An interesting conversation in light of the fact that the winner of Philly's mayoral election took only a plurality. Promoted by MDC.
The talk of strategic voting got me thinking. The current mayoral election in Philly is a good example of a situation where Instant Runoff Voting would make it more likely that the eventual winner would enjoy true majority support.
For example, someone may think Rep. Dwight Evans is the best person of the five for the job, but may also be nervous about recent polls and prefer to do "strategic voting" instead. As many of you know, under I.R.V., an Evans supporter could designate him as their first-round choice, and if he didn't finish first or second, the person's second choice would get the second-round vote. Etc.
I am not an expert, but I wonder if the legislature could permit I.R.V. for Philly only, or if the city council or charter could simply authorize it.
Hmm.
25 Pa.C.S. 3167 states taht "Except as otherwise provided by law, the persons receiving the highest number of votes for any office at any election shall be declared elected to such office, up to the number required by law to be elected thereto," and I wonder if that would preclude the City's doing IRV by itself.
What I'd prefer -- because I believe IRV is a bit complicated -- is an open primary system for the mayor's race. Everyone's in at once; top two (unless someone gets 50%+) move forward to November.
I'd settle
for a Primary runoff of the top two if no one gets 50%.
Actual Use of IRV?
Is there any actual problem with I.R.V. other than the worry that people are too stupid to get it?
Does anyone know of any reports or a good assessment of IRV in London or any of the American cities that have used it? Call me an idealist, but hate to concede that people are too stupid to rank their choice from one, two, and three.
I prefer IRV to top-two run-off because top-two still would have strategic voting. Under top-two, Evans supporters would still feel pressure to vote for their second choice.
----
I support Michael Nutter for Mayor. My slate.
IRV & Cambridge, MA
I lived in Cambridge for quite a few years - their system is essentially the same as IRV. It took some getting used to when I first moved there - but - most people who moved there got the hang of it rather quickly & natives just grew up with it. I'm sure there would be an intitial adjustment period - but - it wasn't really a problem.
Major props to YPP
Dan P. at the Next Mayor is talking about you all on PBS!
I supported Michael Nutter for Mayor.
Dan Pohlig from The Next
Dan Pohlig from The Next Mayor is talking about this thread right now on WHYY.
MDC, we are the biggest
MDC, we are the biggest nerds in the world.
with the possible exception
with the possible exception of Dan Pohlig, who is talking about a blog on public TV.
Proudly wearing my nerd hat
...which I should have had on to cover my ridiculous hair. Last time I trust that stylist.
Anyway, thanks for the content and for opening that thread. It's something I wanted to bring up at some point last night and I think we should continue talking about how we can possibly change this system so that we're not talking about the same situation in 8 years.
Adam B., see you at lunch!
Keep in mind, this doesn't
Keep in mind, this doesn't necessarily solve anything. As we saw with campaign finance law, you have to be prepared for the non-normal scenario.
Say you have IRV or open primary ...
You send the top two people to the november election and they are both Dems.
Katz runs as independent is on the November election with his Republican base.
Tom Knox gets tapped by the Libertarian Party and dumps his millions of dollars into the race.
You would still have a possibility of the winner not having a 50% vote.
You have two strong candidates splitting the Dem machine vote, one candidate pulling the GOP vote and another candidate siphoning off of all parties.
I also don't see anything wrong with the winner having 35% of the vote in a 5 person race and leaving it at that.
Also, if anything, this election goes to show how possible of a little impact strategic voting really is. You would assume Nutter never would have climbed the polls if peopel were going to vote strategically. Obviously the last month of polling, people were voting their opinion, not strategically because they switched to the "underdog".
nope
An open primary with a top-two runoff forecloses the option of later ballot entry. The only way you make it to November is to survive the primary; there's no exterior route. (Otherwise, wouldn't everyone just form his own party?)
I don't think that is legal,
I don't think that is legal, but maybe my understanding is wrong (I am not fluent in all aspects of election law).
In PA elections, there are distinguishing categories of Major Party, based on number of registrations, right? That covers Dems and Republicans. As of right now, Libertarians do not get to have primaries, but have to get signatures to be on the Nov ballot.
If that is the case, wouldn't you have to significantly rewrite election law to do IRV or open primary? I would also assume it would preclude the ability to have Philly do their own election laws separate from the State.
Hmm
Let's suppose that you switched the Philadelphia mayoral race to nonpartisan/open primary. Couldn't you just have everybody vote for the mayoral ticket just as everybody votes for the questions? In other words, "major party" would still be a relevant electoral category. It just wouldn't apply to the mayor's race.
A more complicated matter would be the city council races, where (at least at-large) some parties are guaranteed seats by the charter.
it would require a real change to allow it
But then, once in place, it'd be "anyone who has [x] signatures gets in the primary," and there's no alternate means of November ballot entry.
