Energy intake - Energy expenditure - Change in energy stores = 0
There. Just f(x,y,z)=0. That's what the first law really says. There is no causality implied, it just a first integral of the "motion" of metabolism that the body, naturally, has to follow.
It does NOT say what variables are free to independently modify. It's just a relation the three of them have to obey.
You put some mass that contains chemical energy into your body. Then that mass comes out of your body, minus SOME of the chemical energy. How much of it is extracted? What determines this? I have no idea. That said, I think it is funny to treat it as if the only way to get rid of energy is expenditure. How about:
Energy intake - Energy Ependiture - Energy still in food when it comes out the other end = Change in energy stores
My point is that a lot of people misinterpret conservation of energy assigning it a direction of causality it doesn't have. It's a constraint between variables (not an assignation as in programming languages) which are under complex feedback control systems in the body, and that's why the kind of calories you eat (fats vs carbohydrates) DOES matter, you don't eat abstract pure energy but as you say mass with chemical energy that has to be metabolized.
That's why starvation diets (just counting theoretical calories, which are numbers obtained in laboratories, dismissing the type of macronutrient) often fails in the long term and leave serious damage. The book talks long about this.
I've just noticed that people generally do equate "energy intake" with calories that go into your mouth, and find that interesting.
Change in energy stores = Energy intake – Energy expenditure
According to my biochemistry professor in college, this had one - and only one - interpretation: The change in energy stores is controlled by energy intake and energy expenditure. The right side of the equation controls the left side of the equation. Also, there was an implied "just eat less and exercise more" (that energy intake and expenditure are independent variables).
But Taubes suggests that the causation could be reversed, and that energy intake and expenditure are dependent variables. In other words, a metabolic change could cause a change in energy stores, and energy intake and expenditure would change to match.
That (amazingly!) translates to: if you are losing weight you are likely to be less hungry and likely to have more energy. Or perhaps, if your metabolic state is such that you are losing weight, your hunger will adjust to match your activity level. Or even, if you're burning fat, you will only eat to make up for what you aren't getting from stored fuel.
It makes sense to think that if your energy stores are releasing fuel into the blood stream you would be less hungry (I'm assuming that hunger signals are a function of available fuel to some extent).
Naturally, we eat for nutrition as well as fuel, so I suppose it can't all be broken into thermodynamics, but it's an interesting way of thinking about it!