> The problem with this theory is that while they differ, they don't have opposing bias I think the theory still holds, in that we see different results than i

> The problem with this theory is that while they differ, they don't have opposing bias I think the theory still holds, in that we see different results than if all of these bodies were picked with the same skewed representation at whatever the appropriately weighted average skew would be to match the existing "overall bias". This, of course, very much depends on the exact scope of what we mean by "the theory"; I would limit it to a statement that distance between who is represented by various government bodies can have a moderating influence and that this can be positive and important. I agree it can be overstated and in particular that it doesn't fully (or nearly) compensate for the biases in the present system. You are very much correct to bring up differing terms and term lengths as another moderating influence over time.