There are two effects that are pertinent here. First is the raw velocity. The impact of this is easy to figure out with quickLOAD and a ballistics calculator. See what .02 grains of powder does to velocity, and then see what that difference does to impact at 1,000 yards. This tells you the impact of a 0.02 grain difference *ASSUMING ALL ELSE EUQAL*. It's that last part that makes people scoff at or doubt calculations. All else is never equal. Changes in temperature, primer variation, the way the powder settles, etc - they all change things, even if just a little. But on average, this is a good method.
The second effect is positive compensation due to barrel vibration. Small changes in powder can result in meaningful differences in the bullet's initial launch angle and lateral velocity, and if you get it just right, this can cancel out other errors and result in smaller groups. Unfortunately, there is no way to calculate this with the accuracy required in order to get around testing it.
If you ask long range benchrest shooters (which I am not), they're going to tell you that it's the second effect that dominates *when trying to shoot tiny groups*. This is what they mean by "tuning" the gun. The velocity difference of .02 grains is small, but they're not looking at raw velocity and it's impact on drop or wind, they're looking at the impact on how the bullet leaves the bore, and trying to get that to encourage small groups. Since it's such a complex system, you have to try it to find out what matters and what doesn't.
