One thing that seems to be lost in all of this is that this guy is a nutrition professor, yet he was overweight and had bad health markers. My biggest takeaway is that this was a general indictment of the state of nutrition science and education. They preach the wrong stuff (hi-carb, low fat, whole grains), and apparently can't even keep themselves healthy. There seems to be a fundamental disconnect.

Typical of the propaganda, they mention the junk food that he ate in the first paragraph and wait until paragraph 11 to mention:

"Two-thirds of his total intake came from junk food. He also took a multivitamin pill and drank a protein shake daily. And he ate vegetables, typically a can of green beans or three to four celery stalks."

IMO, it's just another piece of bad journalism. It doesn't even come close to being science.

I think Eva has hit on the most important rejoinder to this article namely:

it is just not sustainable long term.

I don't think this is just true psychologically, by physiologically. All the answers saying "Of course you'll lose weight if you just calorie restrict" need the qualification "within limits," lest we connote that any-one can not get fat/lost weight if they just choose to eat fewer twinkies. We surely know of lots of people with damaged metabolisms who have starved themselves to no avail. They can't all be lying. Typically just not eating so much can work in the short term, but not in the long term and I'm guessing that it stops working (you stop losing reasonable amounts of weight) before a lot of people cave in and decide to eat more twinkies to meet their calorie requirements. The commonest complaint online in any weight loss context is " X worked great for me for a a few weeks, but then I just stopped losing!

We've also seen from the starvation trials that holding people absolutely rigorously to a calorie deficit long term sounds easy but a) drives them completely insane (at which point, talk of "will-power" and "controlling behaviour" become irrelevant) and b) causes them to gain back even more weight than they lost in the first place. Hence the second most common complaint on the dietosphere, "I lost a lot of weight doing X, but then I gained it all back again, and more!"

I suspect that after more than two months the problem would not be that his finite store of "will-power" would suddenly collapse and he'd feel compelled to gorge on an extra 800 calories of twinkies (or consume an extra 800 calories of whole grains and bananas from his old diet). Rather, the problem would be that his body would adapt to continued starvation in the only way that makes evolutionary sense, namely reducing his calorie expenditure through whatever means possible, including burning through his muscle mass (incidentally, on his facebook page you can see he lost about 2kg lean mass in a month). Once his body successfully adapted to a continued intake of 1800 calories his efforts would be severely diminished, leaving him faced with eating an even lower twinkie diet if he wanted to continue to lose weight, at which point he would simply be pitched into a battle between himself and his body's 200,000 years of evolution).

Oh and btw a diet of doughnuts would plausibly give him a diet of around ~50% fat, doubtless substantially higher than his previous diet of whole grain and banana (hence, why he reduced trigs by 39% no doubt).

Looking at his one sample diet day he actually seems to have consumed 888 calories of carbohydrate, 531 of fat and 176 of protein, for roughly 1600 calories. The ratios would be roughly 55% carbs, 33% fat, 11% protein (and this was a two protein shake day!). Of course, granting that he normally consumed approx 2600 calories, what he was really consuming would have been closer to 34% carbs, ~59% fat, ~11% protein (more protein and less fat, if his body was burning more muscle and less fat).

I am sure if he keeps it up he will end up in a state of skinny fat.

Here are his blood results from his facebook:

Total cholesterol: Pre=214; wk10=184 LDL-C: pre=153; wk10=123 HDL-C: pre=37; wk10=46 TC/HDL ratio: pre=5.8; wk10=4.0 TG:HDL ratio: pre=3.3; wk10=1.6 Glucose: pre=94; wk10=75 Blood Pressure: pre=108/71; wk10=104/7

The LDL and HDL are obviously not optimal. He didn't post his triglycerides. The TG:HDL ratio would be considered good tho, I think. He hasn't done a small dense cholesterol reading but I am sure his blood isn't the picture of health he's flaunting it to be.

We also don't know anything about his quality of life. How's his digestion after consuming all the processed crap and gluten. Is he hungry all time? Experiencing sugar crashes? Not thinking clearly? Can he do a crossfit workout on this diet?

Haha, when I was younger I ate a Snickers or a King Cone EVERY DAY. I was skinny, had clear skin, and my blood markers were fine. It took YEARS for the effects to show.

King Cones >>>> Twinkies.

Yeah, if you starve yourself, you'll lose weight. He was starving himself, so he lost weight. This should not be controversial or surprising. I guess it does serve as a useful counter-example to magical thinking about how you can get fat eating very little food, provided it's the bad food. In the long term he'd probably kill himself on that diet but that's the long term. In the short term, there is no magic, and starvation works.

The key takeaway here, which I sort of wished was the main paleo message, is that behavior matters. IF you consciously modify your behavior, you can lose weight. If you just ate twinkies to appetite forever I'm sure you'd end up as a diabetic balloon, but if you carefully plan your twinkie consumption, you'll be able to force any outcome.

