[Election-Methods] RE : Re: RE : Re: RE : Re: Simple two candidate election

[Kevin Venzke]

Dave,

--- Dave Ketchum <[EMAIL PROTECTED]> a écrit :
> Time to give up on this exchange unless we can find a way to a language 
> for communication.

I can clarify anything you want.

> >>Seems to me we are agreed as to goals, but are tripping over what I
> meant
> >>when I said "effectively".  I MEANT to learn how Range works, and then
> do
> >>  the best I could within those rules, such as:
> >>  best - max score to try for winning.
> >>  worst - min score to try for losing.
> >>  soso - this is the hard one - less than max, for I hope for best
> to
> >>win; probably near max to improve odds for soso if best loses.
> >>
> >>I see this as properly using "effectively" as an English word.
> > 
> > 
> > I'm not sure what you mean. The strategy you suggest for "soso" is not
> very
> > good, which to my mind makes it not effective.
> 
> How did we get here?
>   I want best to win, so I certainly want to give all others a lower 
> rating.

This isn't Condorcet though. Range doesn't pay you any special attention
for rating one candidate above all the others.

>   But, if best loses, I want to have done the best I can for soso, my
> second choice.

You're on the right track here. You correctly calculate that you want to do
the best you can for soso. On the other end, you have no way to specify
what should happen if best wins or loses. That works in Condorcet, but not
in Range. The incentive to rate soso below best is gone, in Range.

> In these words I see 3 groups:
>   best - which I hope will win.
>   soso - almost belongs in the same group as best, but I want to 
> indicate preference for best.
>   worst - of which I hope for all such to lose.

The question is whether you want to indicate preference for best (over
soso) even if, under the given election method, it is a bad strategy.

Kevin Venzke


  
_ 
Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers Yahoo! Mail 
http://mail.yahoo.fr


Election-Methods mailing list - see http://electorama.com/em for list info

Re: [Election-Methods] Fwd: FYI - FairVote MN Responds to Lawsuit Against IRV

[Dave Ketchum]

I have to hope the lawsuit fails, for it is attacking ranked choice, which 
is a good way to give voters power.

Could wish for attacking weaknesses of IRV, for it is too ready to declare 
undeserving candidates as winners.

Related to this, I question whether the many claiming to understand IRV 
speak from ACTUALLY understanding its weaknesses.

