If Romney did get a bounce though, it sort of makes sense that he'd get it in a mountain state like Colorado. I think it will be easier for Romney to gain ground in Colorado than even in Ohio.
Colorado just seems like an easier state than Ohio for Romney to pick up. Few people realize that 2011 may have been more important to these two states' political futures than 2010. In 2011 Colorado, including all metro Denver counties (excluding Boulder), voted overwhelmingly against a statewide tax hike for schools (the margin was a little less than 2:1). By almost the same margin, Ohioans voted to repeal Gov. Kasich's new union law. Those two off-year elections help focus voters' attention and toss out a few tea leaves for 2012. Colorado is one of the most fiscally conservative, anti-tax states in the country, and Ohio is not especially conservative on those fiscal issues. Obama is very unpopular in both states, but in an election based on core issues like taxes, unions, budgets, etc., Colorado seems like the most likely to split right while Ohio seems more likely to split left.
That's an analysis that many haven't yet picked up on--in fact, the White House has only recently indicated a shift in strategy that will place Ohio and the midwest ahead of the "New Democratic" states in terms of electoral priority.
The bottom line is this: the Purple Poll is probably where the election is headed, despite Obama's problems with blue-collar white voters. Colorado will end up moving more safely in Romney's direction, and Ohio will emerge, as it is usually does, as the flashpoint of the 2012 campaign. If Romney does drop Ohio, something that looks likely at this point, he will need to find a way to pick off enough suburban Philly voters to win Pennsylvania. If there's a candidate to do it, it would probably be him, as long as doesn't lose too many "Pennsyltucky" voters in mid-state PA or the areas around Pittsburgh.