[EM] What should an ideal single-winner method achieve?

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

[EM] What should an ideal single-winner method achieve?

Toby Pereira
With all the discussion of different single-winner methods and the criteria they pass and fail, I'm interested to know what you think the "ideal" method should hope to achieve. For example, some people might want to maximise utility summed across the voters. Others might want to find the candidate that is closest to the "median voter". For others it might be more about obeying some sort of majority criterion (e.g. Condorcet). Etc.

Personally, the measure that makes most sense to me is to maximise utility. But this doesn't automatically mean score voting (where a score could simply be seen as a utility rating of a candidate), at least in part because strategies that voters adopt might reduce its effectiveness. Obviously a voting method also needs to be simple enough to understand (in terms of voting and understanding how the winner is calculated), and it might be that different types of election suit different methods.

Toby

----
Election-Methods mailing list - see https://electorama.com/em for list info
Reply | Threaded
Open this post in threaded view
|

Re: [EM] What should an ideal single-winner method achieve?

robert bristow-johnson

 

 

1. One person, one vote.  this does **not** maximize utility because it says that if I really really like my Candidate A much more than Candidate B and you only slightly prefer your Candidate B over A, your vote for B>A counts no less (nor no more) than my vote for A>B.  This does not maximize utility but sticks with the principle that all of we citizen voters have an equal franchise in government, which is fundamental to me.  More fundamental than maximizing the mean utility over the electorate..

2. Removing the burden of tactical voting from voters and removing the temptation of strategic voting from campaigns.  This implies some method to prevent or at least impede spoilers from swinging an election.

3. Majority rule, which is strictly defined only for binary decisions.  This leads to Condorcet: If more voters mark their ballots preferring Candidate A over Candidate B than the number of voters marking their ballots to the contrary, then Candidate B is not elected.

4. Election integrity.  This means paper ballots (and, i presume, optical scan ballot box).  The physical instrument that the voter marks must have the candidate names on it, so there is no "registration" or alignment problem where the voter thinks they're voting for "A", but somehow "B" gets marked.  Another thing this means is simple straightforward rules for tallying votes and deciding the winner.  Another would be precinct-summability.

 

That pretty much leaves me with ranked-choice ballots where ties on the ballot are allowed.  Enough ranking levels that every candidate on the ballot can be ranked (so no one is "disenfranchised").  And a good and simple Condorcet method.  I am still leaning toward Tideman RP Margins.

 

and FUCK THE ELECTORAL COLLEGE!!!   line them all up against the wall and shoot them.

(still pissed about today's Supreme Court gerrymandering decision.  the bastards.  we should shoot the five of them.)

r b-j


---------------------------- Original Message ----------------------------
Subject: [EM] What should an ideal single-winner method achieve?
From: "Toby Pereira" <[hidden email]>
Date: Thu, June 27, 2019 7:31 am
To: "EM List" <[hidden email]>
--------------------------------------------------------------------------

> With all the discussion of different single-winner methods and the criteria they pass and fail, I'm interested to know what you think the "ideal" method should hope to achieve. For example, some people might want to maximise utility summed across the voters. Others might want to find the candidate that is closest to the "median voter". For others it might be more about obeying some sort of majority criterion (e.g. Condorcet). Etc.
> Personally, the measure that makes most sense to me is to maximise utility. But this doesn't automatically mean score voting (where a score could simply be seen as a utility rating of a candidate), at least in part because strategies that voters adopt might reduce its effectiveness. Obviously a voting method also needs to be simple enough to understand (in terms of voting and understanding how the winner is calculated), and it might be that different types of election suit different methods.
> Toby----
> Election-Methods mailing list - see https://electorama.com/em for list info
>

 

 

 


--

r b-j                         [hidden email]

"Imagination is more important than knowledge."
 

 

 

 


----
Election-Methods mailing list - see https://electorama.com/em for list info
Reply | Threaded
Open this post in threaded view
|

Re: [EM] What should an ideal single-winner method achieve?

fdpk69p6uq
In reply to this post by Toby Pereira
On Thu, Jun 27, 2019 at 10:31 AM Toby Pereira wrote:
Personally, the measure that makes most sense to me is to maximise utility. But this doesn't automatically mean score voting (where a score could simply be seen as a utility rating of a candidate), at least in part because strategies that voters adopt might reduce its effectiveness.

Yep, maximizing utility is the correct answer (= finding the candidate that best represents the entire electorate), and the best way to get that kind of information is with rated ballots.


Condorcet ranked systems will also typically elect the Utilitarian Winner, but only because real-world voting behavior typically follows a spatial model with unimodal distribution.



Voter strategy on rated ballots is way overblown.

1. It doesn't happen much in real elections.
2. Even with strategic voters, rated systems still tend to produce more representative outcomes than other systems.
3. There are many voting methods based on rated ballots besides pure Score Voting, which have extra features to discourage strategic exaggeration. STAR, 321, MJ, IRNR, etc.  Maligning pure Score, as if it's the only way to tally rated ballots, is a bit of a strawman argument.

----
Election-Methods mailing list - see https://electorama.com/em for list info
Reply | Threaded
Open this post in threaded view
|

Re: [EM] What should an ideal single-winner method achieve?