DWK

On Sat, 22 Dec 2007 22:12:04 -0700 Kathy Dopp wrote:
> -
> What do you think of this lawsuit?
> Kathy
> 
> 
> -- Forwarded message --
> From: Raging Grannie (Wanda B) <[EMAIL PROTECTED]>
> Date: Dec 22, 2007 8:19 PM
> Subject: FYI - FairVote MN Responds to Lawsuit Against IRV
> 
> 
> 
> 
> go to FairVoteMN.org if you have trouble viewing
> this newsletter
> \"FairVote
> 
> []
> 
> 
> For immediate release
> 
> Contact: Kelly O’Brien,
> [EMAIL PROTECTED], 612-227-9102
> Jeanne Massey,
> [EMAIL PROTECTED], 763-807-2550
> FairVote Minnesota Responds to Lawsuit Against Instant Runoff Voting
> 
> Minneapolis, MN (December 20, 2007)—FairVote Minnesota, a nonprofit,
> nonpartisan organization leading the effort to institute instant runoff
> voting in Minnesota, is issuing a response to news of a lawsuit brought
> against the City of Minneapolis and various elected officials by an Eden
> Prairie-based organization that is opposed to the pending use of instant
> runoff voting (a.k.a. single transferable vote or ranked choice voting) in
> the City of Minneapolis.
> 
> The lawsuit alleges that instant runoff voting (IRV) is unconstitutional
> and violates the principle of “one person, one vote.† Additionally, the
> plaintiffs contend that voters won’t understand how to vote using instant
> runoff voting.
> 
> FairVote Minnesota presents the following facts as guidance in the
> discussion of this lawsuit.
> 
> 1. IRV has been upheld on the principle of "one person, one vote" in legal
> challenges following its adoption in Cambridge, Mass. (Moore v. Election
> Commissioners of Cambridge (1941) and in Ann Arbor, Mich. (Stephenson v.
> Ann Arbor Board of Canvassers (1975).
> 
> Sources:
> http://www.rwinters.com/docs/moore.htm;
> http://www.fairvote.org/library/statutes/legal/irv.htm
> 
> 2. No legal challenges are threatening instant runoff voting anywhere it's
> in use—San Francisco, Calif.; Cambridge, Mass.; Burlington, Vermont; Takoma
> Park, Maryland and Cary and Hendersonville, North Carolina.
> 
> 3. Election exit polls in cities using IRV all show voters in overwhelming
> numbers not only understand IRV, but prefer it to the old way of voting.
> The share of voters indicating they understood IRV well or very well the
> first time using IRV: San Francisco – 87%, Burlington – 89%; Takoma Park –
> 88%; 8%; Cary – 95%; Hendersonville – 86%.
> http://www.fairvotemn.org/sites/fairvotemn.org/files/Exit%20Survey%20Summary_2007_FINAL.doc
> 
> 4. As discussed in an opinion piece in the September 30, 2007 Star Tribune
> by attorney and professor David Schultz, the 1915 legal case cited in the
> lawsuit, Brown v. Smallwood, was not about instant runoff voting. It was
> about a voting method that effectively gave Duluth citizens two votes in
> some situations, a clear violation of both the Minnesota and United States
> constitutions. The concern of that decision was based on what the courts
> now call the "one person, one vote" standard. IRV does not violate this
> standard because it does not give anyone two votes. It simply allows voters
> to rank their preferred candidates.
> 
> 5. Schultz further explained that since 1915 American democracy has
> matured. The political process now seeks to provide more choices for voters
> than it once did, as evidenced by numerous ballot access court decisions
> that have made it possible for third party candidates such as Jesse Ventura
> to run for office. The courts, mindful of voters’ demands for more
> options when voting, have properly responded to the demand of citizens in
> interpreting election laws to empower and not limit options on election day.
> 
> 6. The lawsuit’s plaintiffs advocate to keep low-turnout municipal
> primaries and to make them partisan to ensure all parties are represented
> on the ballot. However, IRV shares the goal of ensuring choice on the
> ballot. In fact, it makes sure that all candidates appear on the general
> election ballot, regardless of party. With IRV, no candidate can be
> eliminated in a low-turnout election whose chances might be different in a
> general election.
> 
> The following are suggested sources for further information on instant
> runoff voting.
> 
> Prof. David Schultz, Hamline University. 651-523-2858
> Rob Richie, executive director, FairVote (national). 301-270-4616
> Tony Solgard, former president, FairVote Minnesota. 612-242-5642
> Jeanne Mass

Re: [Election-Methods] RE : Re: RE : Re: Simple two candidate election

[rob brown]

On Dec 23, 2007 8:18 PM, Dave Ketchum <[EMAIL PROTECTED]> wrote:

> How did we get here?
> I want best to win, so I certainly want to give all others a lower
> rating.
> But, if best loses, I want to have done the best I can for soso, my
> second choice.
>

Yes, that's what Range seems to promise, but it doesn't deliver.

A system that allows you to express such a preference, while actually being
in your interest to do so, is what is needed.

Range is not that. At all.

Election-Methods mailing list - see http://electorama.com/em for list info

Re: [Election-Methods] RE : Re: RE : Re: Simple two candidate election

[rob brown]

On Dec 23, 2007 7:28 PM, Kevin Venzke <[EMAIL PROTECTED]> wrote:

> For Range to work as well as hoped, it is important that voters do *not*
> realize how to vote effectively. Or, that they choose not to.
>

You've summed up, in 25 words or less, everything that is wrong about Range.

Very nice job. :)

Election-Methods mailing list - see http://electorama.com/em for list info

Re: [Election-Methods] RE : Re: RE : Re: Simple two candidate election

[Dave Ketchum]

Time to give up on this exchange unless we can find a way to a language 
for communication.

