IRV: Not So Fast, Not So Simple

We have several concerns with Instant Runoff Voting (IRV).  At first it seems simple and appealing, however, a closer look reveals concerns and unappreciated consequences, including:

  • IRV can be confusing and difficult for voters to understand.  It can result in a much bigger ballot, requesting voters to rank multiple candidates and not make a mistake.  A single race may take an entire ballot that today can handle an entire municipal election or a combined state and Federal election.
  • IRV does not solve the problems it is intended to fix.  It claims to prevent the “wrong” candidates from being elected by providing a runoff until one candidate has over 50% of the votes.  In reality, sometimes it does and sometimes it does not.  IRV can cause different candidates to be elected than real runoff voting and different candidates to be elected than when the leading candidate  from the first round is selected.  Often IRV awards a race to a less popular candidate than these other methods.
  • IRV is complex and time consuming to count, especially in races covering multiple polling places.  This is because  elimination requires exact information on each ballot ranking be centrally counted or each round centrally counted.  Each round is critical to the final result, thus each round may require all absentee ballots and provisional ballots be counted and perhaps require a manual recount to determine the eliminated candidate.

This last point is covered in an editorial from Minnesota – counting a municipal race my take eight weeks and involve spreadsheet calculations <read>

Minneapolis won’t have the assistance of scanning machines to tabulate votes if no candidate receives over half of the first choices, delaying results for up to eight weeks. Since Minnesota does not certify machines that can count ranked choice ballots, and because cities are prohibited from using non-certified machines, city elections workers must hand-count each ballot. Fortunately, the city attorney has assured elections officials that they can use Microsoft Excel to help tally. Otherwise, votes would still be being sorted well into 2010.

Update: It did not take 8 weeks to count.  Perhaps because of IRV voters stayed away: <read>

What if they gave an election and nobody came? Well, almost nobody came to the City Election last Tuesday. R. T. Rybak got 73 percent of the vote for Mayor, but that was only 33,217 votes. That’s the smallest amount a winning candidate has gotten running for Mayor since 1910. Rybak’s total was less than 15 percent of the number of registered voters. In the last three municipal elections the LOSING candidates for Mayor polled almost as high as Rybak’s winning total

Update:  But it is eight days and some races are still being counted <read>

Several races may be announced Wednesday, the paper says, but the closest council races, in the Fourth and Fifth Wards, won’t come until a later batch of counting.

Update:  17 days past the election and still counting<read>

But the biggest surprise for O’Connor and the elections staff has been how quickly the hand count process has gone. Originally, O’Connor expected that they wouldn’t be ready to announce unofficial winners until December. But they’ve been able to do so several weeks earlier than planned.

Here are some video’s that graphically explain IRV:

  • A clear explination of IRV, what could go wrong, what the MN Supreme Court said and what happened in Aspen Colorado <video>
  • A more detailed version of how a candidate can loose by gaining support <video>
  • We often think of IRV for a single candidate, but it can be used when we vote for several people for town council. Yet its not simple.  Here is one way of counting the votes <video>
  • Third parties tend to like the idea of IRV.  Watch this video to be disapointed <video>

Update: 3/3/2010 Burlington voters repeal IRV <read>

A raucus celebration at Norm’s Grill in the new North End Tuesday night marked a victories end to the YES on 5 campaign in Burlington. A mix of Democrats and Republicans celebrated the defeat of Instant Runoff Voting.

After a debate that brought out big names like Bernie Sanders and Howard Dean urging a NO on 5 vote, Burlington residents voted to repeal IRV by a vote of 3972 to 3669, a margin of only 303 votes.

“It was a grass roots effort and we’re really proud of the people that worked on this, we’re ecstatic,” said IRV opponent Chuck Seleen while celebrating the victory.

Wards 4 and 7 were the only two to vote YES on 5. Both had much higher voter turnout than the other wards. Mayor Bob Kiss says that is a sign most of the city still supports IRV and he suggests the city continues the discussion and possibly have yet another vote on IRV.

“This is a modest voter turnout. I think a bigger voter turnout might have a different result and particularly in the case because if you have two wards that are voting very heavily and I think other wards are more modest so if we heat up that debate we might get a better view of Burlingtonians’ real view on IRV,” Kiss said at City Hall moments after the results were tallied.

Several city councilors and election observers have suggested the IRV vote is a referendum on Kiss and his administration, as it tries to deal with the mess of Burlington Telecom. Kurt Wright, who was elected back to the city council, Tuesday, by defeating incumbent democrat Russ Ellis, says it is time Kiss gets the message.

Two comments:

Saying “Wards 4 and 7 were the only two to vote YES on 5. Both had much higher voter turnout than the other wards.” is pure spin. With equal lack of usefulness one could say that say that if more voters had come out in other districts it might well have gone down further.

Apparently Howard Dean changed his mind from his statement on this video.

Update: Report: Ranked Choice Voting Causes Confusion, Fatigue In SF’s District 10 <read>

The new system produced some surprising outcomes, and in the wake of the results there has been a flurry of criticism and praise of how RCV plays out for voters.

Facebooktwittergoogle_plusredditpinterestlinkedintumblrmailFacebooktwittergoogle_plusredditpinterestlinkedintumblrmail

Comments are closed.