Of course, if there are 2 sets of candidates such that the merit difference
within the two sets are negligible compared to the merit difference between
the sets, then it's clear that defeating the worse set is more important
than choosing among the better set.
So I mis-spoke when I said that, in ICT, if there's a C/D problem, you
should automatically vote E in 2nd place. All of {A,B,C,D,E} belong in 1st
place.
It's just that, in the special situations where in Approval, you'd not
approve E, _that's_ when you'd move E down to 2nd place in ICT.
The various C/D solutions in Approval can, in some circumstances, call for
not approving E. Those would also call for demoting E to 2nd place in ICT.
It's the same strategy response, in both methods, though it's different in
the 2 methods. Obviously, that response is considerably milder and less
costly in lCT. That's ICT's advantage over Approval.
Of course, in a u/a election, certain conditions would have to be met before
you'd let the C/D solutions make you (in Approval) not approve E, or (in
ICT) move E down to 2nd place.
For instance, maybe E is only barely or marginally acceptable. &/or maybe,
perhaps due to things said by E supporters and organization, there's doubt
about whether E really would implement hir good platform policies or
campaign promises. In a u/a election, to justify the non-approval or the 2nd
place demotion of E, A,B,C, or D should have a (good?) chance of winning.
If E is clearly the big favorite among {A,B,C,D,E}, and the only one
winnable, then you don't want to do the non-approval or demotion. I'm
assuming that E is at least marginally acceptable in a u/a election.
Even if E supporters attack or continually bad-talk A,B,C and D, that isn't
a practical reason to not approve E or to demote E to 2nd, if E's platform
would bring some genuine improvement, and is on the acceptable side.
And if E's lesser improvements would result at least in a more open system,
with more open and honest media, better media and ballot access, etc., that
would be a start toward better improvement, no matter how much E's
supporters are criticizing A, B, C, and D.
Later, with E elected, if things aren't good, then of course E is unlikely
to get your full support next time, and the various C/D solutions would come
into play if E remains marginally acceptable. Or maybe E has by then become
unacceptable and wouldn't be considered for approval or top ranking.
Mike Ossipoff
----
Election-Methods mailing list - see http://electorama.com/em for list info