On Mon, 24 Dec 2007 04:28:57 +0100 (CET) Kevin Venzke wrote:
> Dave,
> 
> --- Dave Ketchum <[EMAIL PROTECTED]> a écrit :
> 
>>Seems to me we are agreed as to goals, but are tripping over what I meant
>>when I said "effectively".  I MEANT to learn how Range works, and then do
>>  the best I could within those rules, such as:
>>  best - max score to try for winning.
>>  worst - min score to try for losing.
>>  soso - this is the hard one - less than max, for I hope for best to
>>win; probably near max to improve odds for soso if best loses.
>>
>>I see this as properly using "effectively" as an English word.
> 
> 
> I'm not sure what you mean. The strategy you suggest for "soso" is not very
> good, which to my mind makes it not effective.

How did we get here?
  I want best to win, so I certainly want to give all others a lower 
rating.
  But, if best loses, I want to have done the best I can for soso, my 
second choice.
> 
> Basically it is not in your interests to rate soso lower than best. Range
> does not make any note of which candidates you like better than which, so
> you're just wasting voting power if you try to indicate it.
> 
> If you want to maximize the power of your vote, you should split the
> candidates into two groups (better than average vs. worse than average, if
> you have no polling info), and use the entire power of your vote to try to
> break any ties occurring between these two groups. Unless there are only
> about three voters, you can't hope to do better than this.

In these words I see 3 groups:
  best - which I hope will win.
  soso - almost belongs in the same group as best, but I want to 
indicate preference for best.
  worst - of which I hope for all such to lose.
> 
> For Range to work as well as hoped, it is important that voters do *not*
> realize how to vote effectively. Or, that they choose not to.
> 
> Kevin Venzke
-- 
  [EMAIL PROTECTED]people.clarityconnect.com/webpages3/davek
  Dave Ketchum   108 Halstead Ave, Owego, NY  13827-1708   607-687-5026
Do to no one what you would not want done to you.
  If you want peace, work for justice.




Election-Methods mailing list - see http://electorama.com/em for list info

[Election-Methods] RE : Re: RE : Re: Simple two candidate election

[Kevin Venzke]

Dave,

--- Dave Ketchum <[EMAIL PROTECTED]> a écrit :
> Seems to me we are agreed as to goals, but are tripping over what I meant
> when I said "effectively".  I MEANT to learn how Range works, and then do
>   the best I could within those rules, such as:
>   best - max score to try for winning.
>   worst - min score to try for losing.
>   soso - this is the hard one - less than max, for I hope for best to
> win; probably near max to improve odds for soso if best loses.
> 
> I see this as properly using "effectively" as an English word.

I'm not sure what you mean. The strategy you suggest for "soso" is not very
good, which to my mind makes it not effective.

Basically it is not in your interests to rate soso lower than best. Range
does not make any note of which candidates you like better than which, so
you're just wasting voting power if you try to indicate it.

If you want to maximize the power of your vote, you should split the
candidates into two groups (better than average vs. worse than average, if
you have no polling info), and use the entire power of your vote to try to
break any ties occurring between these two groups. Unless there are only
about three voters, you can't hope to do better than this.

For Range to work as well as hoped, it is important that voters do *not*
realize how to vote effectively. Or, that they choose not to.

Kevin Venzke


  
_ 
Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers Yahoo! Mail 
http://mail.yahoo.fr


Election-Methods mailing list - see http://electorama.com/em for list info

Re: [Election-Methods] RE : Re: Simple two candidate election

[Dave Ketchum]

On Sun, 23 Dec 2007 23:44:56 +0100 (CET) Kevin Venzke wrote:
> Dave,
> 
> --- Dave Ketchum <[EMAIL PROTECTED]> a écrit :
> 
>>So we look for a method suitable for two, or more, candidates, such as:
>>  Approval - cannot rank 3 candidates as best, worst, and soso 
>>(matters when voter wants to indicate best is better than soso AND soso
>>is 
>>better than worst).
>>  Condorcet - allows any number of ranks plus equal ranking such as 
>>Approval offers.
>>  IRV - like Condorcet, excluding equal ranking and occasionally 
>>awarding disappointing choices as to winner.
>>  Range - like Condorcet but uses ratings instead of ranks.  Ratings 
>>give finer control than ranks, but demand that voters learn to assign
>>them effectively.
> 
Seems to me we are agreed as to goals, but are tripping over what I meant 
when I said "effectively".  I MEANT to learn how Range works, and then do 
  the best I could within those rules, such as:
  best - max score to try for winning.
  worst - min score to try for losing.
  soso - this is the hard one - less than max, for I hope for best to 
win; probably near max to improve odds for soso if best loses.

