Wednesday, January 6, 2010

IRV - Let's Do the Numbers

A referendum whether to continue IRV (Instant Runoff Voting) in Burlington's mayoral elections will be on the March 2nd Town Meeting Ballot.

Watch the video below... an analysis of the 2009 Burlington mayoral election, which was an IRV election. Watch closely as the narrator does the numbers count for that election. Do you think tallying up the results is confusing? I know it's confusing! - I was in Contois Auditorium last March as the results were brought in from all 7 wards and "counted." Have you ever been tricked by a shell game?


  1. This comment has been removed by the author.

  2. I watched the video and frankly it is rife with improper analyses. What it comes down to is the argument that if the election was different, it would have been different. Well, frankly, BFD. It makes assumptions about what people would have done and how they would have voted under different conditions based on party affiliation. And it treats a three-way election as a series of three two-way elections, which is invalid on its face.

    The video winds up claiming that Montroll had "majority support" even though M. came in 3rd on the first round (W=3297, K=2982, M=2554) with only 28.4% of the vote, close to 20 percentage points less than the 48.0% the IRV gave to Kiss.

    In the "traditional" voting scheme the video appears to favor, Wright would have won. In a traditional run-off scenario, Montroll would have been the one dumped while Kiss and Wright went head-to-head - with Kiss the front-runner (i.e., predicted winner) because Montroll voters preferred Kiss over Wright in sufficient numbers.

    Which, despite the sneers about "vote shuffling," appears to be what happened - which is why it's called "instant run-off."

    The point being that in no traditional scenario would Montroll have won and the attempt to prove something entirely different is clever figure-mongering but entirely invalid.

  3. We don't need to be lectured or told how to vote. IRV does not work as advertisd and I'm voting to repeal it.


  4. John -

    First, know that I don't live in Burlington so this has no effect on me. I know nothing about the three candidates discussed beyond the numbers. I'm just an outside observer.

    But when you say "IRV does not work as advertised," I'm confused. IRV is "instant run-off" and the result appears to have been what would have been expected with a traditional (i.e., non-"instant") run-off. So just how did it "not work as advertised?"

  5. LarryE

    In a traditional voting method, top two run-off, we don't know who would have won. You really can't just extrapolate can you?

    That's the flaw with "Instant" runoff, because the two candidates do not get to have debates, and the voters do not get a chance to really view the two finalists up close, Mano y Mano. In addition, having the losers endorse one of the candidates also plays a part in the decisions in a general as well. In about 33% of the time, there is a flip from the winner in the primary than in the general due to increased debates, voter knowledge, and other factors. IRV shortchanges the voter in this respect. Even Terry Bouricius, the Burlington's IRV expert said IRV puts a bigger burden on the voter to get to know more about the candidates. Traditional elections provides much better chance to get to know the candidates.

    As far as how IRV is being clearly falsely advertised, I think it's time to put up that other IRV Video on the 2009 Burlington election that is circulating.

  6. William -

    you can't extrapolate

    No, but that's largely beside the point to the issue at hand. If you're going to say IRV "shortchanges the voter," you could just as easily "extrapolate" that voters are short-changed by not having an even longer campaign, more debates, more ads, more mailings, more whatever, any part of which could hypothetically increase "voter knowledge" and affect voter decisions. Any system, whatever it is, has to stop somewhere.

    other factors

    Significant among which, surely, is to what extent supporters of elimnated candidates prefer one of the remaining two. Which is what IRV is designed to reveal. Yes, again hypothetically those preferences might change in the course of a traditional run-off - but basing an argument against IRV as a system on that gets right back into that sort of speculation about outcomes we were supposed to be avoiding.

    Ultimately, I do find it strange to say that a system that requires greated voter involvement is a black mark against it.

    Two last things and then I'm done with this thread:

    One is that IRV does have some clear positive aspects, particularly, the opening up the political process to more voices by virtually if not completely eliminating the pressure for strict "lesser of two evils" voting. Meanwhile, the arguments against it seem to be based almost exclusively on hypotheticals and digging out the rare exceptional cases or circumstances.

    Two, perhaps more to the point, none of this in any way addresses my original topic, which was that the video designed to shoot down IRV used bogus arguments to reach a patently bogus conclusion (Montroll should have won) when under either a "first past the post" or a traditional run-off he would have come in third.

  7. Anonymous troll comment (without a hint of identification, as requested in comments "policy") not adding to this discussion has been deleted. - JayV

  8. Nice discussion, peeps. My main complaint about IRV is that the proponents write tomes in comments on other blogs telling people how to vote. I don't want to be told how to vote. KISS, KISS. :P

  9. say JayV,

    is it okay to discuss what works and what doesn't work and why we think one is better than another? is that the same as telling you how to vote?

    i tried posting to this before, and have been unsuccessful. let's see if this works.


Thanks for visiting.

Please be considerate... no off-topic, racist, sexist or homophobic comments.

Comment moderation is on.

No anonymous comments will be accepted..