[EM] smith/schwartz/landau

[Kristofer Munsterhjelm]

On 03/28/2018 08:08 PM, Curt wrote:

> Incidentally, this does point to what I believe *is* a major downside of 
> a Condorcet method - it’s by definition entirely unsuitable for figuring 
> proportional representation, because that again would imbuing vote 
> margins with utility concepts.

The Condorcet property (or even the majority property) is incompatible 
with proportional representation. The usual example is something like:

51: X1>X2>X3>X4>X5>X6>X7
49: Y1>Y2>Y3>Y4>Y5>Y6>Y7

Suppose we want to elect four winners. The proportional outcome is to 
have two Xes and two Ys, probably X1 X2 Y1 Y2. But applying the majority 
criterion forces the election of X1, and after X1 is out of the picture, 
it forces the election of X2, and so on until four X-es have been elected.

Majority and Condorcet simply are the wrong tools for the job of PR. 
That said, there are methods that pass Condorcet when there's only one 
candidate to elect, and pass the Droop proportionality criterion when 
there's more than one to elect. One example is CPO-STV, and another is 
Schulze STV.