I see this as properly using "effectively" as an English word.
> 
> "Assign them effectively"? In Range that means rating everybody either the
> maximum or minimum score. That's not "fine control" surely. It doesn't
> matter how many buttons it has if you're not supposed to push them.
> 
> I would rather say that Range hopes that voters *do not* learn to assign
> the ratings effectively. Or if they learn how, they choose not to.
> 
> Kevin Venzke
-- 
  [EMAIL PROTECTED]people.clarityconnect.com/webpages3/davek
  Dave Ketchum   108 Halstead Ave, Owego, NY  13827-1708   607-687-5026
Do to no one what you would not want done to you.
  If you want peace, work for justice.




Election-Methods mailing list - see http://electorama.com/em for list info

Re: [Election-Methods] Simple two candidate election

[rob brown]

On Dec 23, 2007 2:00 PM, Juho <[EMAIL PROTECTED]> wrote:

> Works with humans too. Three friends living in different places might
> agree to meet at a place that has equal distance to all three homes. Or they
> might select a place that minimizes the sum of the distances (maybe they
> will share the travelling costs).
>

What you are describing is very different from what happens with eusocial
animals.

A situation where a small number of people can work things out, and where
reciprocity plays into it (i.e. "you got your way last time", etc) is
completely, totally different from the type of situation where voting is
needed, such as a large scale election where most of the voters don't know
one another.

What you describe isn't eusociality, which is the complete lack of self
interest found in animals that don't reproduce directly.  A 100% rational,
self-interested person may wish to allow someone else to get their way when
preference strengths differ, with the expectation that it will be
reciprocated when the preference strengths are reversed.  But a large
election has no mechanism for reciprocity.

I could imagine a voting system that might address this issue for larger
groups, but it isn't Range.


> Parents could agree to watch a Disney movie with their child although they
> might like a war movie more themselves.
>

Parents and children are, from a Darwinian perspective pretty much the  same
as eusocial animals, motivation-wise.  The children's interests are almost
completely aligned with the interests of the parents, since the children are
the vector for the parents getting their genes into future generations. In
my view, all motivation can be traced back to the attempt to get genes into
future generations.

But in politics / competitive situations rules of course tend to be
> different. And the election methods should be chosen accordingly.
>

I'm with you there.

Election-Methods mailing list - see http://electorama.com/em for list info

[Election-Methods] RE : Re: Simple two candidate election

[Kevin Venzke]

Dave,

--- Dave Ketchum <[EMAIL PROTECTED]> a écrit :
> So we look for a method suitable for two, or more, candidates, such as:
>   Approval - cannot rank 3 candidates as best, worst, and soso 
> (matters when voter wants to indicate best is better than soso AND soso
> is 
> better than worst).
>   Condorcet - allows any number of ranks plus equal ranking such as 
> Approval offers.
>   IRV - like Condorcet, excluding equal ranking and occasionally 
> awarding disappointing choices as to winner.
>   Range - like Condorcet but uses ratings instead of ranks.  Ratings 
> give finer control than ranks, but demand that voters learn to assign
> them effectively.

"Assign them effectively"? In Range that means rating everybody either the
maximum or minimum score. That's not "fine control" surely. It doesn't
matter how many buttons it has if you're not supposed to push them.

I would rather say that Range hopes that voters *do not* learn to assign
the ratings effectively. Or if they learn how, they choose not to.

Kevin Venzke


  
_ 
Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers Yahoo! Mail 
http://mail.yahoo.fr


Election-Methods mailing list - see http://electorama.com/em for list info

Re: [Election-Methods] Simple two candidate election

[Juho]

On Dec 23, 2007, at 22:52 , rob brown wrote:


On Dec 23, 2007 10:49 AM, Juho <[EMAIL PROTECTED]> wrote:
Now bak to the question. Majority vote may often not yield the
optimum outcome (from some chosen high level theoretical viewpoint)
but majority vote may well be considered to be the best practical
method for competitive two candidate elections.

