The really big development this Presidential cycle has been the popular rise of such statistically-oriented political analysts as Nate Silver. They mix math, polls, GDP, magic sauce, wizardry, SCSI-termination mojo into a prediction about the election that is SCIENTIFIC.
This is not a post about those people, or not entirely about them. Instead, this is a post about aggregating all their (and other pundits) predictions and seeing if the magic of crowd-sourcing and the wisdom of the crowds works in political prognosticating.
So, I gathered as many predictions as I could: Nate Silver’s; Sam Wang’s, Drew Linzer’s, John Scalzi’s, Michael Barone’s, Karl Rove’s, Dick Morris’s, Bickers/Berry, Real Clear Politics, DeSart and Holbrook, Jackman, Election Projection, Lanny Davis, and, finally, the prediction of the students of Cornell in Washington’s Policy and Politics class. [UPDATE: added George Will, Cokie Roberts, Matthew Dowd, and Ron Brownstein! Average remains the same.]
[UPDATE II: And now, Larry Sabato. This changes the average to Obama 275, Romney 263, and I added a map in below] The average of all these pundits--statistical, political, partisan, better, worse--is Obama, 274, Romney 264. Which is actually possible, with this map:
Or, by splitting Maine, this one:
[For Obama 275, Romney 263, this map works:
I’m sure I could try to create a model that weighed the pundits’ credibility (Dick Morris: 0), but I don’t want the VRWC to declare war on me.
