Here's a revised calculus of the whole thing.

Average daily Ein for experiment: 1457 kcals
Average daily Eout for experiment: 2840
Claimed Eout before experiment: 2600
Projected caloric deficit: 800
Corrected projected caloric deficit: 1143
Actual caloric deficit: 1383
Difference between projected and actual: 240
Total difference for the duration: 16800 kcals.

Just to give you an idea of how much that is: One pound of fat is 4086 kcals. So that would be a bit more than 4 lbs of fat.

Considering the rigidity of the First Law of Thermodynamics, there shouldn't even be a difference, let alone such a big one. Furthermore, the argument about the change in Eout over the experiment from greater to smaller still holds, so it's hard to see where those calories went.

Further furthermore, considering that a hypothesis is supposed to predict what will happen in the future, and that the calorie hypothesis is based on an exact science, i.e. the First Law of Thermodynamics and that it can't be broken (or actually, it's never been demonstrated to be false yet), and that the calorie hypothesis did not predict the difference between projected and actual, we can only conclude that the calorie hypothesis must be wrong.

We can only conclude the calorie hypothesis is wrong because he is a normal sized person.

If a fat person gave out numbers that did not precisely match the calorie-math, it would instead simply be assumed they were either lying, in denial, or well intentioned but stupid (mistaken).

PJ

It would be assumed that we cannot expect precision about weight management just by counting the calories we eat alone. An easy to read flowchart diagram of what can happen to what we eat would be useful I believe. I'll check if I can find something like that.

In truly precise experiments, the energy balance hypothesis has always prevailed so far. Even Gary Taubes does not dispute this fact. GCBC is about causality and not about refuting the energy balance hypothesis.

Well, sure. There's just more to energy balance than 'merely' calories of the food before it enters your mouth. I get that. But Martin was referring to calories (and most people do about such debates) so I responded to that part. - PJ

But the calorie hypothesis is precisely that we can expect precision about weight management just by counting calories we eat alone.

"What can happen"? The calorie hypothesis is so inaccurate that it can't tell us exactly what will happen?!? Why didn't the hypothesis predict that there would be a 4 lbs of fat difference between projected weight loss and actual weight loss? How does it explain it? What's the margin of error? As far as I know, the First Law of Thermodynamics' margin of error is zero. But maybe the calorie hypothesis allows for some black hole to swallow calories that can't be explained otherwise?

Never mind that "in truly precise experiments, the calorie hypothesis has always prevailed so far". What about this experiment?

Now take the facts and put it through the hypothesis and see if it prevails.

But you don't get it. You think that just counting actual Ein and actual Eout during the experiment is enough. It's not. What the hypothesis is supposed to do is be able to predict what will happen if I start with X Ein-Eout, cut Y kcals, and expect Z Eout and thus an exact weight loss. How come we started with a different Eout than we had during? Not only was it different, but it was different in the opposite direction. Remember, the calorie hypothesis expects Eout to drop. In this experiment, Eout increased.

Of course not. You may think that, but that does not make it so.

We can only count calorie intake with some precision, this is the only part of the equation that we can measure by ourselves without living in a specialized lab.

I get it just fine and there is no such expectations. I'm open-minded, I proved it by changing my mind a couple of times already.

"Change in E" equals "E in" minus "E out". This has stood every tests so far. Even when I believed in Taubes' carbs hypothesis, I still acknowledged this undisputed fact.

If you measure "E in" and "Change in E" does not match what you think it should, then "E out" was not what you thought it was. When we measure all three in a lab setting the equation holds every single time.

Regarding weight per se, many factors comes into play. Water balance, glycogen, etc... Body weight is not only about body fat content or energy.

So we can only conclude that the twinkie diet was not a precise experiment designed to test the energy balance equation, but to show evidence that it still holds even while eating junk for this guy. Although, I'm pretty sure it would still be true for the other humans. That's all there is to it.

Sorry, you still haven't explained where these missing calories went. 4 lbs of missing fat. Where is it? The calorie hypothesis can't explain it. But you can by accusing the guy of lying or of being incompetent? Like I said a few times already, we've already established the guy's reputation and competence. Your horse is dead. Try again.

I don't see where it's claimed anywhere that Haub ate 2600 calories daily before the diet, only that a man his pre-dieting size usually consumes about 2600 calories daily. We can only work from the information given. If this were a serious dietary intervention study, he would have gone into this with a one or two week period of eating his original diet to appetite, to try to record his habitual intake. Never mind changes in energy out, we don't even really know what the change in energy in was.

Yes I did answer this in my last post. I'm sorry that you cannot or will not see that.

I'm sorry you can't see that your answer is unsatisfactory.

That means we can conclude anything we want from this experiment which is to say we can't conclude anything at all.

Brandnew Jimmy Moore Podcast concerning Dr. Haub and his Twinkie-diet