Yeah, well that is pretty much my opinion.  With the exception that  
the "high level theoretical viewpoint" that supports Range (whether  
in 2 candidate elections or ones with 3 or more), by ignoring the  
competitive nature of elections, makes it a pretty non-useful  
theory for elections held among human beings, who have a history of  
being rather competitive by nature ( http://en.wikipedia.org/wiki/ 
War ).


It works pretty good for predicting how non-competive eusocial  
animals such as bees might want to hold elections, however.   
( http://rangevoting.org/ApisMellifera.html , http:// 
en.wikipedia.org/wiki/Eusociality )


Works with humans too. Three friends living in different places might  
agree to meet at a place that has equal distance to all three homes.  
Or they might select a place that minimizes the sum of the distances  
(maybe they will share the travelling costs). Parents could agree to  
watch a Disney movie with their child although they might like a war  
movie more themselves. In that case they are maybe seeking a solution  
that maximizes the worst of the three personal utilities. There are  
many functions that can be used in determining the optimum outcome.


But in politics / competitive situations rules of course tend to be  
different. And the election methods should be chosen accordingly.


Juho


Election-Methods mailing list - see http://electorama.com/em for list info

Re: [Election-Methods] Simple two candidate election

[rob brown]

On Dec 23, 2007 10:49 AM, Juho <[EMAIL PROTECTED]> wrote:

> Now bak to the question. Majority vote may often not yield the
> optimum outcome (from some chosen high level theoretical viewpoint)
> but majority vote may well be considered to be the best practical
> method for competitive two candidate elections.
>

Yeah, well that is pretty much my opinion.  With the exception that the
"high level theoretical viewpoint" that supports Range (whether in 2
candidate elections or ones with 3 or more), by ignoring the competitive
nature of elections, makes it a pretty non-useful theory for elections held
among human beings, who have a history of being rather competitive by nature
( http://en.wikipedia.org/wiki/War ).

It works pretty good for predicting how non-competive eusocial animals such
as bees might want to hold elections, however.  (
http://rangevoting.org/ApisMellifera.html ,
http://en.wikipedia.org/wiki/Eusociality )

Election-Methods mailing list - see http://electorama.com/em for list info

Re: [Election-Methods] Simple two candidate election

[rob brown]

On Dec 23, 2007 11:43 AM, Dave Ketchum <[EMAIL PROTECTED]> wrote:

>  > And then Gore loses, just like he did with Plurality.
>
> But what other voters do also matters.
>

Surely you understood that I had considered that others might do the same
thing.

The point is, the issue of vote splitting with plurality is there with range
as well.

 > People already have the ability to not vote.  I skip candidates and
>  > propositions I don?t know much about or care much about.  Here in san
>  > francisco, I even get to rank candidates, and will rank candidates
>  > equally if that?s how I feel.  But if I actually have a preference, I
>  > will express it as strongly as I can.
>
> This puzzles.  I understand that San Francisco has ranking via IRV -
> but assume that it does not permit equal ranking, which IRV has
> trouble implementing.


I allows you to express a first choice but treat second and third as equal.


If I had my way, it would be condorcet, and would allow equal rankings at
any position.

I will, however, say I much prefer IRV to plurality, and think it is a step
in the right direction.  First we had runoff elections (which people seemed
to understand), then we have an automated version of the same that involves
ranking candidates, and hopefully in the future we will have a better way of
tabulating it that also uses ranked candidates.  As much as I understand the
imperfections of IRV, it still thrills me to get to rank candidates.



>   > With 3-or-more choice elections, I think the benefits of Range
> Voting
>  > would be HUGE.
>  > >
>  > Understood.  Disagree.
>
>  > I think people already feel like their vote doesn?t count for much.
>  >
> Range and Condorcet give benefits whenever there are more than two
> seriously contending candidates.
>

I am well aware that with 3 candidates there are problems that are not there
in two candidate elections.  (
http://karmatics.com/voting/movienite.htmlsums up my view on this) To
be clear, I think Condorcet (and related things
like DSV) addresses them (if imperfectly), while Range doesn't come close.
Approval is somewhere in the middle.

