What's the matter with Ohio?

Plenty.

The New Yorker's George Packer dives into the notion that hard hit whites will turn to the Democrats.

...four sociologists at the University of Arizona, led by Lane Kenworthy, relesed a...study [that] followed the voting behavior of the forty-five per cent of white Americans who identify themselves as working class. Mining electoral data from the General Social Survey, they found that the decline in white working-class support for Democrats occurred in one period—from the mid-seventies until the early nineties, with a brief lull in the early eighties—and has remained well below fifty per cent ever since. But they concluded that social issues like abortion, guns, religion, and even (outside the South) race had little to do with the shift. Instead, according to their data, it was based on a judgment that—during years in which industrial jobs went overseas, unions practically vanished, and working-class incomes stagnated—the Democratic Party was no longer much help to them. “Beginning in the mid-to-late 1970s, there was increasing reason for working-class whites to question whether the Democrats were still better than the Republicans at promoting their material well-being,” the study’s authors write. Working-class whites, their fortunes falling, began to embrace the anti-government, low-tax rhetoric of the conservative movement. During Clinton’s Presidency, the downward economic spiral of these Americans was arrested, but by then their identification with the Democrats had eroded. Having earlier moved to the right for economic reasons, the Arizona study concluded, the working class stayed there because of the rising prominence of social issues—Thomas Frank’s [author of What's the matter with Kansas?] argument. But the Democrats fundamentally lost the white working class because these voters no longer believed the Party’s central tenet—that government could restore a sense of economic security.

Ohio's trending blue these days - but I'd Packer is pointing out a host of reasons to not be complacent. Among them:

“Poorer, less well-educated white people refuse surveys more often than affluent, better-educated whites,” Kohut wrote. “Polls generally adjust their samples for this tendency. But here’s the problem: these whites who do not respond to surveys tend to have more unfavorable views of blacks than respondents who do the interviews.” This statistical glitch is different from the Bradley Effect...

At least on the Pollster chart, a state that leans blue could very well be a toss up.

My money's still on Obama - McCain has a sick amount of electoral ground to make up - but progressives should not expect to coast across the finish line.

(H/t to CJR)