Having done a bit of poking around on this subject, as far as I can tell the model is more or less as follows.
The human body is modelled as a collection of four elements; fat, muscle, water, and bone. The percentages of these different elements can change with diet, with exercise, with different types of exercise. Bone is pretty much constant (though apparently lack of calcium can cause trouble there); water fluctuates a lot. Fat and muscle are more controllable; a given diet and exercise regimen has a target fat percentage and muscle percentage. Starting on the diet/exercise causes the body to approach the target fat/muscle percentage in some manner (it may be asymptotic). For this purpose, lack of exercise also counts as an exercise regimen, and it is one that has a high fat percentage and a low muscle percentage (so if you have been exercising and stop, you gain a fair amount of fat). There is some complicated interaction between the diet and the exercise regimen here. There may be a genetic component also affecting the model.
Each of these four elements—fat, muscle, water, bone—has a certain density, a certain conductivity. There are certain percentages of these elements (I do not know what they are) that would lead to an optimal health (measured as the greatest life expectancy). Given a person’s height, and perhaps a few other measurements, one can estimate the total mass of bone (our skeletons are pretty much standard). From this, and given the optimal percentages, one can estimate the optimal mass of fat, of muscle, for the greatest life expectancy. (Water still fluctuates a lot, as I understand it).
Measurements of these percentages include weight, girth, electrical conductivity, and use of calipers. The first three of these figures measure quantities that are affected by all four percentages; a change in one factor can be masked by a change in the others.
All in all, it’s a far more complex problem than it looks like at first glance. Some heuristics have leaked out into common knowledge; things like “don’t eat too much fatty foods” and “exercise at least a bit”. I am not sure how accurate these heuristics are—presumably there is some reasoning backing them, possibly based on the model vaguely described above. I also suspect that the idea of the ideal weight (based on BMI) is based on the expectation of a certain common maximum muscle percentage.
Having done a bit of poking around on this subject, as far as I can tell the model is more or less as follows.
The human body is modelled as a collection of four elements; fat, muscle, water, and bone. The percentages of these different elements can change with diet, with exercise, with different types of exercise. Bone is pretty much constant (though apparently lack of calcium can cause trouble there); water fluctuates a lot. Fat and muscle are more controllable; a given diet and exercise regimen has a target fat percentage and muscle percentage. Starting on the diet/exercise causes the body to approach the target fat/muscle percentage in some manner (it may be asymptotic). For this purpose, lack of exercise also counts as an exercise regimen, and it is one that has a high fat percentage and a low muscle percentage (so if you have been exercising and stop, you gain a fair amount of fat). There is some complicated interaction between the diet and the exercise regimen here. There may be a genetic component also affecting the model.
Each of these four elements—fat, muscle, water, bone—has a certain density, a certain conductivity. There are certain percentages of these elements (I do not know what they are) that would lead to an optimal health (measured as the greatest life expectancy). Given a person’s height, and perhaps a few other measurements, one can estimate the total mass of bone (our skeletons are pretty much standard). From this, and given the optimal percentages, one can estimate the optimal mass of fat, of muscle, for the greatest life expectancy. (Water still fluctuates a lot, as I understand it).
Measurements of these percentages include weight, girth, electrical conductivity, and use of calipers. The first three of these figures measure quantities that are affected by all four percentages; a change in one factor can be masked by a change in the others.
All in all, it’s a far more complex problem than it looks like at first glance. Some heuristics have leaked out into common knowledge; things like “don’t eat too much fatty foods” and “exercise at least a bit”. I am not sure how accurate these heuristics are—presumably there is some reasoning backing them, possibly based on the model vaguely described above. I also suspect that the idea of the ideal weight (based on BMI) is based on the expectation of a certain common maximum muscle percentage.