The power of paleo is the unconscious modification that takes place within the context of the paleo diet: you can eat to appetite because your appetite now functions properly and will regulate how much you eat. It doesn't work for everyone but for those who it does, congratulations, you now don't have to think about how much you eat anymore. Oh and if you have gluten or dairy issues you're probably happy too. And yeah, your long term health will almost certainly be better.

But again: when it comes to weight loss, especially large weight loss like going from 33% to 24% BF, your behavior is the key. Adherence to diet is the strongest predictor of success, and this just adds to that simple observation.

First of all, just because your cholesterol went down does NOT mean you are healthier. THere is plenty of evidence that cholesterol levels do not equate automatically with health. Plus the fact that trigs were not posted is suspicious. The other big thing is, sounds like he was eating a lot of carbs before the special diet, so it may not have been a big difference switching to twinkies. Is a whole grain really much healthier than a twinkie? Perhaps not if you eat a lot of whole grains but much less calories of twinkie. PLus he was still eating some veggies and kinda healthy stuff. I'd like to have seen his before and after macronutrient profiles, PUFA intake, and trig levels to get a better understanding of what is going on. And just lowering cholesterol means nothing. Lots of people with high cholesterol are healthy and lots with lower cholesterol have heart attacks and die.

As for weight loss, it is no surprise that heavy caloric restriction does work, but it is just not sustainable long term. He went from an ad lib probably carby diet to a restricted intake carby diet. So no particular surprise he lost weight. I agree with him that calories do count. But they are only one part of a very complicated puzzle.

I'd say, on any controlled diet, you are going to lose weight.. Especially if he had no retricions before doing this. Wether that weight stays off and if it actually healthy is another story.

Seth Roberts has a hypothesis that novel tastes cause the body to lose weight while familiar tastes cause the body to gain weight.
http://www.blog.sethroberts.net/2010/05/20/losing-weight-by-eating-new-food/

Here is what I think about the dumbass diet...

http://hbfser.wordpress.com/2010/11/08/the-prof-mark-haub-nonsense/

Before his Twinkie diet, he tried to eat a healthy diet that included whole grains, dietary fiber, berries and bananas, vegetables and occasional treats like pizza.

"There seems to be a disconnect between eating healthy and being healthy," Haub said. "It may not be the same. I was eating healthier, but I wasn't healthy. I was eating too much."

Surprise, surprise. His previous diet of wheat, bananas, and pizza was unhealthy. Even with all the Twinkies, his diet is improved in a lot of ways. This "junk food" diet could be a lot worse. He's eating veggies, drinking whole milk, cutting fruit, there's no liquid sugar in his diet, and I don't see a lot of added seed oils (not sure about the snack cakes). No wonder he saw improvements.

Maybe I'm reading too much Matt Stone, but I see metabolic derangement in this guy's near future.

Wait, wait, there's one more thing. The blood work improved because blood work always improves when one loses weight. If, speaking very generally, LDLs, those little packages for things that are not blood-soluble, are carrying fats to the cells (and carrying cholesterol too, but let's not overcomplicate), and HDLs are carrying fats from adipose tissue to be used for fuel (and probably not, alas, dipping into your endothelium and popping back out to remove cholesterol that is lodged there) -- then it makes sense that your LDL would go down and your HDL would go up when you lose weight. And in fact, no matter how you do it. Presumably when Haub reaches a new equilibrium and stays there, without gaining or losing weight, the numbers will get worse again. (Though of course we all know he is not going to stay at the new weight.)

I found the idea mentioned in a few places on Stephan's post, which Jae rightly calls our attention to. One of the commenters there also links to his blog post on the topic, which is excellent. As it turns out, this is the gentleman "Raj" who gave this same link here on this thread. Anyhow he makes the point that if anything Prof. Haub's numbers probably could have improved even more than they actually did, based on his own experience. Seems right to me.

What frustrates me is that his whole point with this is to prove that for weight gain/loss, all that matters is calories in/calories out. But anyone who's read "Good Calories, Bad Calories" hits their head on the desk whenever they see this mantra repeated. (which is often, both my head and desk are probably dented at this point.)

He doesn't list his triglycerides explicitly, but you can figure them out by multiplying "TG:HDL ratio" by HDL, right? Using that I see a before trigs of 122 and an after of 74. From all that I've read, trig levels are directly and strongly tied to carbohydrate consumption, so this says to me pretty clearly that his junk diet is lower in carbohydrates than his "heathy" before diet. (Due to overall calorie restriction and also probably due to increased fat intake.) So the point that he's making isn't even for real, he didn't just lower calories, he lowered carbohydrates a fair amount. I'd like to see him try this same diet but eating pure sugar. Or how about nothing but bread? (Okay fine, he can still have his multivitamin.)

And at first I too was surprised that his HDL went up, but then I looked at the actual numbers. He went from a disturbingly low (especially for a nutrition expert) HDL of 37, to a still very low 46. (what's the margin of error on these tests?)

I think this result is based on his own metabolism and maybe even frame of mind...he wanted to prove something, but yes wholeheartedly agree I would like to see this pulled off long term. You would see the decline over a longer period.