[EM] Re: Chris: Approval vs IRV

[Kevin Venzke]

Chris,

--- Chris Benham <chrisbenham at bigpond.com> a écrit : 
> CB: Suppose that  for some single-member seat; previous election 
> results, pubblished polls and all other indicators say that the
> result in the next election will be:
> 
> 60%:Labor
> 40%:Conservative
> 
> The Labor party machine trys to impose some candidate that local Labor 
> party supporters don't like, or like decidedly less than
> than some other pro-Labor candidate.  Under IRV, this third candidate 
> can run as a Labor-clone, possibly badging himself  as
> "Independent  Labor", which at a minimum means that he directs his 
> preferences to Labor and if elected, will help make up the
> numbers in parliament  on which  Labor  tries to base a government. For 
> this reason,  the Labor party will direct their supporters
> to give that  candidate their second preferences.
> The threat of  the (potential) rebel-clone (to the official candidate) 
> is "I will appeal to regular Labor supporters to supplant you as
> their representative, without them being inhibited by any anxiety about 
> splitting the anti-Conservative vote".

The reason I doubt this scenario is that if Independent Labor is really a
substantially different candidate, would the Labor party really, without a
doubt, instruct their voters to put the independent second?  Because by not
doing so, by asking their voters to put the independent third, it seems to me 
that this increases the odds that voters will elect the official nominee.  
(That is, increases the odds in comparison to the independent's.)

If the party says to dump the independent, then voters have reason to be
inhibited by anxiety about splitting the anti-Conservative vote.


I was thinking about the defection issue generally...  I know that in this
scenario:

49 A
24 B
27 C>B

You do not like to elect B, probably because you suppose that the B voters are
stealing the election from C.  If the B voters tack on the second pref for C, then
most methods suddenly elect C, which is a gross LNHarm failure.  IRV dodges this
by not electing B to begin with in the above scenario.  But then the C voters wish
they had compromised, showing a WDSC failure.

I can't see any way to fix this so that no one is ever punished.  I guess I would 
rather punish B voters for voting B>C (by electing C, their compromise) than punish 
C voters for voting B in second (by electing A, their least favorite).  My reasoning
is that punishing C>B voters is more likely to discourage C from entering the race.

Kevin Venzke
stepjak at yahoo.fr


	

	
		
Créez gratuitement votre Yahoo! Mail avec 100 Mo de stockage !
Créez votre Yahoo! Mail sur http://fr.benefits.yahoo.com/

Dialoguez en direct avec vos amis grâce à Yahoo! Messenger !Téléchargez Yahoo! Messenger sur http://fr.messenger.yahoo.com