Thoughts from
San Diego on Law, Politics, and Culture

 
Don't fret, little Bushies
By Tom Smith

Yes, the polls show Kerry has closed the gap, and that's nothing to be thrilled about. But the polls are not very good indicators of who is going to win. We don't have very good indicators of who will win, but the best we have are futures markets, odd as that may seem. When Kerry gets within 3 or 4 points of Bush on the betting markets, then it will be time for the serious questions, like do I want the 3 month MRE variety value pack, or should I just choose singles that look like they might taste good after 6 months in the bush, waiting for the isotopes to decay. (Yes, I exagerate somewhat, but a Kerry victory would definitely increase my national security anxiety level.)

Trust me. I've looked at this. I'm too lazy to link to my sources, but it is just remarkable how good a job even completely play markets do processing information from polls and other sources respecting future events. Of course, this does not mean the market can predict the future. It just does a better job of doing so in light of currently available information. The markets obviously discount polling numbers in various ways. It's hard to say what the market is "thinking." It's important to understand the market is smarter, a lot smarter, than any of the individuals who trade in it, most of the time. So it is not just a matter of what the smart money is thinking. Having said that, it may be the market thinks the polls are inaccurate, that swing voters tend to be impressionable but change their minds easily, that Bush will do better in a Townhall format, that there's a good chance of an October terrorist surprize, who knows.

So, here's Bush on tradesports.com, which I like better than IEM because its graphics are better. It's now a 60-40 race for Bush, down from roughly 67-33. A big drop for Bush, reflecting his poor performance in the debate. But 60-40 is still a big spread. Not time to panic yet. Later, maybe. If any Kerry supporter will give you even odds on Kerry winning, take it! If somebody is mouthing off about how Kerry is going to win, ask them what he thinks Kerry's odds are of winning. If he says 70-30, ask him for a bet on those odds. You give 30 bucks to honest intermediary, he gives seventy, and the winner gets the pot. That's fair, isn't it? It's amazing how few people would dream of putting their money where their mouth is, in that cliched, but so very salutary phrase. It does have a way of shutting people up, and when it doesn't, you can make some dough.

I suspect the polls are biased, not necessarily in favor of the Dems, but in favor of being newsworthy. Newsweek's numbers seem awfully volatile to me. I didn't believe them when they said Bush was 14 points ahead or whatever it was, and I don't believe them now, with Kerry 2 points ahead (or whatever). Zogby just seems pro-Democratic. Gallup somewhat Republican, perhaps. Rassmussen seems to use bigger samples, which can't hoit. State polls strike me as potentially more useful. You can waste hours looking at this stuff on realclearpolitics.

posted by thomas at 10/04/2004 10:43:00 AM