According to this story, a city initiative has just been filed in Burlington, Vermont, to ask the voters if they wish to repeal Instant Runoff Voting. Earlier this year, the Progressive Party had won the Mayor’s race (Burlington has partisan city elections), even though the Democratic Party nominee had been expected to win. That caused some Burlington voters to dislike IRV.
Hmm… funny how that works! D’s or R’s lose and the law must be changed.
Progressives have won elections in Burlington without IRV. This isn’t about D’s and R’s not liking a progressive being elected.
Kiss owes IRV for his win, since he had the least first place votes.
Maybe voters are not happy with the IRV elected Kiss:
“Burlington Slapped With Lawsuit Over Telecom”
December 16, 2009 …The Kiss administration arranged for $17 million in city loans to support the struggling communications firm. Most City Council members said they knew nothing about the loans.
#
“Editorial: BT viability panel off to a rough start” December 24, 2009
The opening session of a panel set up by the City Council to assess the long-term viability of Burlington Telecom raises one question: Why bother?
The seven committee members — all with day jobs — are charged with reporting their findings to the council by Jan. 15. That’s a mere 25 days from the first meeting Monday, a period that includes three weekends and the year-end holidays.
Compare that to the months Mayor Bob Kiss kept Burlington Telecom’s financial problems and BT’s license violations from the City Council, regulators and Burlington residents.
Meanwhile, the Kiss administration again attempts to spread a cloak of secrecy over matters having to do with Burlington Telecom
Joyce McCloy apparently supports this repeal, which would mean candidates could win on the first round with 40% of the vote — down from the current majority threshold with IRV. The “spoiler” dynamic would be back, and candidates could become mayor who were the last choice of a majority of voters.
Kiss finished second in first choices in 2009 behind the Republican nominee, contrary to what Joyce writes. It was not surprising that Kiss earned more second and third choices from the Democrat and independent(who had double digit support as well).
They leading opponents of IRV are among backers of the Republican who lost the mayoral race in 2009 (including he candidate himself, who seems to be leading the charge) and the Democratic mayoral candidate who lost in 2006. Both heavily outspent Kiss, but lost — and are taking it out on the system, willing to junk the majority rule principle in the process.
4761/9781 is not a majority.
4313/8980 is not a majority.
#4 By this standard, an awful lot of runoff elections don’t have majorities, either. Compare the Burlington IRV election with the runoff election for City Council District 7 last year. In the Mayoral election, only 7% of the ballots cast in the first round were exhausted. In the runoff election for City Council, turnout was 44% lower than that of the general election.
IRV is an EVIL FRAUD — since it ignores most of the votes in a place votes table.
P.R. and A.V. Now = REAL Reform
#5 515/940 is certainly a majority.
Turnout for the 2008 Burlington town meeting day election was higher than that that for the 2009 Burlington town meeting election day. Does that imply voters were confused by the IRV election in 2009, and boycotted the election? Or perhaps some of the folks were coming to vote in the presidential primary on the same date, maybe?
In 2009, voters were probably interested in the mayoral election, and many didn’t pay any attention to the city council race or were relatively indifferent. Those who were interested showed up for the runoff. Turnout for the runoff was about the same as for the town meeting day election in 2007 when there was no mayoral election and no presidential primary.
If there had been a conventional runoff in Burlington in 2009, turnout may well have been higher than it was. Certainly turnout was generally higher for mayoral elections prior to the adoption of IRV.
In Los Angeles, a recent city council district special election saw a turnout in the runoff that was 23% higher than the first round, and 93% higher when the district was last contested in a general election.
The 2009 instant runoff election in Burlington,Vermont suffered from nearly every pathology in the book:
Non monotonicity – where with instant runoff, a voter can hurt their preferred candidate by ranking them first.
A spoiler effect – in this election, Kurt Wright was the spoiler.
The “no show” paradox – Wright supporters who also supported Montrol would have helped him if they hadn’t shown up to vote at all.
Majority failure -the candidate supported by the most voters did not win.
Incumbent protection thanks to name recognition.
Centrally counted votes – instant runoff opened up the election to fraud because votes were not counted where cast.
Please google:
“Voting Paradoxes and Perverse Outcomes: Political Scientist Tony Gierzynski Lays Out A Case Against Instant Runoff Voting”
“Vermont Legislative Workshop”+
“Assessment of Instant Runoff Voting”
and
“Burlington Instant Runoff Election riddled with pathologies”
If you want to help third parties,then make ballot access easier, provide public funding of campaigns, and work hard.
