[EM] Real IRV Election, Disputable Result

[Jonathan Lundell]

At 11:06 PM +1100 3/21/06, Anthony Duff wrote:
>I, like James, have thought about this.  It is 
>particularly relevant in the common Australian 
>electorate, where the voting pattern is (with A 
>left, B squeezed centre, C right, extreme and 
>other random candidates ignored):
>
>45 ABC
>5 BAC
>5 BCA
>45 CBA
>
>Even though with IRV, burying your favourite’s 
>greatest perceived opponent has little purpose, 
>people do it.  One reason is that your favourite 
>recommends it, another is that it is 
>psychologically satisfying.  People do this, 
>even though they can barely decide between A and 
>C.
>
>If IRV were quietly substituted by a condorcet 
>method, then B type candidates would be elected.

Alternatively, we might conclude that 90% of the 
electorate prefers an even chance of electing 
their end of the spectrum to trying for a 
compromise.

"So, because you are lukewarm, and neither cold 
nor hot, I am about to spit you out of my mouth." 
Revelation 3:16.
-- 
/Jonathan Lundell.