[EM] Real IRV Ranked Ballots

[Adam Tarr]

At 04:44 PM 3/2/2004 -0500, Eric Gorr wrote:

>Also, wanted to thank Adam Tarr for providing the following two examples:

(you're welcome; examples snipped)

>For the moment, none have been able to mount a defense for why IRV selects 
>the winner it does.

The best defense of IRV against these examples that I can find on the list 
is this:

http://groups.yahoo.com/group/election-methods-list/message/12095

Not great (in my opinion, of course).  Basically, the argument boils down 
to this: people will see this situation coming, and one of the parties will 
see their first place support crumble as a result.  In essence, the 
election will be decided by which party blinks first, as oppose to which 
party has majority support (or, in this case, about a two-thirds 
majorty).  How anyone could see this as a good result is really beyond me.

>The current IRV supporter rejected the examples outright, calling the 
>voters 'oddly split' and rejecting that examples such as these would ever 
>occur.

That's a pretty funny defense of IRV in this case.  I would suggest 
"building up" to this example.  Maybe start with a British party example, 
and say there was an election that went:

35% Labour>Lib Democrat
16% Liberal Democrat>Labour
15% Liberal Democrat>Conservative
34% Conservative>Liberal Democrat

Other than giving the lib dems a bit more first place support than they 
usually do, this is a stock example, something you could probably find in a 
real British election.

And then add in a couple more extreme parties, say socialists and 
anti-immigration, or green and libertarian, whatever.  Ask him/her, at what 
point did the example become unrealistic?  At some point, either the person 
you're talking to will recognize how tortured their logic is, or you will 
recognize that the person you're talking to simply does not want to change 
their mind.

-Adam