Brad, it's desperately important that we inform the folks who support Instant Runoff Voting (IRV) of the truth about IRV and convince them to choose a simpler and better strategy.

Here's the truth: IRV is not an improvement. It is worse in many ways than the current system where each voter chooses one candidate.

1. The ostensible motivation for IRV is to prevent the "spoiler" problem so that voters aren't forced to hide their support for smaller parties, with the eventual goal of breaking the two-party duopoly and allowing more parties to be represented in government. IRV does not accomplish this; in fact IRV prevents third parties from winning.

It may appear to you that IRV lets you safely vote for a third party and rank a major party in second place to prevent your ballot from being wasted. Unfortunately, this is only true if the third party has no chance of winning. Once the third party gains non-negligible support, IRV puts you in the same position of having to choose the lesser of two evils. That is to say, IRV only provides the *illusion* of letting you vote for a third party; it only allows you to do so if your third-party vote is purely symbolic.

2. IRV additionally introduces many disadvantages. IRV is complicated and hard to explain to voters, which is a showstopper since voters have to understand the system in order to have any faith in the results. IRV is expensive and error-prone to implement, because the vast number of permutations on the ballot prevent precincts from totalling up their results as a small set of numbers; instead all the data for each individual ballot must be collected in a central location for processing.

And, perhaps worst of all, IRV exhibits extremely pathological behaviour whenever an election becomes close. In such situations, voting FOR a candidate can actually cause the candidate to LOSE. Such crazy behaviour is simply unacceptable.

(The basic reason this happens is that even though you are marking many candidates on your ballot, only one of your entries is going to count; the entries above it will be eliminated and the entries below it will be ignored. The problem is that you can't predict which! Which entry counts depends in a complicated way on the rest of the ballots.)

***

IRV is the wrong path. Fighting for IRV not only wastes the energy of the people who want voting reform; it risks wasting the patience of the populace who will no longer be open to other voting reform ideas when IRV is adopted and it fails.

Condorcet, as you mentioned, is one possibility. It behaves much better than IRV and is probably the most accurate method for choosing the best candidate. But it is still very complicated to explain and implement, probably more complicated than IRV, and i consider that fatal.

The best practical answer is approval voting.

Approval voting is extremely simple to explain and implement. Here is the explanation in only 15 words:

"Vote for as many candidates as you like. The candidate with the most votes wins."

In approval voting, every vote counts. Period.

Approval voting even works with existing ballots.

And it's trivial to see that there can never be a spoiler in an approval voting election. Your decisions about each of the candidates are completely independent.

Please, please consider promoting approval voting instead of IRV. I need your help to make this idea take hold.