thermodynamics any more than you can explain growth at puberty by simple thermodynamics. Energy conservation, although true, tells you very little about self-regulating systems w.r.t. the environment.

I'm wondering if it is possible to inject some insulin into a person, then put him on the scales (denying him any food/drink) and observe the weight increase. Obviously that is not possible.

What if we take a person, force him to do some exercise so that he loses 1kg of weight, and then provide him with food/drink that weighs 900g combined? Will this person gain weight over the initial amount?

What I'm trying to say, I think, is that while the simple calorie calculations may not be able to explain everything, surely weight increase can not come "from nowhere" without the abundance of food/liquid.

This is not intellectually honest. It's not what he means, and you know it. Of course weight-gain comes from food, the problem is the relation between what you eat and what you store. If insulin or lack of can change the scales from 1:1 to 10:1, then it doesn't matter very much how much you eat any more, does it? Since you can realistically change only about 20-30% of both eaten food and burned calories.

I'm not saying insulin does this (though it might very well) but for the sake of the discussion please don't fight the wrong idea.

Two things I suggest you look at: Zucker rats. An obese strain of rat. You can semi-starve it from infancy and it will get just as obese. It will be smaller, but the body fat percentage will be the same.

The second thing to look at are overfeeding studies. It's very hard to overfeed people with protein. The only successful studies had to be done on prisoners, and they barely gained weight. With carbohydrates, it's the opposite. People would eat 10,000 calories a day and complain of being hungry. Weight gain is easy to introduce.

- Fat is not an energy battery. Carbohydrates are for energy. Fat is for providing nourishment when food is not readily available, and for warmth.

- Damage from starvation is not to fat, it's to muscles. Muscles are high-maintenance tissues. If there aren't enough nutrients to maintain them, they break down. It only takes a day or two of fasting for this to occur; the longer the fasting period, the greater the damage.

- Weight management is about calories consumed versus calories expended. If you want to call that thermodynamics, so be it, but it demonstrates that you don't understand what thermodynamics is, or that you don't understand how the human body works.

- My science is medical science. It's backed by prevailing medical opinion. Hell, I made sure to check WebMD and various other medical sites to make sure i wasn't pulling it out of my ass. It's very simple: eat more than your body can convert into energy or expel through various bodily functions, and that remaining mass remains in your body, as fat. Too much of that fat, and you have "obesity".

Puberty is red herring. However, medical science has shown that puberty is triggered by the quantity of food one eats -- suggesting a mass trigger for puberty, which explains why fat little boys and girls hit puberty early, but gymnasts look like little girls into their 20s.

Yeah. That's about as useful as saying that alcoholism is caused by over-drinking.

There are all sorts of diseases and conditions that cause obesity, and we know for a fact that the primary cause is not eating too much (eating too much or too little is the effect): menopause-induced weight gain, drug-induced weight-gain, cancer-induced weight loss.

The idea that the recent (post-1970) epidemic of obesity might have something to do with the massive amounts of refined carbohydrates we began consuming (in response to low-fat hysteria) is not a bizarre one. Sugar is a brand-new element to the human diet and now we're eating 100 pounds per year per capita.

And with your last statement, you aren't seriously denying the primacy of growth hormone in childhood growth, are you? Because otherwise, your argument about mass triggers is rather tangential.

- Most post-menopausal women are not obese. Indeed, most do not gain much weight post-menopause. - Drugs, by and large, do not result in weight gain. A few drugs do, but only a few. - Cancer weight loss is a product of the body not being able to digest enough food for the body to use for maintenance, usually because nausea induces vomiting before the food is fully digested. - As for the sugar: the problem boils down to (again) overeating caused by the taste of sugar stimulating a desire to eat more (sugary) food. In the end, overeating is still the problem. - Finally, as for human growth hormore: the effects of premature puberty have been documented in cultures where growth hormone is not present in the cows such as in India and China (among the portions of the population wealthy enough to over-eat). Thus, it is clear that growth hormone has a negligible, or at best minor, effect on obesity.

If we can explain vertical growth as a hormonal effect, and overeating as the result, than there is no a priori reason that a similar explanation is impossible for horizontal growth. In fact there is great evidence for exactly that explanation.

Your other objections were equally tangential. You seem determined to discuss tangents rather than the central points. And what evidence do you have for your sugar hypothesis? Sugar is no different from other nutrients, hmm? So glucose and insulin aren't related to FFA uptake by the fat cells. And glycogen from insulin isn't a necessary molecule for that process? I'm curious as to what research has caused you to declare all the textbooks wrong. It must be compelling.