No subject

Tue Mar 5 15:20:04 PST 1996

       A. In
Subj:  [EM] Hitler-Stalin-Middle Example Again Reply
Date:  Mon, Mar 4, 1996 2:19 AM EDT
From:  election-methods-list at
X-From: DEMOREP1 at
there was--
<I leave it to folks to think of examples of circular results and <the fewest
number of votes needed to elect under the <Condorcet tie breaker (especially
with the D. 33 votes each <example).

In each of the 42-41-16 first choice Hitler-Stalin-Middle A. B. C. examples
circular results can occur if 27 voters of the 41 voters vote for the
candidate having 16 votes to get
42 beats 41
41 beats 16
43 (16 + 27) beats 42.

The general case is that the candidate in the middle based on first choice
votes will try to get his/her supporters vote their second choices for the
candidate with the least number of first choice votes to produce circular

For the Condorcet tie breaker, if the middle candidate (41) gets 1 more of
his/her supporters to vote for the last candidate (16), then he/she wins the
Condorcet tie breaker (assuming the voters for the highest and lowest
candidates do not make any second choices). That is, 
42 beats 41, margin 1
41 beats 16, margin 25
44 (16+28) beats 42, margin 2

The 41 candidate is the least beaten Condorcet tie breaker winner.

The 42 candidate must thus encourage his/her supporters to vote their second
choices for the 16 candidate to help the 16 candidate beat the 41 candidate.

However to do so there is the major risk that the 16 candidate will win both
against the 41 candidate and the 42 candidate.

       B.  In the Hitler-Stalin-Middle 33-33-33 D. example if 1 voter in each
of 2 groups votes a second choice for the other candidate, then there is
(assuming no other second choice votes)
34 (33 +1)-33
34 (33 +1)-33.
Thus with 35 votes (33 first choice votes and 2 second choice votes), a
candidate could be elected (assuming no other second choice votes).

     C.    The result of the above is strategic fun and games regarding
second choices (which just might be regarded by "average" voters as being
totally corrupt as the gerrymander in legislative elections).
          With 4 or more candidates the fun and games would really get
complex for second and later choices.
          The above should also show why there probably will be much
truncation of voting if an additional choice is much different from an
earlier choice producing lots of plurality winners.

       D. The above raise the elementary questions-
Should plurality winners be allowed ? I say no.
Should all executive and judicial elections be nonpartisan to lessen partisan
extremism ? I say yes.

More information about the Election-Methods mailing list