I found with my Ruger that keeping the balls out in the chamber as far as possible gave better accuracy for me. I also grease the exterior around the balls. Seating the balls deeper would result in more grease, may mess up accuracy.
I'd rather use a filler than seat the balls down in the chamber.
No matter what you do, the simple geometry of the ball and chamber, means that the ball will have to travel at least .22 inches to reach the end of the chamber in a 44. So does it make a major difference whether it travels .5 inches or .22 inches, I don't know. I do know that if the ball starts moving too fast before it hits the forcing cone and then the rifling, there will be more inertia against beginning to spin in the rifling. Whether it would be enough to cause the ball to strip the rifling and lose accuracy, I also don't know. It would certainly seem more liklely.
Now for the most part, such guns aren't that accurate anyway, that it may not matter as much as several other factors that go into accuracy of such a gun.
I have a mould that throws a small cylindrical .456 bullet that is only .45 inches long. It is flat on both ends with two grease grooves. My Ruger likes these bullets, but kicks enough to cause sore wrists.