Kristofer Munsterhjelm-3
On 28/06/2019 00.41, [hidden email] wrote:

> On Thu, Jun 27, 2019 at 10:31 AM Toby Pereira wrote:
>
>     Personally, the measure that makes most sense to me is to maximise
>     utility. But this doesn't automatically mean score voting (where a
>     score could simply be seen as a utility rating of a candidate), at
>     least in part because strategies that voters adopt might reduce its
>     effectiveness.
>
>
> Yep, maximizing utility is the correct answer (= finding the candidate
> that best represents the entire electorate), and the best way to get
> that kind of information is with rated ballots.
>
> http://leastevil.blogspot.com/2012/03/tyranny-of-majority-weak-preferences.html
>
> Condorcet ranked systems will also typically elect the Utilitarian
> Winner, but only because real-world voting behavior typically follows a
> spatial model with unimodal distribution.
>
> https://www.sciencedirect.com/science/article/abs/pii/S0304406815000518
>
> http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.504.3181&rep=rep1&type=pdf
>
> Voter strategy on rated ballots is way overblown.
>
> 1. It doesn't happen much in real elections.

Could you give some sources for use of rated ballots in actual elections
with a large electorate (e.g. not polls or surveys)? I don't think I've
seen any such, and it would be useful to have.
----
Election-Methods mailing list - see https://electorama.com/em for list info
Reply | Threaded
Open this post in threaded view
|

Re: [EM] What should an ideal single-winner method achieve?

Toby Pereira
In reply to this post by robert bristow-johnson
I definitely understand this position. But under your number 3, you have put "If more voters mark their ballots preferring Candidate A over Candidate B than the number of voters marking their ballots to the contrary, then Candidate B is not elected.", which as we all know can be violated under a Condorcet method because of the possibility of a cycle. But obviously I understand that your principle would be to stick to that as much as possible. But by having majority rule as a principle rather than just something for use in practice where possible, it does also mean that the "best" winner (rather than most practical winner) between A and B can change with the addition of candidate C. And arguably this is a contradiction.

The principles for a vote can arguably change depending on who is voting and what they are voting on. If it is a group of friends voting on where to go to eat, then I see it as perfectly reasonable for a minority to have a say if they absolutely hate one of the options. But with national elections for public office, it becomes a more competitive thing with people doing their utmost to elect the candidate they want rather than find the best compromise. But is that necessarily the best thing for society?

I would say we do ideally need a system that is simple, removes the burden of tactical voting as much as possible and is also perceived as fair by voters. But if there were several methods that largely passed these criteria, then I would still favour one that in practice is more likely to elect the highest utility candidate. Bear in mind also that under a Condorcet method, tactical voting can lead to bad results, and that under certain assumptions, score voting can lead to the Condorcet winner being elected more than under an actual Condorcet method. https://rangevoting.org/DH3.html https://rangevoting.org/AppCW.html

Toby


On Thursday, 27 June 2019, 21:07:45 BST, robert bristow-johnson <[hidden email]> wrote:

1. One person, one vote.  this does **not** maximize utility because it says that if I really really like my Candidate A much more than Candidate B and you only slightly prefer your Candidate B over A, your vote for B>A counts no less (nor no more) than my vote for A>B.  This does not maximize utility but sticks with the principle that all of we citizen voters have an equal franchise in government, which is fundamental to me.  More fundamental than maximizing the mean utility over the electorate..

2. Removing the burden of tactical voting from voters and removing the temptation of strategic voting from campaigns.  This implies some method to prevent or at least impede spoilers from swinging an election.

3. Majority rule, which is strictly defined only for binary decisions.  This leads to Condorcet: If more voters mark their ballots preferring Candidate A over Candidate B than the number of voters marking their ballots to the contrary, then Candidate B is not elected.

4. Election integrity.  This means paper ballots (and, i presume, optical scan ballot box).  The physical instrument that the voter marks must have the candidate names on it, so there is no "registration" or alignment problem where the voter thinks they're voting for "A", but somehow "B" gets marked.  Another thing this means is simple straightforward rules for tallying votes and deciding the winner.  Another would be precinct-summability.

 

That pretty much leaves me with ranked-choice ballots where ties on the ballot are allowed.  Enough ranking levels that every candidate on the ballot can be ranked (so no one is "disenfranchised").  And a good and simple Condorcet method.  I am still leaning toward Tideman RP Margins.

 

and FUCK THE ELECTORAL COLLEGE!!!   line them all up against the wall and shoot them.

(still pissed about today's Supreme Court gerrymandering decision.  the bastards.  we should shoot the five of them.)

r b-j


---------------------------- Original Message ----------------------------
Subject: [EM] What should an ideal single-winner method achieve?
From: "Toby Pereira" <[hidden email]>
Date: Thu, June 27, 2019 7:31 am
To: "EM List" <[hidden email]>
--------------------------------------------------------------------------

> With all the discussion of different single-winner methods and the criteria they pass and fail, I'm interested to know what you think the "ideal" method should hope to achieve. For example, some people might want to maximise utility summed across the voters. Others might want to find the candidate that is closest to the "median voter". For others it might be more about obeying some sort of majority criterion (e.g. Condorcet). Etc.
> Personally, the measure that makes most sense to me is to maximise utility. But this doesn't automatically mean score voting (where a score could simply be seen as a utility rating of a candidate), at least in part because strategies that voters adopt might reduce its effectiveness. Obviously a voting method also needs to be simple enough to understand (in terms of voting and understanding how the winner is calculated), and it might be that different types of election suit different methods.
> Toby----

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

 

 

 


--

r b-j                         [hidden email]

"Imagination is more important than knowledge."


 

 

 

 

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


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