This is the first presidential election (going back to the 1960s) I've ever seen where I did not know who would win two weeks out. (In 2000 the guy I "knew" would win two weeks out subsequently had a DWI he lied about revealed, lost the popular vote, and then got appointed by people his father had put on the Supreme Court, but he ended up being the guy sworn in.)
This year is different. I don't know.
But given all the available evidence, I feel pretty good about it. Team Obama feels they have 281 EV lined up pretty securely, and I agree. And of that 281 (which doesn't include VA, FL or CO), if New Hampshire and Iowa went south it would still be 271.
We are in good shape and nobody on our side would want to trade places with Romney.
There will be some late big-money smear ads and Romney's paid turn-out operation will be good... but our organization is better.
Being a poker player gives you a real-word sense of odds. Pocket AA versus pocket KK wins 80% of the time. It loses 20% of the time. Pocket AA is where you want to be, but you do lose a reliable 20% to KK, over time.
But Nate Silver odds are probably not quite right because the individual states don't really behave randomly/indeoendently. If Obama wins Virginia that datum contains evidence of how he will do in Nevada because the datum suggests something about where Obama is nationally. So the computer odds models should probably contain some "stickiness." Some slight cluster effect in simulations where state outcomes change the odds. Say Virginia is 20% and Nevada is 80% (just hypothetical round numbers). That means that one in 25 simulations will feature winning VA while losing NV. But the odds are derived from imperfect evidence, so winning VA should bump NV up to 90% because it increases the likelihood that all the odds should be slightly more Obama-friendly. (From Romney's perspective, the same for losing Nevada, in the opposite direction.)
Meaning that it is hard to be only slightly ahead. If you are really ahead then that is probably worth 5% bonus points. (And the same for being behind... it is harder to overcome than it seems.)
So I think our odds might be better a little than they seem, and what evidence we have today suggests that we might get a nice little surge of momentum going into the last two weeks.
So I'm feeling okay about it all. I'll say 75% today, and if we get a sustainable post-debate 2% bounce that would move it to 85%. A 3% bounce would come close to sealing the deal.