Just some numbers for comparison. A while back I cast up 5 lots of 1,000 each in .440, .454, .457, .490, and .530 moulds. The balls were weighed, and the weights plotted in .1 grain intervals. No ball was more than .3 grains under weight, or more than .2 grains over weight. Over 97% were within a .2 grain spread (i.e., plus-or-minus .1 grain).
Those weren't specially selected in any way; that was "mill run" after the mould was up to temp. From past experience I would say that is typical of balls cast in a good mould, ranging from .40 caliber to .54 caliber. I don't have enough experience to know the ranges on smaller balls, and have never really tested in larger moulds other than a .735". As I recall, the range from the .735 mould was slightly larger.
Having done similar experiments in the past, I know that I can tighten up the overall range a bit with a couple of tricks. I'm just not convinced it is worth the bother for rifles using iron sights.