Cloneproof SSD

MIKE OSSIPOFF nkklrp at hotmail.com
Thu Jan 25 18:14:16 PST 2001



>The amount of *cloneness* is a question of degree even with 2 choices.
>
>Choice A exists (versus no other choice).
>
>Choice B comes along.
>
>Possibilities
>
>all to zero A [> < ] B zero to all
>
>Examples--
>
>100 A > B 0
>0 A < 100 B
>
>99 A > 1 B
>1 A < 99 B
>
>51 A > 49 B
>49 A < 51 B
>
>Some voters, of course, may rank A and B as being equal.
>
>Assuming that only the *all* (100 percent) values indicate clones is a bit 
>of
>a stretch (especially in public elections).

In any of your examples, has anyone ranked a candidate between A
& B? Or ranked A over someone without ranking B over him?

A 2-candidate election is probably what would be called a "trivial case"
of clones.

>
>That it, I very highly doubt that there would be any *100 percent* clones 
>in
>public elections so that worrying about *cloneproof* methods is something 
>of
>a waste of time.

Suppose in a 100,000,000 voter Presidential election, A & B are clones
except that a few voters votes A over C without voting B over C.

Obviously, the bigger the election, the more voters could do that
without changing the result.

Though true clones are unlikely in a many-voter election, it's less
unlikely that 2 candidates will be so similar that only a few sporadic
voters will rank them other than as clones, and that those will be
a small enough minority of the electorate as to not change the result
from what it would be if those 2 candidates were true clones.

The point is: Running a candidate who is sufficiently similar to another
candidate won't have a spoiler effect on that other candidate.
That similarity needn't be so great that the 2 candidates are true
clones.

If a method lets that 2nd candidate spoil the 1st one, even when they're
true clones, that tells you that it can also do so under the more
realistic conditions when they're merely near-clones.

Steve Eppley has written a criterion about near-clones, but it's
apparently quite complicated. It might be very useful, if he can
simplify it some.

Mike Ossipoff


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



More information about the Election-Methods mailing list