QUICK question: small electorate, many candidates

[Jack Santucci]

Gang,

I need your advice. The voting rule needs to accommodate:

1) about 20 voters

2) a ballot with the same number of candidates

3) quick implementation (e.g., via something like demochoice.org)

4) low probability of a tie

No need to prove the merits/demerits of any one rule with this. There won't be any publicity.

Thanks!

Jack

[William Waugh]

How many winners from one election? If 1, I think Approval Voting. Not sure if anything can be done about the probability of a tie, though. Maybe Score{50, 49, 0, -49, -50} would reduce that.

On Friday, November 16, 2018 at 10:26:29 AM UTC-5, Jack Santucci wrote https://groups.google.com/forum/?fromgroups#!topic/electionscience/VCtppGcPhlk

[Jack Santucci]

Yes, one winner.

We've converged on Borda just out of necessity... lots of tools for getting it done, etc. It's a bit of a rush job.

On Fri, Nov 16, 2018 at 11:17 AM William Waugh <2knuw...@snkmail.com> wrote:

How many winners from one election? If 1, I think Approval Voting. Not sure if anything can be done about the probability of a tie, though. Maybe Score{50, 49, 0, -49, -50} would reduce that.

On Friday, November 16, 2018 at 10:26:29 AM UTC-5, Jack Santucci wrote https://groups.google.com/forum/?fromgroups#!topic/electionscience/VCtppGcPhlk

--
You received this message because you are subscribed to the Google Groups "The Center for Election Science" group.
To unsubscribe from this group and stop receiving emails from it, send an email to electionscien...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[William Waugh]

Well, Borda's not any easier to count than Score, is it?

[Andy Jennings]

On Fri, Nov 16, 2018 at 8:26 AM Jack Santucci <jack.s...@gmail.com> wrote:

Gang,

I need your advice. The voting rule needs to accommodate:

1) about 20 voters

2) a ballot with the same number of candidates

Are all of the voters candidates?

 

3) quick implementation (e.g., via something like demochoice.org)

4) low probability of a tie

No need to prove the merits/demerits of any one rule with this. There won't be any publicity.

Thanks!

Jack

--

[Mark Frohnmayer]

You can use http://star.vote to run a STAR Voting election quite easily. With that many candidates, a little more nuance than approval may be desired.

[Jameson Quinn]

I agree, star.vote would be the solution I'd recommend for this case.

[Jack Santucci]

Thanks, all. I've passed it along. Have a great weekend. 

Sent from my iPhone

[NoIRV]

On Friday, November 16, 2018 at 12:42:45 PM UTC-5, Mark Frohnmayer wrote:
> You can use http://star.vote to run a STAR Voting election quite easily. With that many candidates, a little more nuance than approval may be desired.

Who owns star.vote? I think it would be nice to make it so you can optionally change the maximum rating, say to 8 or 10 or 20, perhaps with 5 being the default with "(recommended)".

[Andy Jennings]

Do you have the election data?  Can you post it (maybe anonymized)?

I'm especially interested in seeing the rankings if each voter was also a candidate.  It would relate to some discussion we've had here about "choosing the speaker for a parliament".

~ Andy

[Jack Santucci]

Sadly I do not. 

Sent from my iPhone

[NoIRV]

The best way to elect a leader of a parliament is to use Approval Voting to filter off anyone with under 50% support*, and then use a good election method like Score Voting or 321 on the remaining candidates.

*Or 37.5% if you think having majoritarian speakers is bad.

[Andy Jennings]

NoIRV,

Your suggestion is not bad, but my goal is to find a method which is resistant to a caucus (where a majority of the legislators meet beforehand, choose some extremist from their party to be speaker, and impose that choice on everyone else).

We've discussed some methods in the past that can do this, so it would be good to see some real data on this type of situation.

~ Andy

On Sat, Nov 17, 2018 at 5:33 PM NoIRV <xyzxyz...@gmail.com> wrote:

The best way to elect a leader of a parliament is to use Approval Voting to filter off anyone with under 50% support*, and then use a good election method like Score Voting or 321 on the remaining candidates.

*Or 37.5% if you think having majoritarian speakers is bad.

[NoIRV]

Is it not a tautology that in a 2-party system, either (1) one party can prevent some outcome without the consent of the other, or (2) one party can cause the outcome without the consent of the other? We need to end two party domination.

[Andy Jennings]

NoIRV,

You are correct that a majority can, in theory, generally do whatever they want.  Which means that the majority party, if they vote in lockstep, can pass whatever legislation they like.

But it also means that the minority party can agree with a few of the most moderate legislators from the majority party, form a new majority and "roll the speaker" if the existing speaker is too extreme.  Which should happen far more often than it actually does (in my opinion).

As far as I can tell, caucuses exist because they increase the power of experienced legislators (at the expense of inexperienced ones).  Further, experienced legislators are more likely to come from "safe" seats which, I believe, are more likely to be more extreme.  Even more so if gerrymanders have increased the number of safe seats.

Speakers are not rolled because:

1) It's much easier not to.  It takes a very acute issue to make moderate legislators from the majority party want to join forces with the minority party.

2) Many legislators fear getting "primaried" next election if they turn against their party just as much as they fear losing a general election.

For those reasons, I feel like if we had some voting method that always chose a good centrist legislator and the legislature had to use that method to choose their speaker, then we should end up with a moderate from the majority party as speaker.  Most importantly, it would be done automatically and without drawing a lot of attention (as does a defection or rolling the speaker).  Such a speaker would then be a centrist and would hopefully be difficult to roll.

Such voting methods exist and were discussed in other threads.

I do admit that I can see ways this might fail.  I'm not sure these arguments are 100% right, but they seem more right than wrong to me...

On Mon, Nov 19, 2018 at 8:30 PM NoIRV <xyzxyz...@gmail.com> wrote:

Is it not a tautology that in a 2-party system, either (1) one party can prevent some outcome without the consent of the other, or (2) one party can cause the outcome without the consent of the other? We need to end two party domination.