[EM] IRV - disincentives

[stephane.rouillon at sympatico.ca]

The answer is no.
Clearly if a and b are eliminated,
the ballot of your voter had no impact on it
(a>b) or (a>b>c>d).
So those last preferences kick in only once a and b are eliminated co you are better with the full preferences to
avoid at least d over c.

However their is an incentive for unsincere ranking
b>a>c>d if you think most voters who vote for b in 1st place
do not prefer a as 2nd but most voters for a in 1st place prefer b as 2nd. This is because IRV is non-monotonic.
Blake and other specialists on this list had a name for that: I think it was "digging" or something alike...

> De: Eric Gorr <eric at ericgorr.net>
> Date: 2004/03/30 mar. PM 06:39:39 GMT-05:00
> À: election-methods at electorama.com
> Objet: [EM] IRV - disincentives
> 
> How would others answer this question:
> 
>    Is there a disincentive in IRV to rank candidates
>    you do not want to see win?
> 
> I am thinking the answer is NO, but there is something nagging at me 
> which says the answer could be YES.
> 
> Consider a case with four candidates and a voter who casts a ballot:
> 
>     a > b > c > d
> 
> but does not want to see either C or D win the election. It does not 
> seem to me that the voter would be any better off if they had cast a 
> ballot:
> 
>     a > b
> 
> preventing the possibility of their ballot being used to increase the 
> vote total for C.
> 
> Which is then where the disincentive might come into play...the lower 
> the vote total for a Candidate, the less of a mandate that candidate 
> might believe they have...but is this enough of a reason to answer 
> 'YES' to the question above?
> 
> ----
> Election-methods mailing list - see http://electorama.com/em for list info
>