Majority Tie Breaker

Steve Eppley seppley at alumni.caltech.edu
Wed Jun 19 17:01:54 PDT 1996


Demorep1 wrote:
-snip-
>Each will almost always produce a majority of all voters winner
>(despite their well known possibilities of not picking a Condorcet
>head to head beats all winner). 
-snip-

[The above paragraph referred to Runoff, MPV, and Approval.]

What's the definition of a "majority of all voters?"  The three 
methods you mentioned often manufacture a FAKE majority and a FAKE 
mandate.

Here's an example of what I mean, using Runoff:
In the preliminary election, 10 candidates fragment the vote and none 
win more than 20%.  In the runoff between the top two, neither is 
liked by a majority but the voters have no other choices--one of the 
two is guaranteed to win.  (Remind you of the Russian election?)

When it comes to standards, I'd prefer that methods elect the
*sincere* favorite candidate of a majority, if there is one.  
I don't care about fake majorities; why do you?

---Steve     (Steve Eppley    seppley at alumni.caltech.edu)




More information about the Election-Methods mailing list