[EM] Why my methods are unbiased by accepted definition

Michael Ossipoff mikeo2106 at msn.com
Mon Jan 22 05:10:29 PST 2007

It still isn't possible to send long postings, because of modem shutting 

Here's what bias means to most people:

A method is large-biased if the largest states have more s/q than the 
smallest states.

Imprecise, but it can be improved if we specify that we're talking about a 
hypothetical country with arbitrarily many states. "Large states" means an 
arbitrarily large number of states at the top end. Small states likewise 

With so many states in "large states", local-positioning-caused s/q 
disparity, caused by different states being in dilfferent parts of their 
cycles, averages out and cancels out. We're left with the disparity in the 
cycles' average s/q.

My methods transparently make cycles' average s/q equal, as nearly as 

Mike Ossipoff

More information about the Election-Methods mailing list