> Isn't this a fairly easy problem to solve? Don't you just need to make a reasonable guess, calculate your expected weight change over some time period, then follow the diet to see what actually happens to your weight, and adjust your guess accordingly?

It's easy if you do the work, weigh your food and find the number that's unique to your situation, which could take a month or so adding or taking away calories to find. Having a model to account for all the ephemeral things that plays into that unique number consistently for the population at large is the hard part

Nutritional science is hard because adherence to the diet is to be taken with a grain of salt, you have to just trust participants. Where outside of athletes I doubt may people have the energy to weigh everything, which also starts to feel unhealthy mental health wise making you seem crazy.

I get alot of push back for it myself, and have to plan around not weighing and counting calories to not get to anxious about it since I am so aware of how a small bit of extra food can have me 'lose' so much progress