[EM] To Bill Lewis Clark re: stepping-stone

[Adam H Tarr]

Eric wrote:

>At 7:17 PM -0500 1/24/04, Bill Lewis Clark wrote:
>
>>  It's nowhere near as good as Condorcet
>>(IMHO) but it's not "change for the sake of change."
>
>Apparently, it is.

My position on IRV's advantages over plurality is this:

AS LONG AS you have two major factions that have comfortably more than two 
thirds of the first-place preference between them, IRV does a good job of 
preventing minor party candidates from "interfering" in the two party system.  
In this respect, it manages to solve the "spoiler" problem in its most commonly 
described form (i.e. the extremist party leeching votes from one of the major 
parties).

IF, however, the electorate has more than two factions that have significant 
first-place support, IRV's results can become erratic, and results can (in 
various ways) turn out even worse than plurality.  The most commonly mentioned 
problem is monotonicity violations, although the example I gave in the previous 
message has nothing to do with that.

(IRV shares other flaws of plurality, although that's not relevant here.)

-Adam