The theoretical concept of CICO is thermodynamics and indisputable, but the point that the data demonstrates and that these Harvard scientists are discussing is that all of the parts of the equation are so error-ridden and faulty that to combine them together in practice is not effective. Everything from the faulty results of burning food in a calorimeter, to users guesstimating the ingredients in food they cook, to the wide amount of false data inputted to calorie tracking apps, to the extremely complex and still not well understood G.I. tract, to measuring how much each individual can absorb, to how much cooking effects it, to how much the personal biomes between gut and colon play an effect, etc all add up to the reality that there is too much error to pretend the system is useful.

If using the system got you to eat differently, and eating different made you lose weight, then congratulations, but that doesn't prove that the food you ate was actually X calories or that you actually burned Y calories, and that that deficit is why you lost weight, nor does it mean that someone else who ate as much as you did would have their body change in that way.