Condorcet Truncation Example

[Markus Schulze]

Dear Hugh Tobin,

I cannot agree with your statement, that (if a Condorcet Criterion
method is used) only omnissent voters can vote strategically.

In the USA, there are elections for parliament every two years.
The parties do know very well, where their supporters and their
adversaries are and how they will vote. If they didn't, how could
gerrymandering occur?

The only reason, why the parties in the USA haven't asked their
voters to vote strategically is the fact, that the USA has only two
parties and in a two party system it is not possible to vote
strategically.

Outside the USA, it is very common that the parties ask their 
supporters to vote strategically.

********************************************************************

The example, I presented in my last email, was only an example to
show, that it is possible that voters are punished for not
truncating. Rob Lanphier asked for such an example.
I agree with you that voting in reverse order is a very much 
more usefull strategy.

If a Condorcet Criterion method would have been 
established in the USA, then the parties would present the
example of my last email to explain to the voters why they should
truncate.

It has been said several times in this election method list, 
that impartial people will tell the voters that it doesn't make
any sense to vote strategically if a Condorcet Criterion
method is used. I don't agree that it doesn't make any sense.
And I don't agree, that they will tell that. I believe, that
they will demonstrate the most curious and most absurd situations 
and pretend that these situations are typical for a Condorcet 
Criterion method, because nobody is interested in being impartial,
but everybody is interested in being interesting.

********************************************************************

I want to show a way, how it is possible to vote
strategically if a Condorcet Criterion
method is used:

Suppose you are right-wing and you are living in a left-wing
constituency (e.g. 60% Democrats, 40% Republicans). Suppose there
are R+L candidates, R right-wing candidates and L left-wing
candidates. How do you have to vote to make a right-wing
candidate win this constituency?

  You give your first R preferences to the right-wing
  candidates as follows: The more moderate the candidate, the
  better the preference.
  You give your last L preferences as follows: The more radical
  the candidate, the better the preference.

What will happen?

  The moderate right-wing candidate will lose against the moderate
  left-wing candidate, because you are living in a left-wing
  constituency.

  The moderate left-wing candidate will lose against the radical
  left-wing candidate, because all the right-wing voters and a few
  left-wing voters will give the radical left-wing candidate a
  better preference than the moderate left-wing candidate.

  The radical left-wing candidate will lose against the moderate
  right-wing candidate, because all the right-wing voters and a few
  left-wing voters will give the moderate right-wing candidate a
  better preference than the radical left-wing candidate.

  If this is done well enough, then the moderate right-wing
  candidate will win this constituency.

Example:
  
  There are one right-wing candidate A, one moderate left-wing
  candidate B and one radical left-wing candidate C.

  The opinions of the voters are as follows:

  40 voters prefer ABC.
  25 voters prefer BAC.
  10 voters prefer BCA.
  25 voters prefer CBA.
  
  Then B is the Condorcet winner.

  Now suppose the 40 right-wing voters use the above mentioned
  strategy. Then:

  40 voters vote ACB.
  25 voters vote BAC.
  10 voters vote BCA.
  25 voters vote CBA.  

  Now A, B, and C are in a tie. And if you use
  Smith//Condorcet [EM], then A wins.

I don't agree with Hugh Tobin's statement, that the voters have 
to know very exactly how the other voters vote to be able to 
vote strategically. 

To use the above mentioned strategy you only have to know whether
you are living in a right-wing or a left-wing constituency. I believe,
most voters know whether they are living in a right-wing or a
left-wing constituency. And I believe that most voters don't care
who is elected if not a candidate of the own party is elected.

*********************************************************************

What will happen, if a Condorcet Criterion method is established
(e.g. in the USA)?

I believe, in every constituency the minor party will ask their 
supporters to use the above mentioned strategy.

Then the major party will ask their supporters to use the above
mentioned strategy as a counter-strategy against the minor party.

At the end, that party will win, that succeed in making their
supporters give their first preference only to those candidates of the
own party, who can win against the moderate candidates of the other
party. But this would mean, that at the end, the Condorcet Criterion
method results in Hare's STV method.

**********************************************************************

Summary:

-Hugh Tobin said that (if a Condorcet Criterion method is used)
 the voters won't vote strategically, because they would have to
 know very exactly how the other people vote to be able to
 vote strategically.
 I don't agree with Hugh Tobin. I believe that the parties know
 very well how the voters of a constituency will vote. And
 I believe, that if a party will ask their supporters to vote
 strategically, most of their supporters will.
 The only reason, why the parties in the USA have never asked
 their supporters to vote strategically is the fact that there are 
 only two parties in the USA and that it is not possible to 
 vote strategically if there are only two parties.
 In Germany it is the normal case, that the parties ask their
 voters to vote strategically and that the voters do that.

-Hugh Tobin said that (if a Condorcet Criterion method is used)
 the voters won't vote strategically, because they would have to
 fear that if they vote strategically the elected candidate is
 worse than elected candidate if they don't vote strategically.
 I don't agree with Hugh Tobin. The parties will tell their
 supporters, that if the candidate of the own party is not elected
 than it doesn't matter who is elected. And the supporters will
 believe that.

-If every party asks their supporters to vote strategically, then
 the Condorcet Criterion method will result in Hare's STV method.

Markus Schulze (schulze at speedy.physik.tu-berlin.de)