So it is hard to say what happens if Mitt Romney, the establishment candidate, gets the nomination. It is a convenient aphorism that "Democrats fall in love, Republicans fall in line." The 1/3 of the country that is committed Republican will vote for whoever is the nominee, just as most disappointed liberals and progressives will fall in line and vote for Obama. But what this says for Congressional elections or for the middle of the electorate is damn hard to say.
Part of the problem is that it is possible to project such a wide range of divergent scenarios as *likely*. Instability in the Middle East, a perennial favorite to stir things up, has now reached epic proportions. I do not find it unreasonable to say that 2012 bears some remarkable and unfortunate resemblances to 1912 in terms of economic instability and international instability caused by the decline of the traditional great power (US/England), (EU/Austria-Hungarian Empire) and the rise of new powers testing their own influence and finding their position in the world. Northern Africa to Central Asia does a nice imitation of Central and Eastern Europe if we swap, say, Syria for Serbia. Meanwhile, China may or may not suffer economic or ecological catastrophe or its own ethnic uprisings and rural discontent. South and Central America remain in flux, with Brazil's growth making it a regional leader and a source of envy in the region.
All of this makes it rather hard to forecast the state of the economy, or even whether we will be at war or not, ten months from now. And if we are, who gets blamed? If Iran closes the Straits of Hormuz and the U.S. and EU respond militarily does the country rally around the President? If the nascent job recovery again peters out from any of a variety of causes, who gets blamed? The Republican House or the Democratic President?
Closer to home, several swing states that should have reverted back to the GOP have extremely unpopular Republican Governors. Maine, Florida, and OH, all of which went for Obama in '08 and could have flipped in 2012, have very unpopular Republican governors. Wisconsin, which arguably could have flipped (but I think was less likely than any of the above), has been radicalized by Scott Walker. Even if he survives recall, I think it will be hard for a Republican to carry the state.
This is not to say that Republicans are unlikely to gain states. Indiana is almost certain to flip back, as is VA. But there are so many wild cards that the simplistic formula and correlations (as well as the obsession with poll results) really don't capture things very well.
My final thought is that whenever voters go with "electability," they generally back the wrong candidate. That was the Dem experience in 2004 with Kerry, and the R experience with McCain in 2008. By contrast, Dems rejected the conventional wisdom on electability in 2008 -- although in fairness there were competing theories on electability in 2008. (Hilary supporters argued that America still wasn't ready for a black President, whereas Obama supporters argued that Hilary had such strong negatives that she couldn't win.)
The problem with "electability" is that it requires voters to set aside their own feelings on the basis of what they think other people will think in a general election months in the future. The problem with this is that people are generally bad at predicting what other people will think and feel and are lousy at predicting the future. As a result, voters in primaries who focus on electability either vote based on regurgitated popular wisdom of the moment, or on an assumption that other people won't respond to the same things that they respond to in a candidate. Neither is a particularly good predictor. However, since it is impossible to rerun the election after the fact with the other candidate, it is not something easily disproved.
To conclude, when Republican candidates start campaigning against heartless big business and greed, something funny is going on. This is especially so when you consider that Republicans are generally far better than Democrats at focus group testing and messaging. Sure the Republican base will turn out in droves to vote out Obama, just as the Democratic base turned out in droves to vote out Bush. What tipped the balance for the critical number of voters was the connection many felt with Bush on a personal values level. While Democrats scoffed at the articulation of this as "Bush is the guy I would have a beer with," the statement unpacks to "Bush is the guy I think is most like me and will look out for my interests." Voters still have generally high opinions of Obama on personal qualities, even if they think he is weak and ineffectual. What remains to be seen is if Obama can establish a personal connection with swing voters the way Bush did in '04.