Comments on: Predictable Instapundit

By: john b

john b — Tue, 12 Oct 2004 23:13:32 +0000

Steve – I thought WTA was ‘winner of the popular vote’ (which is easy to analyse), not ‘winner of the electoral college’ (which is hard to analyse)? I’m pretty sure that WTA contracts last time paid out on Gore not Bush.

By: Giles

Giles — Tue, 12 Oct 2004 22:17:11 +0000

uncle I spent alot of my student years doing polling and believe me waht people tell is important is what they think every one else thinks is important. But at the end of the day they make their mind up generally on very individual issues. In Australia Latham and Labor wanted to make Iraq and issue and the polls in say march would have told them it was – but in reality it wasnt – it was decided on how happy each average John was. As for middle class elite, that refers to what ever the perjorative term you have for the chaterati 0 Islington dinner party set, north shore chardonay set or whatever wherever you live. Basically teachers, lawyers, acdemics, accountants, people in the arts etc.

By: Steve Carr

Steve Carr — Tue, 12 Oct 2004 21:21:29 +0000

The IEM WTA market has the probabilities—assuming for the moment you can read these off from prices—at around 55% for Bush. (The IEM is a little strange now because the WTA market has been broken into four smaller markets, but that’s what the numbers sum to.) The vote-share market, though, which I’ve always thought was more important because you can know—without debate—how accurate the market’s prediction was, has Kerry getting 49.2% of the two-party vote and Bush 50.5%, which is pretty close to 50-50.

By: John Quiggin

John Quiggin — Tue, 12 Oct 2004 20:15:34 +0000

Steve, you’re quite right that this is a win for the betting markets. I noted this on my blog. I’m 50-50 again on the US election, so if someone can supply me with the current IEM odds, I’ll estimate the posteriors on the EMH following a Bush win in November.

By: Uncle Kvetch

Uncle Kvetch — Tue, 12 Oct 2004 19:13:39 +0000

Just to clarify: in that Newsweek poll, the Iraq war was the third most common response to the question “Which ONE of the following issues will be MOST important in determining your vote for president this year?”

By: Uncle Kvetch

Uncle Kvetch — Tue, 12 Oct 2004 19:11:15 +0000

At the end of the day, outside of the middle class elite, most of the electorate are not going to decide their vote on the basis of a small war in the Middle East. I don’t know what your source is for this, but in the latest Newsweek poll the “situation” in Iraq was third, after “Terrorism and homeland security” (which, as we know, is virtually [and distressingly] indistinguishable from the war in Iraq for much of the American public, and the economy. And as long as I’m here: would you please clarify for me just what is a “middle class elite”? We didn’t cover that one in my intro to sociology.

By: Randy Paul

Randy Paul — Tue, 12 Oct 2004 16:40:36 +0000

Post hoc ergo propter hoc: Thy name is Glenn Reynolds.

By: Jason

Jason — Tue, 12 Oct 2004 16:20:30 +0000

I’m with Steve, the Bayesian inference is by far the most interesting tidbit here (although I didn’t agree with JQ’s reasoning completely). Now, what would a Bush victory do to your prior? Admittedly on IEM at the moment, it is only a 55% predicted Bush win.

By: Giles

Giles — Tue, 12 Oct 2004 16:10:18 +0000

I would also add that you posted tha you thought it significant that Latham won the debates on Iraq. The problem wasn’t that some people didnt want Iraq to be an issue, but rather that the majority of the electorate wasn’t interested. And I think that what Reynolds is alluding to is that this may also be the case in the US. At the end of the day, outside of the middle class elite, most of the electorate are not going to decide their vote on the basis of a small war in the Middle East.

By: Steve Carr

Steve Carr — Tue, 12 Oct 2004 16:10:18 +0000

John, you’re absolutely right that the election was no referendum on the war. But you’ll remember that you made it an impromptu natural experiment of the virtue of the EMH. I think you said that since Howard was heavily favored in the betting odds on the day the election was announced, if he won, we would have to revise upward—to 0.55, I believe—the probability that the EMH was right. That’s a lot more interesting than Glenn Reynolds’ bloviations.

By: praktike

praktike — Tue, 12 Oct 2004 14:22:17 +0000

Ignore the Instapundit. If you take away a germ’s medium, it will die.

By: dave heasman

dave heasman — Tue, 12 Oct 2004 14:03:20 +0000

I thought that Australia has had low unemployment and fairly constant economic growth > 2% p.a. ? That’s usually enough to get a governing party reelected.

By: G. Svenson

G. Svenson — Tue, 12 Oct 2004 14:02:43 +0000

“Admittedly, Reynolds isn’t “the international press”,” Now there’s a quote ripe with potential..

By: raj

raj — Tue, 12 Oct 2004 13:38:12 +0000

Just wondering, does Reynolds pay you people to hype his blog? He’s an otherwise obscure law perfessor from Tennessee.

By: Snowball

Snowball — Tue, 12 Oct 2004 13:19:28 +0000

“I don’t know much about Australian politics, but I agree with Instapundit that the press would have made the election into an Iraq referendum if Latham had won. ” Well, you’re half right. If you did know much about Australian politics, you’d know that, if Latham had won, the press would have “made” the election about a government that had past its expiry date, a reaction against yet another cynical, fear mongering and deceitful campaign and an appreciation of a party presenting a policy platform that paid some attention to a raft of long neglected social concerns. Sadly, that’s not the story the press have a chance to tell.