My point in talking about two candidate elections is that the main
differentiating effect Range has (allowing people to voluntarily express
their opinion with less than maximum weight) could equally apply to two
candidate elections as well.  By looking at it where the complexities of
3-candidate elections are absent, its absurdity might be more apparent.

Election-Methods mailing list - see http://electorama.com/em for list info

Re: [Election-Methods] Simple two candidate election

[Dave Ketchum]

On Sat, 22 Dec 2007 22:46:17 -0800 rob brown wrote:

 > On Dec 22, 2007 8:04 PM, Jan Kok <[EMAIL PROTECTED]
 > > wrote:
 >
...
 >
 > Yes, some people might vote weak preferences in their first Range
 > Voting election, then "learn their lesson" when their preference lost
 > and they would vote strongly thereafter. But I sort of wonder about
 > the intelligence of people who would cast a *weak* vote for some
 > choice X, and then get upset when X lost. What were they thinking when
 > they voted?!?
 >
 >
 > Well, I once again mention the 2000 election. Someone who likes Nader
 > best, Gore second best, and Bush the least would probably say "Oh cool!
 > Before, I couldn?t say that I liked Nader best because it would take my
 > vote away from Gore who really needs it.  Now with this newfangled Range
 > voting system I can express that.  I?ll give 100 to Nader, 80 to Gore,
 > and 0 to Bush.  Woo hoo!"
 >
 > And then Gore loses, just like he did with Plurality.

But what other voters do also matters.
 >
 > What exactly have you guys solved again?
 >
 > The thing is, even more so that plurality, Range ASKS people to vote
 > "honestly".  I didn?t feel dishonest at all when I voted for Gore (even
 > though I liked Nader better.  I just felt like I was giving a vote to
 > the candidate that could best use it.  (some, of course, people feel
 > otherwise, and interpret a plurality to mean "I like this candidate best")
 >
 > Range, however, has a strong implication that you are SUPPOSED to
 > express more "rich" preferences than, say, approval.  After all, why
 > else would they go to the expense of offering the more complex ballot?
 >
 > I know that?s an opinion, and is psychological speculationbut I *am
 > * a UI designer, and psychological speculation is what I do. :)
 >
 > By the way, I?m not about to go on a crusade to implement Range Voting
 > for two-choice elections. I don?t think the benefit of perhaps
 > slightly better outcomes would be worth the effort and cost. I?m just
 > responding to Rob?s question: Yes, I think there could be _some_
 > benefit, sometimes, to using RV in binary-choice elections.
 >
 >
 > Well, if range is implemented, I?d assume it would apply to two
 > candidate elections as well, wouldn?t it?

Sure - we start with two candidates and Plurality.

Then we get into trouble when there are, sometimes, more than two candidates.

So we look for a method suitable for two, or more, candidates, such as:
  Approval - cannot rank 3 candidates as best, worst, and soso 
(matters when voter wants to indicate best is better than soso AND soso is 
better than worst).
  Condorcet - allows any number of ranks plus equal ranking such as 
Approval offers.
  IRV - like Condorcet, excluding equal ranking and occasionally 
awarding disappointing choices as to winner.
  Range - like Condorcet but uses ratings instead of ranks.  Ratings 
give finer control than ranks, but demand that voters learn to assign them 
effectively.
 >
 > My point, however, is that on two candidate elections, where you don?t
 > have the potential for cycles and all these other problems you have with
 > multicandidate elections, it makes more clear the absurdity  (in my
 > opinion) of asking people to weaken their vote.

The ability of the methods above to handle more than two candidates does 
not demand any weakening of votes when there are only two candidates.
 >
 > People already have the ability to not vote.  I skip candidates and
 > propositions I don?t know much about or care much about.  Here in san
 > francisco, I even get to rank candidates, and will rank candidates
 > equally if that?s how I feel.  But if I actually have a preference, I
 > will express it as strongly as I can.