Do you think it would need
Do you think it would need to be a state-wide change or do you you think Philly could fight to have their own rules?
Also, would the ballot have listed what party they are registered to or would it just have a list of names?
good question
Probably statewide, but you'd think Republicans would go for anything that gave their party a chance to win here.
The ballot would list . . . whatever the law said it should.
Well obviously we are
Well obviously we are discussing what we thinnk the law should do. ;) I was curious of your thoughts on that as well as what is done in current open primary areas.
how would it give Republicans a better chance to win? They are already (pretty much) guaranteed a top two spot in November.
The thing is . . .
. . . under the current system, if Sam Katz can't beat John Street in 2003, then no person labeled "Republican" can really win. But give people a way to reach November without a party label, and more folks have a shot.
I actually rather like the Jungle Primary
as in Louisiana. With a top two runoff, you're guaranteed that one person will walk away with 50% + 1, which seems to me more democratic. I don't like the idea of candidates winning with a plurality.
Ideally, we'd have a primary, a primary runoff, and then a general election. (That also happens to be the way they did it in the old south and New York City: other places where one party dominated).
It just seems wierd,
It just seems wierd, inherently.
Basically you're having a primary for a primary.
If the primary is the election,
as it has been for 50 years, then why on earth shouldn't we?
Your statement is making the
Your statement is making the assumption nothing changes and is a logic flaw.
I used the word "if" for a reason
All other things being equal, would the process be more democratic if we had some form of a primary runoff? I would say so, yes.
So let's follow this
So let's follow this conversation ...
Basically, we have a City that is predominantly one party and some people from that party what to change the election rules that would most likely send two candidates to the november ballot from their party as opposed to one from each.
The problem is, these are primaries. These are ways the party chooses their candidate.
Personally, I feel any party, with proper minimum level of support, should be able to have a candidate on the November ballot. It is a way to insure everyone has a voice.
Just because a City is overwhelmingly one party, it should nto be justification to effectively remove them from the November ballot.
If you want to eliminate
If you want to eliminate party based primaries, then eliminate party representation from the elections as well.
I want three rounds, if needed
1. Primary
2. Primary runoff if no candidate gets 50% +1 in round 1.
3. General election.
Three days is a waste
Do you own stock in a campaign flyer printing press or do you rely on campaign street money to feed your family?
I think to do in three days what can be done in one (with nonpartisan IRV) is a waste of time and resources. Rather than engage the apathetic public, they would be burdened about having to follow three different elections a year and vote three times a year. ("What are we voting for this time?")
Why do you think three days would better than one?
I don't see it being fairer to the favorite candidates or the outsiders. It just favors those that take the label of minority political parties (Republican, Green, or Libertarian, etc.) who get to stand in on an election as an apparent equal to whoever emerges from the two Dem primaries campaigns (increasing from 5 months to what, 7 or 8?). This is a system that only campaign geeks would follow or care about.
-----
I support Michael Nutter for mayor in November.
Thanks for the personal attack
No, I have no financial stake. The question I'm trying to answer is how we can ensure that a candidate will always win with 50% + 1. I suppose we could have no parties and two rounds.
Well, do we have to have a
Well, do we have to have a situation where the candidate has 50%?
Also, do they "really" have 50% after two rounds of voting?
Say the top two candidates get 10,000 votes each and three others each had 9,000.
You have a second election or runoff, but the supporters of the three candidates don't vote because they no longer care, their guys is out of it. In all technicality, the winner only has 50% of the votes of the two remaining candidates and not of the people who actually voted.
Again, I really don't see what the problem is with a multiple candidate race won with less than 50%.
Why do people think that is a problem?
Me either.
Me either.
What if
the person who most people like least still gets a plurality of the votes. Allowing that person to win seems undemocratic to me.
Democracy is voting for who
Democracy is voting for who you want. It isn't voting for who you don't want.
Indeed. And IRV lets voters
Indeed. And IRV lets voters practice that maxim more fully.
That's a pretty good way
of defining what we currently have as best.
As Aardhart writes below,
As Aardhart writes below, the problem is when the plurality winner would lose head-to-head against the 2nd-place (or lower) finisher. This forces voters to either vote strategically rather than for whom they think would do the best job; or to vote for a candidate who they know will not win.
You mentioned Dan's post where he suggests that Greg's ward would've endorsed miller, but his ward does not account for all 5,000 of his votes, even assuming 100% of his in-ward supporters would have followed his advice. With IRV, with Greg coming in fourth, then the ballots of those who selected him as their first choice are re-examined. Those who made no second choice are discarded, and the others distributed according to their 2nd choice.
