knb comments on Political impasse as result of different odds ratios - Less Wrong

5 Post author: calcsam 19 November 2011 08:01PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (10)

You are viewing a single comment's thread.

Comment author: knb 20 November 2011 01:47:58AM *  1 point [-]

I doubt the professional politicians who are making big decisions actually have such wildly disparate predictions. Rather it seems that political amateurs have lousy predictions about who is likely to win--not surprising since the stakes for amateurs are pretty low. Intrade has Obama's chances at 51% for reelection. I would be surprised if the GOP leaders in Washington are banking on winning the trifecta.

*Edited to fix a typo (of should have been if -- in the last sentence).

Comment author: gwern 20 November 2011 02:15:11AM 0 points [-]

What makes you think the professional politicians are any better than say, Nate Silver or the pollsters, all of whom go into the Intrade stew? Politicians routinely make epic mistakes ranging from their personal life to grand strategy (in fact, one such example - Newt Gingrich and the government shutdown - has multiple parallels in modern American politics, from Gingrich himself being back to the recent near government shutdown).

Comment author: knb 20 November 2011 08:32:19PM 1 point [-]

You misunderstood what I was saying. I was referring to this:

In other words, the Republicans believe they can achieve complete victory so that they can enact their whole agenda, while Democrats believe they can block this victory. On a key component — Obama's re-election — Republicans believe they will defeat Obama with 1:6 odds, while Democrats believe this event has only 3:2 odds.** With a 9-fold difference in this key perception, it seems highly unlikely the two will reach a compromise.

The post conflates the outcome of an opinion poll of the general population with the beliefs of the Democrats and Republicans who will actually make big budget decisions. There is good reason to expect the professionals have more accurate expectations than a bunch of poll-taking amateurs. I wasn't saying that the politicians are more accurate than Intrade, I was saying they're more accurate than poll-takers. Actually, I wouldn't be surprised if a lot of politics professionals use Intrade to estimate probabilities.