posts to days old
of topics
with text

Instant Runoff Voting

Paul (16 Jun 2004 5:03): Whether you already support Instant Runoff Voting or have never heard of it, visit They need 200,000 signatures of registered Washington State voters by December 2004.
The idea is: when you vote for, say, President, you can rank your choices: if your first choice wins, okay; if not, your vote counts toward your second-choice. Cambridge, MA has been using this system for 60 years to elect its 9-member council (the IRV rules for electing to multiple-seat bodies ensure that, say, if 33% of Cambridge wants Republican councillors then the council will include 3 Republicans).
Paul (14 Nov 2004 11:35): Having read a lot at, I no longer strongly support Instant Runoff Voting. They prove not only that Condorcet (pairwise) voting is better, but also that IRV is hideously bad whenever a third candidate is a viable winner.
Paul (14 Nov 2004 12:08): Consider the following fantasy:
44% - 1st:Nader, 2nd:Kerry
22% - 1st:Kerry, 2nd:Bush
33% - 1st:Bush

IRV would say: Kerry has fewest 1st-choice votes, so loses; now Bush wins.

But note that the pairwise defeats here are cyclical. This is an inherent ambiguity of the voters' preferences:
Nader beats Kerry (44% vs 22%)
Bush beats Nader (55% vs 44%)
Kerry beats Bush (66% vs 33%)
Condorcet drops the smallest defeat, Nader's over Kerry, then Kerry wins.

And if the Bush voters ranked Bush, Nader, Kerry? Then
Nader beats Kerry (77% vs 22%)
Bush beats Nader (55% vs 44%)
Kerry beats Bush (66% vs 33%)
dropping the smallest defeat, Bush's over Nader, then Nader would win.
m (15 Nov 2004 0:26): wait - by smallest defeat, you mean defeat in which the victor got the lowest percentage?
m (15 Nov 2004 1:10): Ok. I thought about your description of condorcet for a while, and I'm convinced that in an election with three political factions, no one group can swing the election their way by changing how they rank the candidates (which I initially thought would be the case after looking at your example where bush voters didn't rank the other candidates). I guess two groups can collude and make things go their way indefinitely, but that's the same thing as reducing the number of groups.

When there are more than three, it's confusing.
m (16 Apr 2015 9:57): Bruce Schneier on the problems with this year's Hugo Award:

(1) I think the best choice would be to do nothing. It's not at all obvious that this is anything other than a temporary aberration, and that any fixes won't be subject to a different set of abuses and need to be fixed again. I think the worst situation would be a series of rule changes in a continuous effort to stave off different abuses. I don't think highly of a bureaucracy that tinkers with election rules until it gets the results it wants.

(2) If we choose to ignore (1), the second-best choice is to modify the electorate. The problem isn't the rules of the vote; the problem is that a voting bloc was able to recruit voters from outside the usual community. Trying to fix that problem by changing the voting rules is very difficult, and will have all sorts of unintended consequences.

(3) If we choose to ignore (1) and (2), this is the thread to discuss how to fix the voting rules.

and here's the follow-up where he summarizes possible fixes to the Hugo voting mechanism
m (16 Apr 2015 10:00): But I mostly just wanted to point out that George R.R. Martin's blog IS SECRETLY A LIVEJOURNAL (maybe not so secretly)
m (16 Apr 2015 10:47): BLAST FROM THE PAST

Current Location: Santa Fe
Current Mood: depressed depressed
m (17 Oct 2016 12:40): holy shit I think Maine is going instant runoff