[EM] Correction. Big CS fault?

[MIKE OSSIPOFF]

Sorry, another error in my "CS Approval, 3 coalitions" message:

I said that the Dems would improve their score by voting for Nader
coalition candidates, but that isn't so if the Repubs have more
voters than the Nader coalition. In that case, by voting for Nader
candidates, the Dems put themselves farther from the Republicans as
much as they put themsleves closer to the Nader voters. So if the
Repubs are more numerous, the Dem candidates score better if their
voters don't vote for Nader candidates.

So in that way too, CS Approval is similar to ordinary Approval--
the Dems have no reason to vote for Nader candidates (in CS Approval
that's true only if the Repubs are more numerous than the Nader
coalition voters).

Though the Nader voters could increase the Dem candidates' scores
more by voting only for the Dem candidates, eliminating any difference
between Nader & Dem ballots, doing so would also improve the Repub
candidates' scores, and so nothing is gained by Nader voters voting
only for Dems. ...except that it would have the odd effect of positioning 
the Nader candidates where the Dem candidates are, making a score tie 
between those 2 sets of candidates. That sounds like an
error that CS makes, when it lets voters strategically put their
candidate where another candidate is, to gain a tie with that candidate.

The ability that CS confers, to put one's candidate anywhere one
wants to seems to be a serious fault of CS. CS voting is like campaigning, 
but it takes insincere campaigning to a whole new level.

Unless I'm mistaken about that fault, or unless it can be fixed, it
seems to spoil CS.

Or, on the other hand, that could be all wrong. That fault
has just occurred to me, and I sometimes make errors when I write
things as they seem to be at first glance (sorry about that). But
right that CS fault seems to be genuine.

Now, I'm almost afraid to click "send", for fear that I'm wrong about
that big CS fault. But since I can't find an error now, I'll bravely post 
this anyway.

But even if CS is never suitable as a voting system, it still seems
useful for comparing the merits of various distance measures.

Mike Ossipoff







_________________________________________________________________
Tired of spam? Get advanced junk mail protection with MSN 8. 
http://join.msn.com/?page=features/junkmail

----
For more information about this list (subscribe, unsubscribe, FAQ, etc), 
please see http://www.eskimo.com/~robla/em