Duplicate bridge was created to "eliminate" the luck factor. That is partnerships are compared only against other partnerships playing the same cards.
Rubber bridge is a different animal. Here, "luck of the draw" is clearly a major factor in the result, at least until you get to a "law of large numbers" point where the luck factors cancel out.
Suppose there are two partnerships, A and B, playing ONE rubber. From their play at duplicate, we "know" that partnership A plays one trick better than the other. Clearly partnership A would have better than a 50-50 chance at rubber as well. But how much better?
Is there any computer software, statistical formula, or mathematical algorithm that could estimate the relative chances of partnerships A and B winning the rubber?