Try Fusion voting (which strengthens third parties), approval voting, or other election methods which are proven to work and don’t harm election transparency.
Botton line – until we can count votes the plain old vanilla way, we can’t even consider the complicated algorithm, the central counting and the shuffling of the votes that IRV requires.
How many times WILL IRV elect a Stalin or Hitler clone when the *middle* is divided — i.e. the S/H clones are in the final IRV top 2 — or even 2 S clones or 2 H clones ???
The S/H clone winner will, of course, claim a mighty majority *mandate* for his/her/its EVIL party hack extremist agenda.
I have better things to focus on than correcting Joyce McCloy’s plethora of errors and assumptions that she posts here. But I did wish to point out one that offends me to all end and needs a response.
She claims that somehow we can’t count IRV by hand, transparently in the public view. That is completely false, and there are many in the IRV supporting movement that are also working for election security, transparency, and accuracy. Sarasota, for instance, is a prime example where IRV activists started an initiative to require paper ballots and audits, as well as IRV. And contrary to her statements, Minneapolis conducts a statistically significant audit of the ballots by hand. So this fallacy she keeps promoting that until we get hand counted and secure voting processes in place, IRV has no place is retarded. For one, Joyce still wouldn’t support IRV (she is a partisan democrat who cares not for diversity of choice and competition, unless they bear a (D) after their name on the ballot). How I really infer that is that she promotes runoffs and plurality election by defending the status quo (meanwhile these systems continue to encourage a two-party duopoly, see Duverger’s Law for more on that).
These partisan Dempublicans who want no further competition, at least contrary to the opinions of 1/5 of our nation, if not significantly MORE, are just exporessing their party line opinion. In fact, I would personally make the case that over half this nation who is ELIGIBLE to vote has grown weary of the TWO PARTY system and wants election systems that allow more competition than two choices. I would make that case because less than 50% of those ELIGIBLE to vote do so in presidential elections, let alone the elections lower down the tier like state house or municipal government.
So I want to say, very clearly, Joyce McCloy doesn’t represent my views, my morals, and my vision for the future. Neither do most partisan Dempublicans or Republicrats. Those of use wanting more from our Democratic Republic are growing, getting better organized, and we intend to continue the struggle for better, more representative, and more INCLUSIVE AND COMPETITIVE election systems. I invite Joyce and her 5 or so buddies to work on implementing their wet dream of approval voting or range voting somewhere so we can see its effects in real world elections.
I invite those who resonate with this disgust with the status quo the Dempublicans and Republicrats represent JOIN THE MOVEMENT. Get in touch with Fairvote.org and get off your butts. Let’s make sure 2012 is spoiler free, yet inclusive of folks like Nader/Paul/Kucinich/McKinney/Brown/Bednarik/ETC. We need better choices than “the puppet on the left” or the “puppet on the right.”
Jim Riley draws a lot of assumptions out of voter’s motivations from some raw data that doesn’t come close to providing that information. Anyone can form assumptions about what occurred from 2008 to 2009. This is called a “hypothesis” in science. Jim, you are skipping the step where you test your theory with experimentation/research, and turning it into your own theory without polling voters or anything to back up your theory. I see lots of similarities to Joyce’s work there. 🙂
I call it gutter science.
See, to know why less people showed up, you would have to call and ask some. That is how SCIENCE works. You can’t just make up why, show me less showed up, and then promote that as science. That is NOT science. that is your opinion backed up by NOTHING. Again, I call this GUTTER SCIENCE.
“I have better things to focus on than correcting Joyce McCloy’s plethora of errors and assumptions that she posts here.”
Well, here goes:
“Non monotonicity – where with instant runoff, a voter can hurt their preferred candidate by ranking them first.”
Non-monotoncity theoretically could have happened under some very limited and contrived circumstances in this election, sort of like balancing a pencil on its point. But it didn’t.
“A spoiler effect – in this election, Kurt Wright was the spoiler.”
A rather loose definition of “spoiler”. Wright was leading in the first and second rounds, and came in a close second to Kiss in the decisive third round. he had a good chance of winning, unlike candidates that are normally called “spoilers”.
Plurality would have elected Wright and Montroll would have been the “spoiler” to Kiss! Under top two, Wright and Kiss would have been in the runoff.
