[EM] Some brief campaign argument

[MIKE OSSIPOFF]

I forgot to add: CR was one of the nominated methods.

Though any pairwise count method will likely do very well most of the
time, the margins methods don't have the absolute guarantees that I'd
like to have. Approval and the various defeat-support Condorcet versions
do well in terms of those guarantees.

Truncation, whether sincere or
strategically-intended, can take victory away from a sincere CW and
give it to the truncators' candidates, with the margins methods.

SSD & Cloneproof SSD comply with all 4 of the majority defensive
strategy criteria listed at

http://www.electionmethods.org

Approval meets FBC and WDSC.

The margins methods meet none of those criteria.

The fact that the margins methods will often do ok doesn't reassure
me, considering the fact that all it takes is some lazy or strategic
truncation to mess up the outcome with the margins methods.

Among the best methods, which I consider to be Approval, CR, and the
defeat-support Condorcet versions, I don't recongnize any significant
merit difference, and so I rank them based entirely on estimated
public acceptability.

On this list, some people have shown that at least some people can
misunderstand Approval. I'm tempted by CR's great familiarity and
popularity.

Approval, CR, & Condorcet (SSD, PC, Smith//PC) are difficult to
choose between. Approval's great simplicity could be the deciding
factor in some cases. But CR's familiarity & popularity could too.
Lots of people seem most interested when told about ranking methods.
That seems true of the people I've spoken to about voting systems.

So that leaves nothing  but doubt about which of Approval, CR, or
Condorcet(defeat-support) would be most winnable.

As I said, the issue that I nominated and that we elected was
"Which voting system would be the  best public proposal?"

But, as I was saying, we should also have a ballot section that asks
us to vote with the intent of electing the one with the most
merit, without regard to winnability considerations.

In that ballot-section, it's easier for me to rank the methods:

1. Cloneproof SSD
2. ordinary SSD
3. Approval
4. CR

(all others ranked lower)

I consider Approval better than CR because I don't want my candidate(s)
defeated because those who agree with me vote sincerely in CR and
the opponents' voters vote strategically. That isn't a problem in
Approval, where all voting is strategic. (But, as I said, I don't
consider that a significant merit difference).

Mike Ossipoff

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com