We’re still a long way from actual primary voting, but it’s never too early for polling! In any case, Not Paul Begala at Blog, P.I directs our attention to the latest round of polling regarding the upcoming Democratic primary. He also puts up a handy chart (which I have shamelessly hot-linked):

As you can see from each and every poll listed, the vast majority of Dems prefer “not Hillary” in the primary. Just for the sake of argument, let’s assume the undecideds break exactly evenly for the other four options listed. Breaking things down into either “Hillary” or “Not Hillary”. We get this:

I have no idea why Mellman’s numbers don’t add up to 100%, but it is what it is. In any case, each of the polls is clearly outside the typical margin of error for these sorts of things (usually it’s around + 4%). NPB’s contention is that Mellman’s numbers are the best, and given that he’s a real political professional, I’m inclined to believe him. Given that, there’s a clear 9pt lead for camp Not Hillary, and yet given the breakdown that NPB provided and the way that plurality (majority rules, first past the post) voting works it’s highly likely that camp Hillary will actually get the nomination.
To my mind, this is the single greatest flaw in our current voting system, and while there are flaws with other systems, it is my firm opinion that this particular issue trumps a lot of them. By voting for your candidate, you can hurt the chances of candidates you prefer, thusly helping a candidate you prefer least to win. The spoiler effect, essentially, causes a violation of the indepencence of irrelevant alternatives. Removing (or adding) a candidate from (or to) the choice set, changes the ordering of the other choices, and I think that’s a big problem.
Other voting systems do a better job of this, but are subject to other issues and, frustratingly, can be extremely complicated to set up and administer. The easiest and most practical system is probably Approval Voting, but it probably wouldn’t be to the benefit of the two major political parties, so I’m not holding my breath for its implementation.

