> But that is issue with CICO, it's like magic box and we only have to count IN, as much at what makes the calculation correct later on, also with OUT. That is not very scientific.

The whole physics is a magic box. We find qualities of materials after we measure them, because that's how the model is constructed.

Let's say that I have a model called mass in volume out (MIVO). It states that mass of a liquid predicts its volume. We measure how much volume a certain amount of water takes, vary the amount, see if the model works, and it does. But a MIVO-denier is not happy, because our measurements with water doesn't reproduce with mercury. Well, duh, because they have different densities. MIVO-denier is still not satisfied, because now they have to go through the tedious task of finding the density of the material they have to work with before being able to apply the model. Furthermore, they have to keep the temperature constant, not mix liquids, etc. Yet, MIVO works regardless of how difficult it is to quantify, measure, or use.

PS: Thanks for the reference. I had missed it.

"Calories" are simply a unit of energy. The energy released by the utilization of ATP can be measured in any unit of measure you like for energy. It won't change the fact that the unit can be calories.

This is like arguing about Celsius vs some other unit.

Likewise, the CICO model has been replicated over and over again in high quality studies that don't rely on self-reported intake. And this model and information are used by thousands of people every year to manipulate their bodyweight and composition at will in sports like bodybuilding or ones with weight classes.

Indeed, treating the body as a black box makes this even easier as tracking caloric intake and your bodyweight is sufficient to do this and apps like Macrofactor do all the math for you.