As a generous approximation, lets say 50% of his votes went to Donna, and the others split evenly between Cindy and Irv. That would get us to:
DRM: 12026
Cindy: 9313
Irv: 8877
Donna still doesn't have a majority, so we look at Irv's ballots, and distribute them to DRM or Cindy. Any votes that went from Greg to Irv are pretty obviously anti-DRM votes, and therefore would have Cindy or noone for third choice. I'd guess that most 1st choice Irv votes would be followed by either Cindy then Greg, or Greg than Cindy. Since Greg's already out, they all go to Cindy. Since nothing's ever perfect, let's assume 15% of Irv's ballots have no valid choices left, and 15% have DRM as the remaining valid choice.
That gets us to:
DRM: 13358
Cindy: 15526
"I would win if no one else
"I would win if no one else ran" I do not view as a valid reason.
The number game also doesn't work because, who says Cindy Bass is the best anti-incumbent supporter? Who says that, if you did a runoff of antis, Irv wins?
Really, voting doesn't need to be that complicated. I could somewhat justify the argument for that type of process for a November election, but these are primaries. This is how your party is choosing your candidate.
When it comes down to it, even though DRM didn't have the majority fo the vote, even more people didn't want Cindy or Irv or Greg.
You can't blame the system. Cindy, Irv and Greg failed to reach enough people with their vision.
Which also leads to the other issue. It isn't just about "we need DRM out" because if that is the case, other candidates would bow out. The core of it is, Bass, Irv and Greg wanted to win, not to get DRM to lose.
No personal attack
No personal attack was intended. I don't think an actual personal should be understood. One obviously cannot feed a family with campaign street money for one person. It was a joke.
My counter-offer, no parties and ONE round with IRV.
-----
I support Michael Nutter for mayor in November.
I have no problem with the
I have no problem with the outcome of this mayoral primary. I think Nutter would have won head-to-head against anyone, as well as in any 3-way or 4-way race. As well as with IRV or any other method.
The problem is in situations like in the 8th district or last year's Texas governor's race, where the winner of a 4-way race quite possibly could not have beaten any of the other candidates head-to-head.
I think any change to IRV or regular run-offs would also be accompanied with a change to non-partisan elections. This wouldn't mean that Democrats wouldn't be elected. It would just mean that they wouldn't have to then run against a patsy in a mostly meaningless election after winning the actual election.
I prefer IRV to traditional top-two run-offs, because one day is cheaper than two and I think that the voters can be educated to use it fine (until I am confronted with actual evidence; I would have thought stories would be easier to google). With top-two run-offs, even though only two names can be on the ballot for the run-off, I think write-ins have caused the winner to have less than a majority, which causes confusion.
-----
I support Michael Nutter for mayor in November.
not necessarily
If you get 50%+ in the primary, that's it. No run against a patsy to follow.
That's the Louisiana system
I think we should have a primary, a primary runoff, and then a general election (as in New York).
I'd add to that a general
I'd add to that a general runoff if nobody gets 50% in the general election, in order to prevent 3rd-party spoilers. Georgia, for example, uses this system of a runoff after both primary and general elections. In practice, the Georgia system comes close to solving the problems addressed by IRV, but does not present the same user-interface problems. Also, it still allows members of a political party the opportunity to choose their nominee, unlike the Louisiana system.
Expensive, but fair
I think that's how Paul Coverdell won his Senate seat in 1992. It otherwise would have gone to the Democrat with a Libertarian acting as spoiller.
Why separate elections
Why separate elections rather than IRV?
It's a question
of setting ease of use against the expensive of holding another election.
I guess the real question is
I guess the real question is ...
How much and how many people feel the primary system is bad? I sort of have the feeling that an overwhelming majority of the average voter doesn't see or realise a problem.
Also, I think it is a bit of a conclusion jump that a vote for the non-incumbent is a vote for anyone that isn't the incumbent. As in the 8th, are people voting for Paulmier because they don't like DRM or because they like Paulmier better? I believe Dan even said that Greg supporters were going to DRM when he was kicked off.
I would be curious if there are any actual studies on such topics.
Let's let the voters decide, then
With IRV, for little additional expense, you don't have council members with less than 1/3 of their party supporting them. If Donna wins under IRV, that's still disappointing, but at least she can say she has the majority of her party behind her. Holding office with 32% of your party's support is just embarrassing.
When I was passing out lit
When I was passing out lit for Greg the other day, the committee people there noted my dedication. They said that they had to hand it to me because I worked hard all day. They then noted that Greg was lucky to have a friend like me. Well, I think that Greg has a lot of friends like me.
It may help to separate his supporters from his Committee people supporters. Many of his supporters would probably not have supported another campaign, but probably would have voted for Irv or Cindy. Many of his committee people supporters would be supporting Donna, because they live in areas where Donna is very popular, and they don’t want to rock the boat. Plus, they just don’t know the other candidates. Personal relationships out way ideas in the Ward system all the time. I’m not saying that that is how it should be, but that is how it is.