[EM] IRV majority vs Ghost Majority of a Bottom Method
Bart Ingles
bartman at netgate.net
Wed Dec 13 22:18:40 PST 2000
Instant Runoff Voting supporter wrote:
> Dear Bart,
>
> When we are using IRV, a majority is fifty percent plus one or more of
> the total voters that voted in the race. And, it is only possible for one
> candidate to have a majority.
>
> With IRV, the voter is allowed to change his vote in the event his
> candidate becomes last in the count of the first choices. This is an option
> of the voter.
>
> A majority in a Bottom Method is merely a Ghost Majority made up from
> the addition of real votes and votes cloned from lower choices. And, more
> than one candidate may have a majority.
> The voter is not allowed to change his vote, he must stay with the loser.
>
> Regards, Donald
But Donald, IRV _is_ a Bottom Method. By attempting to mandate a
majority when none exists, you end up counting detractors as
supporters. Then your majority-by-fiat will elect someone despised by a
majority.
Example:
49% A(10) B(9) C(0)
25% B(10) C(1) A(0)
26% C(10) B(9) A(0)
On a scale of 0..10, candidate B is rated by every voter as being very
near that voter's favorite. In other words, B is a Top Candidate (and
is almost certainly the Approval winner). Plurality winner A is another
top candidate, with nearly a majority of first-choice votes.
With IRV, a Bottom Method, candidate C wins with a manufactured
majority, even though B is despised by 74% of the voters (a Bottom
Candidate if there ever was one). Surely you must agree that a majority
consisting of 26% supporters and 74% detractors is a Ghost Majority, if
that term has any meaning at all.
And you are right, with a Bottom Method such as IRV, the 49% who
supported A are not allowed to change their votes, and must stay with
the loser.
Regards,
Bart
More information about the Election-Methods
mailing list