Circular Stand-Off

[Steve Eppley]

Donald Davison wrote:

> The following is Schulze's Example:
>    40 voters prefer A > B > C.
>    35 voters prefer B > C > A.
>    25 voters prefer C > A > B.
>    Candidate A prefers candidate B to candidate C.
>    Candidate B prefers candidate C to candidate A.
>    Candidate C prefers candidate A to candidate B.
> 
> Donald writes: I would call this a Circular Stand-Off. All three 
> candidates are still contenders - any one of them could be the 
> winner if only the "correct" candidate would withdraw. 

There is no "correct" candidate to withdraw (or to elect) when there 
is a sincere circular tie.  See the discussion in this list regarding 
the merits of the Smith//Random method, which randomly elects one of 
the top cycle.

> It is merely a question of which candidate gives in first. This 
> stand-off could last long after "Schulze is Dead". (punch line 
> of a famous joke).

No, it can't last long because the JITW procedure provides candidates 
only a short period of time (a few days, maybe a week) to withdraw 
before the final result is tallied.

> One value of regular IRV is that it is willing to make a decision.

Every method makes a decision.  Some make better decisions than 
others.  JITW//IRV and JITW//Condorcet make better decisions than 
plain IRV, because IRV is mindless while JITW allows human 
intelligence to mitigate the methods' flaws.

> The pairwiseguys' zeal to elect third party candidates gets in the way
> of real reform like IRV.

Plain IRV isn't real reform.  It would leave in place the same two-
party system with expensive distorting primary elections, since plain 
IRV is so prone to spoiling.  See Gary Cox' book _Making Elections 
Count_ where he demonstrates that IRV (like "Plurality Wins" and 
"Runoff of Top Two") is subject to Duverger's Law.

I don't consider myself to be a zealot.  Pairwise methods will defeat 
the candidates whose policies I prefer most, and elect moderate 
centrists.  Donald is apparently still clueless about who would win-- 
it wouldn't be any of the existing third parties.  It would probably 
still be Democrats and Republicans, if those parties include 
moderates among their nominees.  If for some reason those parties 
only nominate non-moderates, then the moderates would have an 
incentive to run and win as independents, and possibly establish new 
centrist parties leaving behind the non-moderate rumps.

By providing a lot of competition to be at the median, we'll gain a 
lot of information from the voters about their preferences regarding 
the median and near-median policy alternatives.  The empirical 
evidence about plain IRV is that hardly any candidates will compete, 
so we can't expect to learn much about the voters' preferences if we 
use IRV.

Actually, it's CV&D who want their third party chums to be elected.  
They for some reason believe that the identities of the people 
elected to be representatives are more fundamental than the policies 
adopted by the body of representatives.  The only reason CV&D began 
publicly promoting IRV is because the respondents to John Anderson's 
December 1996 poll asked CV&D to pursue single-winner reform.  Prior 
to that, in their zeal for proportional representation, they claimed 
that ALL single-winner methods are by definition a two-party system.
To CV&D, IRV is a useful tool to teach people about the STV 
proportional representation method.  They don't deny that IRV isn't 
as good as other methods at electing the best compromise candidates.


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