Submitted by Clay Shentrup, Mar 30, 2007 20:49

I don't know what definition of "majority" Rob Richie has in mind, but in my previous example with three candidates in an IRV election, the candidate who is preferred to both of his rivals by a huge majority loses. If either of those rivals had dropped out, however, he would have won. So the claims that IRV elects majority winners, and eliminates spoilers are bogus.

In this form of IRV, if no candidate wins an initial majority of first choices, the top two advance, and the ballots cast first for the eliminated candidates are added to the runoff candidate who is ranked next on each ballot .The more popular runoff candidate will ALWAYS win. Period.

This paragraph contains more of Rob Richie's typical misinformation. First of all, an instant top-two runoff is not a "form of IRV". IRV is the use of the Single Transferrable Vote system in a single-seat election; so you successively remove the candidate with the fewest first-place votes, until a "winner" remains. A top-two runoff, on the other hand, means a second contest between the top two vote getters in the main election, if no one earned a majority. This is currently done in at least 27 countries, using a second election when a runoff is required. The idea of an instant top-two runoff is that the voters simply list their ordered preferences, and then we look only at the top-two candidates, and compare them head-to-head - no second election is required. Perhaps Rob Richie should try to sell this method to third parties, instead of IRV, since we know that IRV leads to two-party duopoly.

But even then, it should be noted that an instant top-two runoff is not the same thing as the traditional delayed form in practice, due to voter strategy. Here's an example.

Finally, Richie's statement that "the more popular runoff candidate will always win" completely misses the point. The problem is that the most popular candidate often won't make it to the runoff. In my original runoff scenario, Edwards would win with a huge majority to either McCain or Obama, yet when all three run, Edwards loses??

Let's watch that in slow-mo. In this scenario, if Edwards and Obama were running head-to-head, IRV would say that Edwards was the better candidate. If Edwards were running against McCain, IRV would still say that Edwards was the better candidate. But when all three run together, IRV suddenly "changes its mind" and decides that Obama is now better than Edwards. IRV contradicts itself, and Richie's contrived notion of "core support" is a rather poor excuse for this madness.

Furthermore, McCain is a spoiler in this example, despite Richie's insistence that IRV will eliminate spoilers. It will certainly reduce them, which is a benefit over plurality voting. But Range Voting completely eliminates spoilers, and is a much greater improvement over IRV than IRV is over plurality. Why not go for the gold?

Finally, it is ironic for Richie to continually bring up the notion of tactical voting when talking about Range Voting, when Range Voting is known to be much more resistant to tactical voting than IRV. In fact it performs substantially better under a 50% strategic electorate than IRV does with a 100% honest electorate. The Olympics make an even greater case for the superiority of Range Voting: a method such as IRV would perform about as poorly under completely honest judging as Range Voting would perform under completely strategic judges who were willing to give a 0 or a 10.0 to any competitor as needed. Obviously the Olympics can't afford to have a bad voting method, so they use Range Voting - yet Rob Richie erroneously tries to use this as an argument against Range Voting. Unfortunately Rob Richie can't claim ignorance here, because we've explained this to him many times before.

While IRV does offer some important advantages over our embarrassingly poor plurality system, one must wonder why it is that Rob Richie (head of FairVote) is willing to use deceptive and misleading tactics to evangelize for it.

Note: Comments are screened, and in some cases edited, before posting. We reserve the right to reject anything we find objectionable.