Webster explicitly directly puts each state’s s/q as close to 1 as possible, and therefore as close to each other as possible. No pair of states could be any closer in s/q.

But, as I recently found out, Webster, when the frequency distribution is uniform, also puts each cycle’s expected overall s/q as closed to 1 as possible. That (given the uniform distribution) would give to everyone in the U.S. the same representation expectation, if it isn’t known what part of their cycle the states are in. Or, to put it differently, if we’re looking at it with regard to cycles, and not position within cycles.

But if we look at it with regard to position within cycles, there can, with any method, be no such thing as unbias or equal expectation.

So Webster, WW, CW & AR, achieve unbias and equal representation expectation to the extent that it’s achievable, and in the way that it’s achievable.

Mike Ossipoff


----
election-methods mailing list - see http://electorama.com/em for list info

Reply via email to