Feeds: 
symbol identifies content for paid subscribers only.
TIGHTENING.
I've been finding it vaguely odd to watch liberals grow agitated over mild tightening in the final days of polling. These are the same folks who've been loving Nate Silver's FiveThirtyEight.com, but one of Silver's innovations was to examine past elections and build his algorithm such that it predicts that the race will tighten in the days before the election. It tightened for Kerry, for Gore, and for Dole. It's a common feature of elections. That's part of why his model has a sounder claim to accuracy than competing methods of prediction. No one who's been reading his site should be surprised by this.
COMMENTS (6)
Nate's work has been very interesting. But people do get frazzled, even when they "think" a desired outcome is probable.
I do probabilistic forecasting of tennis matches. Your guy can be up a set and a break, and you still start sweating when the opponent gets to 0-30. It's been interesting seeing Obama do what the great players do: play the right shots regardless of the score.
Posted by: Andrew | November 2, 2008 11:11 PM
Forecasting tennis? That must be, umm, interesting. I really can't imagine how you would approach that. But elections are much easier, fortunately.
Nate's doing a great job and I'm not worried about any tightening. That always happens. I knew that before Nate was born. But I must bow to his superior statistical analysis. He's good, and I hate doing that kind of math. I'm an analytic geometry guy.
Posted by: fostert | November 2, 2008 11:19 PM
On the one hand, Nate's model is, like all such models, a theory of elections, and we don't know how accurate the theory is until the see the reality. On the other hand, before conventions he posted a graph of how the two bounces might overlap and play out, and he pretty much nailed that.
Posted by: DonBoy | November 2, 2008 11:21 PM
I'm not sure why anyone thought that John McCain would only get ~41 percent of the national popular vote. I guess it's possible that we could see a blowout of epic proportions like that, but it's not very likely.
Even what "tightening" we've seen hasn't been such a big deal. Holding McCain at 45-46% of the vote would be pretty impressive.
Posted by: rufustfyrfly | November 2, 2008 11:33 PM
So long as the "tightening" is undecideds going to McCain, and not a drop in Obama's top line number, it is not too surprising, and as long as Obama's at 49+, it shouldn't matter. Those people were going to come home for McCain sometime -- he wasn't going to finish at 42%.
Posted by: Bobby | November 2, 2008 11:45 PM
Uhm, why would we liberals get agitated?
Mr. Ezra, it might have a little to do with the fact that we got our asses handed to us in '80, '84, '88, '00, and '04. Note that the final Gallup poll in 1980 (October 26th) had Carter up by 6-7 points. Note that Dukakis had a ten point lead in late August of '88. Note that the MSM did a hit job on Gore in '00. Note how the Republicans stole that one. Note that Kerry wiped the floor with Bush--making him appear like a drunken mongoloid--in all three debates in '04. And we STILL lost.
Note that it should not surprise you that we liberals are "agitated."
Robert
Posted by: Robert Murphy | November 3, 2008 12:59 AM