The amount of weight that someone loses doing a program is pretty easy to measure accurately. How that is apportioned (fat loss plus muscle gain) is harder to assess accurately, because it requires pretty accurate measurements of percent body fat, a notoriously difficult measurement. You can see this if you just play with the calculations a bit. A couple percentage point change in either the starting or ending percent body fat can alter the estimate for muscle gain by a non trivial amount.
What I have read is that as you get fatter, skin fold measurements of body fat get trickier. In Jeff’s case, he started with about 33% body fat and a BMI close to 32. I think that gets into the range where skin caliper measurements become less reliable. Furthermore, he starts off with a pretty big belly, which suggests he may have started with a lot of visceral fat, which I don’t think skin fold measurements can pick up.