[EM] New clones definition?
Joe Weinstein
jweins123 at hotmail.com
Wed Apr 3 14:40:16 PST 2002
Rob's proposal sounds familiar, as well as maybe even plausible. Let's try
it out on a very simple election:
100: A>B
(A and B are the only candidates, everyone loves A and detests B.)
Delete B from ballots, you get: 100: A
Delete A from ballots, rename B as A, you get: 100: A
The proposal makes A and B 'clones'!
Well, if you don't like an example which starts with just two candidates and
therefore will reduce to no-contest when one is deleted, let's try again
with a third candidate, e.g. start with: 100: A>B>C.
Either way - deleting B, or deleting A and then renaming B as A - you end up
with: A>C. So again, A and B are 'clones'.
More generally, the proposal will make 'clones' out of any two candidates A
and B which are adjacent in preference for all voters - no matter how voters
distribute among the alternatives A>B, B>A, A=B.
As the first example illustrates clearly, such 'clonehood' lacks one
property that some people (including me) find essential. Namely, suppose
voters learn that the winner is some (unspecified) member of a given set of
'clones'. Then there should be overall indifference as to WHICH of the
'clones' has won!
Joe Weinstein
Long Beach CA USA
----Original Message Follows----
From: Rob LeGrand <honky1998 at yahoo.com>
Reply-To: honky98 at aggies.org
To: election-methods-list at eskimo.com
Subject: [EM] New clones definition?
Date: Tue, 2 Apr 2002 21:55:37 -0800 (PST)
I just thought of a possible definition of clones:
Candidates A and B are clones in a ranked-ballot election if and only if
deleting A from the ballots and renaming B as A is equivalent to deleting B
from the ballots.
Example:
49:Reagan>Anderson>Carter
33:Carter>Anderson>Reagan
18:Anderson>Carter>Reagan
Deleting Carter gives:
49:Reagan>Anderson
33:Anderson>Reagan
18:Anderson>Reagan
Deleting Anderson gives:
49:Reagan>Carter
33:Carter>Reagan
18:Carter>Reagan
Now the sets of ballots are identical except for the name of
Anderson/Carter,
so Anderson and Carter are clones in this election. On the other hand,
deleting Reagan gives:
49:Anderson>Carter
33:Carter>Anderson
18:Anderson>Carter
Which isn't identical to either of the other sets of ballots, so Reagan
isn't a
clone of either of them. Can this definition ever fail? Has it been
suggested
before?
--
Rob LeGrand
honky98 at aggies.org
http://www.onr.com/user/honky98/rbvote/calc.html
__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://taxes.yahoo.com/
_________________________________________________________________
Send and receive Hotmail on your mobile device: http://mobile.msn.com
More information about the Election-Methods
mailing list