Allright, some of you may remember how I was all excited about Instant Runoff Voting a while back. I have since heard about something called the Condorcet method. For a full rundown, check out the wikipedia article on Condorcet method.

Anyway, IRV still seems like a pretty good idea to me, as it solves some (but not all) of the problems with simple Most-Votes-Wins solutions like ours. However, there is a case where IRV will choose a canditate such that there is another candidate more people prefer. That's confusing and abstract, so if you car, check out the example from the wikipedia article.

I haven't looked at any of the implementations of the Condorcet method suggested in the article, but a major weakness I see, and I emphasize it is major, is how complicated the Condorcet method seems. I don't think the American populace is going to buy any voting system that can't be explained in a few sentences. To whit:
"Everybody votes for one person. The person with the most votes wins." -> "Hyup, sounds good"
"Everybody ranks all the canditates, putting a 1 by their favorite, 2 by their second favorite, and so forth. We tally everyone's first choices, and if no candidate gets a majority, then we eliminate the candidate with the least votes. Then ballots are re-calculated where if the highest ranked candidate is eliminated, we look at the next highest ranked candidate. After that, if there is still no majority, we eliminate the candidate with the fewest votes again and repeat." -> "Hmm... I guess that sounds allright..."

I won't even try with condorcet.... Anyway, our election system needs to change and IRV is still the best thing i've seen so far.