Nailbiter or Blowout?

So this election is still more than 5 months away and lots lots lots can and will happen before then, so While I am peripherally following the polls and election news, I'm not obsessing over them since their current predictive power is so low. Also, I really hate the "horse-race" way the US media covers elections.

BUT.

Sometime's it's hard to resist, peeking just a little. It's like when I'm playing a FPS and I take a break to see what my kills/deaths ratio is before the round is over. (It's usually not very good.)

The national, bottom-line, "who's gonna win" polls don't look very good right now.


Granted, you can say that given the recession and sluggish recovery, this is pretty good. Carter and Bush I had bad economies and they fared much worse. Being tiedup is not terrible -- it's better than being losing, right? Still, in an incredibly polarized country, being a president who was hugely popular at inauguration and has now sunk to even odds for re-election is a worrisome thing.

But here's the odd thing. When you break down the polling into state by state data, it looks much better for Obama:


According to Kos, this map, were it to hold, would yield a 329-209 electoral college blowout. And that's losing Florida. Most of the swing states aren't even close; if nothing else the margins are well beyond the error bars.

Do how do we reconcile this apparently conflicting data? Is Obama just so deeply unpopular in the Bible belt that it skews the national numbers? That would make sense mathematically, but still is a little surprising.

So, with the usual caveats that it's too early to say anything meaningful, how shall we place our bets? For me, I'll lay my chips on a close election, <4% between O & R in popular vote but with a solid EC victory. Dunno about 120 points, but I'm gonna predict the EC won't be closer than 50 points.

What do you all think?