[EM] Request comments on MMP?

[Adam Haas Tarr]

Olli wrote:

>D'Hondt favours large parties, Sainte-Laguë is neutral, modified 
>Sainte-Laguë makes the first seat more difficult, which favours 
>larger parties.
>
>If we regard both methods as algorithms to find a suitable quota, 
>d'Hondt rounds down, while Sainte-Laguë (Webster's) rounds off. Which 
>digit the fraction starts with does not depend on the size of the 
>party, so the method ought to be neutral.
>http://www.aps.org/apsnews/0401/040117.html
>
>The method that favours small parties is Adams' method, which rounds 
>up. It can be used in organizations. A local association can send one 
>delegate to the national conference for each beginning 25 members, 
>for instance.

Another way to look at it is this: both d'Hondt and Saint-Lague (and Adams) 
attempt to minimize the error in the allocation.  Sainte-Lague considers both 
over-representation and under-representation to be errors, while d'Hondt only 
considers under-representation to be an error.  Adams only considers 
over-representation to be an error.

-Adam