This puzzles.  I understand that San Francisco has ranking via IRV -
but assume that it does not permit equal ranking, which IRV has
trouble implementing.
 >
 > I think people already feel like their vote doesn?t count for much.
 > Expecting them to reduce it further is nuts, in my opinion.
 >
 > With 3-or-more choice elections, I think the benefits of Range Voting
 > would be HUGE.
 >
 >
 > Understood.  Disagree.

Range and Condorcet give benefits whenever there are more than two 
seriously contending candidates.

-- 
   [EMAIL PROTECTED]people.clarityconnect.com/webpages3/davek
   Dave Ketchum   108 Halstead Ave, Owego, NY  13827-1708   607-687-5026
 Do to no one what you would not want done to you.
   If you want peace, work for justice.





Election-Methods mailing list - see http://electorama.com/em for list info

Re: [Election-Methods] Simple two candidate election

[Juho]

On Dec 23, 2007, at 3:39 , rob brown wrote:

> Say you've got an election for two candidates.

> Let's further assume that there are a large enough number of voters  
> that we can assume that most don't know one another.  This isn't a  
> pizza party or a local club, but a large scale vote.  Also assume  
> that while some people are somewhat on the fence, a lot of others  
> have very strong feelings about it on one side or the other.  And  
> assume anyone is welcome to abstain from voting.
>
> So once again, the question is this:
>
> In this case, do you consider a majority vote optimum?

The election seems to be a typical competitive election where people  
want to win or defend what they think is best for them or for some  
others or for all.

Under these typical circumstances majority based methods may well be  
the recommended systems to use. I wouldn't say that majority vote is  
"the optimum". It may well be that the minority favourite (that may  
have e.g. better average utility) would be the ideal alternative to  
elect.

Sometimes it may also be optimal (from one point of view) to elect  
the majority favourite instead of the the one with best utility -  
just to respect the majority opinion and to avoid further fighting  
that could occur if the best utility alternative would be chosen.

Majority vote is a good and well working tool (works in competitive  
and strategic environments too) if one wants to give one equal vote  
to everyone (a rather good democratic principle).

Now bak to the question. Majority vote may often not yield the  
optimum outcome (from some chosen high level theoretical viewpoint)  
but majority vote may well be considered to be the best practical  
method for competitive two candidate elections.

Juho


P.S. There are also circumstances where methods like Range work well,  
but that is not typical e.g. in the competitive political  
environment. One could try to give also an opportunity to the  
alternative with best average utility to win. One could e.g. first  
arrange a Range vote and after that a majority vote between the  
majority winner of the first election and the utility winner (if they  
differ). At the second round the majority would thus be offered an  
opportunity to donate the victory to the best (Range style) utility  
alternative (now they know how popular that alternative is). In many  
circumstances also the informative Range votes could however be  
strategic (many voters would feel no need to weaken the position of  
their favourite and thereby invite others to pick the (Range style)  
utility winner at the second round).



___ 
To help you stay safe and secure online, we've developed the all new Yahoo! 
Security Centre. http://uk.security.yahoo.com

Election-Methods mailing list - see http://electorama.com/em for list info

Re: [Election-Methods] Borda-elimination, a Condorcet method for public elections?

[Markus Schulze]

Hallo,

the following example demonstrates that the Baldwin
method violates reversal symmetry.

Situation #1:

5 ACB
4 BAC
2 CBA

The initial Borda scores are 14 for candidate A,
10 for candidate B, and 9 for candidate C.

Candidate C is eliminated, because candidate C
has the lowest Borda score.

The new Borda scores are 5 for candidate A
and 6 for candidate B.

Candidate A is eliminated, because candidate A
has the lower Borda score.

Thus, candidate B is the Baldwin winner.

Situation #2:

The individual rankings are inverted.

5 BCA
4 CAB
2 ABC

The initial Borda scores are 8 for candidate A,
12 for candidate B, and 13 for candidate C.

Candidate A is eliminated, because candidate A
has the lowest Borda score.

The new Borda scores are 7 for candidate B
and 4 for candidate C.

Candidate C is eliminated, because candidate C
has the lower Borda score.

Thus, candidate B is the Baldwin winner.

Markus Schulze




Election-Methods mailing list - see http://electorama.com/em for list info