[EM] Re: IRV Letter

[Adam Tarr]

  Jeffrey O'Neill wrote:

>(1) IRV is better than plurality.
>I'm assuming that you and most everyone on this list agrees with this.

Probably a safe assumption, but many think the advantage is marginal.  I 
think the advantage is significant AS LONG AS you only have two strong 
parties and all other parties are more extreme.  But that's a significant 
caveat.

>(2) It is easier to explain IRV than Condorcet.
>Convincing a group to change from plurality to any other voting system is 
>really hard.  IRV is easier to explain because it is similar to a runoff 
>election.  While Condorcet isn't so hard to explain when a winner exists, 
>the concept of cycles is very difficult to explain to the average 
>person.  I think your letter is much too complicated for its intended 
>audience.  People aren't going to get it.

I think that it's entirely acceptable to just leave out the explanation of 
cycles, or relegate it to a footnote.

"The voter ranks the candidates.  From the rankings, we know every voters' 
preference between any two candidates.  We then compare every candidate 
one-on-one against every other candidate, using those preferences.  The 
candidate that wins every one of his or her on-on-one matchups is the winner".

Anyone smart enough to ask about cyclic ties is smart enough to hear the 
answer.

>(3) IRV could be a stepping stone to Condorcet.
>I don't think it makes sense to sabotage efforts to enact IRV.  If 
>anything I think it makes more sense to support IRV as a stepping stone to 
>Condorcet.  Once people are familiar with ranked ballots, they can more 
>easily be persuaded to consider Condorcet.

Maybe.  Or maybe people will be reluctant to change again after going 
through the reform fight once.  Maybe it will be harder to recruit 
volunteers for a new campaign.  Maybe the major parties would oppose a 
second round of reform when they realize it could help the small parties 
even more.  Maybe the results of IRV will be bad, and people will be more 
likely to oppose more reform as a result.

Not saying these things will happen, but they are distinct possibilities.

>(4) Approval doesn't help get to Condorcet.
>While approval voting is certainly easy to explain, you have to get around 
>peoples' knee-jerk reactions of "one person one vote." (I know that OPOV 
>is only about redistricting, but most people don't).  Also approval 
>doesn't provide you with a good stepping stone to Condorcet.

Maybe, or maybe the positive experience with voting reform will help spur 
on the movement.  Maybe people familiar with reform will be more receptive 
to ranked ballots.

Not saying these things will happen, but they are distinct possibilities.

>(5) Condorcet and Approval may violate state constitutions.
>This is beyond the scope of this email, but something to think about.

Wasn't IRV struck down in Detroit or somewhere a long time ago?  I'm just 
trying to say that all voting reform gets attacked as unconstitutional in 
one place or another.  But with the exception of a few outlier methods 
(like Borda) I don't really think the criticisms are valid.

-Adam