[EM] More non-altruistic attacks on IRV usage.

[Dave Ketchum]

Condorcet is easy for voters to move to for it is a strong, but  
simple, step up from FPTP and:
1.  Ranking means ability indicate order of varying desires of liking  
candidates.
2.  But ranking is much less of a task than Score's rating where you  
have to calculate the difference in value of A vs B, and express this  
difference as a number.
3.  More detail below.

Not against PR here - PR is not suitable for electing a single-winner.

On Nov 26, 2011, at 10:31 PM, robert bristow-johnson wrote:
> On 11/26/11 6:58 PM, matt welland wrote:
>> On Sat, 2011-11-26 at 16:56 -0500, robert bristow-johnson wrote:
      The next two are related, though not directly quoted.
>>   On Sat, 2011-11-26 at 1:39 AM, robert bristow-johnson wrote:
>>
>>     On Sat, 2011-11-24 at 10:47 PM, Dave Ketchum wrote:

Initial topic is IRV.

>>> the counterexample, again, is Burlington Vermont.  Dems haven't  
>>> sat in
>>> the mayor's chair for decades.
>> Is this due to a split of the liberal vote by progressives or other
>> liberal blocs? Or is it due to a truly Republican leaning  
>> demographic?
> Burlington is, for the U.S., a very very liberal town with a well- 
> educated and activist populace.  it's the origin of Ben & Jerry's  
> and now these two guys are starting a movement ( http://movetoamend.org/ 
>  ) to get a constitutional amendment to reverse the obscene Citizens  
> United ruling of the Supreme Court.
>
> the far north end of Burlington (called the "New North End", also  
> where i live) is a little more suburban in appearance and here is  
> where the GOP hangs in this town.
>
> the mayors have been Progs with an occasional GOP.  it is precisely  
> the "center squeeze" syndrome and IRV didn't solve that problem. and  
> without getting Condorcet adopted, i am not sure how it will be  
> reversed.
>
>> Also, do folks generally see approval as better than or worse than  
>> IRV?
> they don't know anything about Approval (or Score or Borda or  
> Bucklin or Condorcet) despite some effort by me to illustrate it  
> regarding the state senate race in our county.
>
> to attain some measure of proportional representation w.r.t.  
> geography, state senate districts are either divided ( http://www.leg.state.vt.us/lms/legdir/districts.asp?Body=S 
>  ) or, in the case of our county, have an unusually large number, 6,  
> of state senators all elected at large.  this means that besides  
> running against Progs and GOP, the Dems are running against each  
> other.  as a consequence, even though we are allowed to vote for as  
> many as 6, everyone that i know (bullet) votes for 1 or 2 or maybe  
> 3.  effectively, it is no different than Approval voting.
>
> but the only voting methods folks generally see here are FPTP, FPTP  
> with a delayed runoff, and IRV.  and, thanks to FairVote, nearly  
> everyone are ignorant of other methods to tabulate the ranked ballot  
> than the STV method in IRV.
>> To me Approval seems to solve the spoiler problem without introducing
>> any unstable weirdness and it is much simpler and cheaper to do than
>> IRV.
> unless one were to bullet vote (which would make Approval degenerate  
> to FPTP), there is no way to express one's favorite over other  
> candidates that one approves of.  it forces a burden of tactical  
> voting onto voters who have to decide whether or not they will vote  
> for their 2nd favorite candidate.  i've repeated this over and over  
> and over again on this list.  while Score voting demands too much  
> reflection and information from voters, Approval voting extracts too  
> little information from voters.  both saddle voters with the need  
> for calculation (and strategy) that the ranked ballot does not.   
> both Score and Approval are non-starters, because of the nature of  
> the ballot.  but a ranked ballot is not a non-starter, even if we  
> lost it recently here in Burlington.  we just need to unlearn what  
> FairVote did and decouple the concept of ranked-choice voting from  
> IRV.

Back to promoting Condorcet:

It is easier to understand the basics the voter needs to know:
1.  Voting is the same as for IRV, except equal ranking is also  
permitted.
2.  A voter familiar with FPTP can express the same thoughts, with the  
same definitions and power, by approving of a single candidate and  
ranking only that candidate.  Often few will want to approve more than  
one for offices such as Clerk or Coroner (but makes sense for ballots  
to permit ranking for the rare incidents of more controversy in even  
such offices).
3.  To emphasize point 2, a voter satisfied with FPTP voting is not  
seriously handicapped by not instantly learning Condorcet details -  
what is already known is enough to pick and rank a single candidate.
4.  Condorcet counting, unlike IRV's, requires reading all that the  
voters vote in one pass at each reading station and then combining the  
readings at one location to determine results.
5.  Do not have FPTP's need for primaries.
6.  Do not have FPTP's need for runoffs - because voters can express  
themselves more completely, the leader is deserving of winning with  
less than the majority that is truly needed with FPTP.

That the x*x matrix used in counting tells much about candidate  
strength, among the weaker as well as the stronger, can be useful to  
all who are interested in this data.

While there are a zillion Condorcet methods, picking among them is not  
that difficult if you select only among those that award winning to  
the CW when there is one (and there usually is - it takes three  
leading candidates to form a cycle and thus not have a CW).  Further,  
the situation we often complain about - having election power owned by  
two parties - says no cycle until some third party grows some muscle.
>
> -- 
>
> r b-j                  rbj at audioimagination.com
>
> "Imagination is more important than knowledge."

Dave Ketchum
>
> ----