Dave
You still don't get it. Bullet size has little bearing on group
size (group size is determined by subtracting the bullets diameter
from the group, and then measuring the "outside to outside
dimension"). For example, a .22 could group 2" and a .50 could
group .000". that is the size of the group, not the size of the
hole! A 9mm is, under ideal conditions, say, out of a barrel
tester, given all things being equal (ie: barrel tester, perfectly
designed and constructed ammunition) is more accurate than a .45
under the exact same conditions.
The question was, which is more accurate? The answer is: the
9mm. All of the other answers that were given, were conditional
or relative, as it applied to them. But, after scientific
testing, the absolute answer to the question is still the same.
Also, a quick side-bar, the .38, is not even in the game, as far
as absolute accuracy is concerned.