“The “no show†paradox – Wright supporters who also supported Montrol would have helped him if they hadn’t shown up to vote at all.”
Nope. Kiss still would have won if Wright supporters who ranked Montroll above Kiss didn’t show up to vote.
This was ably refuted on Fairvote’s website.
“Majority failure -the candidate supported by the most voters did not win.”
The term “majority failure” is tossed around in a sloppy way by the anti-IRV crowd. If you’re referring to the fact that IRV does not always elect a Condorcet winner, then you should favor Condorcet voting instead of fusion or approval.
If you’re talking specifically about the Burlington election, then the top two runoff system that IRV replaced also would have excluded the Condorcet winner from the final round. Plurality would have given a less favored candidate than IRV.
“Incumbent protection thanks to name recognition.”
Well, then I guess fusion voting is out the window. Many incumbents have won under fusion, including Hillary Clinton (D/WFP-NY) in 2006. Obviously, plurality is as well.
“Centrally counted votes – instant runoff opened up the election to fraud because votes were not counted where cast.”
Can you name one IRV election that was stolen because of central counting? We can consider calling it a “pathology” if you can come up with one. If you have a specific accusation of fraud to make against the Board of Elections in Burlington, it is your duty as a citizen to bring it forth. If not, cut it out.
Attn — all math folks —
ALL election systems have problems with 3 or more choices.
i.e. if a third choice is added to 2 current choices, then such third choice may —
1. beat both C > A > B
2. be beaten by both A > B > C
3. beat one but be beaten by the other (requiring a tie breaker). C > A BUT B > C — *circular tie*.
Much more complex with 4 or more choices.
See [Kenneth] Arrow’s Theorem on the internet.
There are defective systems and then there is the super defective IRV — which ignores most of the data in a place votes table.
-1—2—3
A1–XX–XX
B1–XX–XX
C1–C2–XX
XX = IRV IGNORED VOTES with just 3 choices — worse and worse with 4 or more choices.
i.e. ALL of the A2 and B2 votes might just happen to be for C — i.e. a possible *moderate* middle choice.
A.V. for executive / judicial offices — Pending major education about head to head math — i.e. voting YES/NO on each choice AND having Number Votes 1, 2, etc.
P.R. for legislative bodies.
#11 In 2008, the number of votes cast in the presidential preference primary by ward was 1576, 1228, 1578, 2189, 2147, 1792, 1985; the votes cast for councilor were 1326, 1099, 1389, 1793, 1913, 1594, 1943. That is 0.950 correlation. Correlation between the votes per councilor and number registered was -0.161. The lowest ratios between councilor votes and presidential primaries was in Ward 4 (82%), where there was a single unopposed candidate, and in Ward 1 (84%), where there were 4 candidates, but the winner had 77% of the vote, to 11%, 10%, and 1% for the also-rans. The best performance was in Ward 7 (98%), where there was a very close race (15 vote, 0.77% plurality), and with the only race with 2 candidates from parties holding presidential preference primary that day. In the other 4 wards, the ratio was 98% or 99%.
2008: Turnout was driven by the presidential primary on the same day.
In 2007, turnout for the councilor races was 39% of the presidential primary, with 53% and 49% for the two northern wards (4 and 7), 38% and 37% for the two southern wards (5 and 6), and 27%, 28%, and 28% for the three central wards. Incidentally, almost as many ballots were cast in Ward 4, as in Wards 1, 2, and 3 combined.
No presidential election, no mayoral election, not so many people show up.
In 2009, councilor votes cast were 87% of councilor votes cast in 2008, and 172% of the number cast in 2007. The correlation between the councilor votes cast per ward and mayoral votes cast per ward was 0.994. Highest participation relative to the mayoral race was in Wards 4 (97%) and 7 (99%), which had the two closest councilor races; lowest was in in Wards 1 (88%) and 5 (89%) which had the most lopsided councilor.
2009: Turnout was driven by the mayoral election.
Votes cast in Ward 7 Councilor race:
2007: 981 (top of the ballot)
2008: 1943 (presidential preference primary)
2009: 1696 (mayoral election)
2009: 940 (runoff)
Conclusion, turnout for the 2009 runoff in Ward 7 was similar to the turnout in 2007, the last time that the councilor races were top of the ballot.
I like this YOUTUBE Video on the Burlington Election. It documents the problems pretty well.
http://www.youtube.com/watch?v=img9y2AYTQA