> Is this really true and known empirically?
If you are very careful about how you define “in” and “out”, to the point where it is nearly circular, and are concerned only about weight, not health, yes. It is still problematic in a number of ways, e.g.: if you just attack thr calories in side and have a relatively sedentary lifestyle, you will disproportionately lose lean body mass rather than fat, this not only is less than ideal as an end outcome, it also makes further weight loss more difficult because it reduces resting metabolic rate (and, therefore, minimum calories out) and makes it harder to do additional exercise to increase calories out.
It is likewis known empirically that, as actionable advice, its not particularly useful on its own besides those problems, for the same reason “don’t write bugs” is as advice on how to create correct software. Most serious weightoss advice incorporates CICO, but also adds layers on top dealing with the pragmatics of how to acheive that in a maintainable way (and usually also has at least some thought in how to do it in a way dealing with problems of naive CICO like those discussed above.)
Also, composition of gut microbiota seems to have an effect on metabolism and obesity, at least in mice; that should not be surprising, given that gut microbiota "predigests" some nutrients for us. [1]
[0] https://thecounter.org/scientists-calorie-counts-usda-kind/
It's absolutely true that if you count calories, you can manage the input such that you lose weight. But it is not a solution for the vast majority of humanity.
Why? Evolutionarily, we've had very little time where constant surplus was the norm, and mostly are adapted to situations where surpluses are rare and temporary abd shortages common.
Fighting that is hard.
Sure, our bodies might become good at that if we survive for another million years with first world standards of living as the norm. But they aren’t now, for good—evolutionary—reasons.
Of course, 2000cal worth of HFCS is gonna have a different biological impact than 2000 cal worth of kale. CICO is a gross over-simplification.
Is this really true and known empirically? Could the body be better at accessing calories in some foods and not others?
Thought experiment: suppose I wrapped a small chocolate bar into airtight plastic, then I swallowed it. I have just swallowed 200 calories, but my body will access/process/burn 0% of it since it can't break down the plastic.
At least in this one (unrealistic) example, CICO is false.