Thursday, November 06, 2008

I thought it might be interesting to keep track of FiveThirtyEight.com's projections during the last week of the campaign, in order to compare the progression to the eventual result. I'm sure that kind of historical data is available somewhere on the site, but I wasn't able to find it, so I'll make a note of it here. To be updated daily if I actually remember this between now and the election.

Popular VoteElectoral College
ObamaMcCainObamaMcCain
October 28, 200852.146.5348.2189.8
October 29, 200852.046.6344.1193.9
October 30, 200852.046.4346.5191.5
October 31, 200852.246.4349.7188.3
November 1, 200852.246.6344.0194.0
November 2, 200851.947.2332.9205.1
November 3, 200852.046.1346.5191.5
November 4, 200852.346.2348.6189.4
Actual52.945.7365173

Update: They say races traditionally tighten up in the last week or so. I'm chalking up the November 2 numbers to that phenomenon.

Second update: Exact numbers aren't in yet, but we're getting close, and things are looking awfully good for Nate Silver (if anything, he may have been on the low end). The jury's still out on Missouri (McCain up by about 6,000), North Carolina (Obama up by about 14,000), and Omaha, Nebraska (which has one electoral vote that could go either way). Once those are in, I'll update this table to reflect the actual election results.

Post-election update: Not too shabby.

2 comments:

Anonymous said...

It's like, a day old, and you're already two days behind. Confidence inspiring, it is!

Mike said...

Or am I?