Your restated original post has me curious to mark my brass for my next slow fire event.
If we think through the problem as you have restated it becomes an interesting experiment. If we assume that the pressure ring is unsupported by the chamber we can look at it this way:
The firing pin falls, the primer detonates, the powder burns. Now the gas is pushing the brass outward, the case is stretching backward, at peak pressure the "pressure ring" is as far back as it is going to get with only the brass to contain the pressure.
If the chamber is out of round, it will expand to fill that out of roundness, from the thinnest part of the brass towards thicker parts. This out of roundness can come from further forward in the chamber.
If the brass is out of round, or thinner/weaker in one spot, it will expand more to exploit that weakness, expanding until the brass is in contact with steel or until the pressure subsides as the bullet moves down the bore.
If the chamber is perfectly round, and the brass perfectly uniform, the extractor pressure will add stress to that one side, and it would expand more (although I doubt this is a measurable phenomena as the spring pressure on the extractor is nowhere as stable as chamber or bolt head steel).
If there chamber is perfect, and the brass perfect, and no extractor pressure, then gravity would be pulling the top and compressing the bottom, so we would expect more expansion on the top.
Like water flowing to the lowest point, pressure always tries to take the path of least resistance. However with the amount of variables involved, I doubt that any two rifles would produce the same brass flow.
If I had to rank factors in what I perceive to be importance it would be: Chamber uniformity, brass uniformity, extractor pressure, gravity, in that order.
Now I have to mark my loads to measure reference the top of the camber and bring along a micrometer to see if I can find a difference in the pressure ring roundness.
Machine guns are awesome until you have to carry one.