McCain Camp Thinks PA and WI Stronger Than MI. Really?

The McCain withdrawal from Michigan was done ostensibly to redirect resources to more winnable states like Pennsylvania and Wisconsin. But is Michigan really the weakest of these three states?

Michigan had been the symbolic heartland of McCain's blue state appeal. He had won a surprise victory there in 2000, and despite underperforming against a virtual favorite son in the primaries, HRC had handed him an opportunity on a silver platter with the long drawn out battle over seating Michigan's delegation to the DNC. It was well understood that uncommitted was a vote for Obama in the primary, and uncommitted lost by 15 points.

How did Obama do in next-door Wisconsin? He won it by 18. He lost Pennsylvania by a less-daunting 10.

Even with the polls as they stand today, McCain is doing better in Michigan than he deserves to be. McCain should be losing in Michigan by 11.7% based on its 2004 margin and the swing since '04 in the RCP average. Instead, state-specific factors are keeping McCain to within 7 in the RCP average and 7.5 in FiveThirtyEight's projection.

McCain is currently losing PA by 9.2% in RCP and by 7 in the FiveThirtyEight projection. There, he should be losing by 10.8%. In Wisconsin, RCP has him down by a more promising 5 (due to a poll released 9/17 that had him down 1) but 538 projects a more sobering 9.6%. I tend to think that geography (cities within commuting distance of Chicago), a more progressive Democratic electorate, and same day voter fraud registration should make Wisconsin a McCain-underperform state.

I won't lie. All three states are extremely uphill at the moment. But I could make a case for dropping Minnesota, Wisconsin, and Iowa long before I'd make the case for Michigan.

I have put together a spreadsheet showing where McCain should stand in all swing states given national polling and the 2004 baseline. Michigan is one of the two states where McCain is performing strongest relative to 2004 if you compare the polling averages to the expected outcome. The other is New Hampshire.

And the other states? McCain is outperforming (in most cases, losing by less than he should be) in Nevada (+3.7%), Minnesota (+3.7%), Ohio (3.5%), Pennsylvania (2.7%), Missouri (2.1%), Wisconsin (1.4%), and Florida (0.6%). And he is underperforming in Colorado (-0.7%), New Mexico (-1.0%), Iowa (-3.3%), Virginia (-3.6%), North Carolina (-4.6%), and Indiana (-11.3%).

Interestingly, all of McCain's underperforming states are red states which were thought to have a decent cushion of Republican voters. McCain can't afford to give up on any of them, except possibly Iowa. It's like they're being magnetically drawn to swing state status by giving Obama just enough support to be competitive. 

And so are the blue industrial heartland swing regions, who refuse to let their exalted status die hard by giving McCain 3 or 4 extra points, and enabling the Maverick to compete there. 

I've made my position on this clear. Regardless of what the national polling says, McCain should target the states he'll need to forge a winning coalition to get to 270. That means aggressively competing in states that could represent Electoral Votes 300 or 310. By any measure, Michigan should be in this grouping, though it probably sits right on the borderline.

Also, Michigan carries more symbolic import than a Wisconsin or Minnesota, which haven't gone Republican since 1984 and 1972 respectively. It is home to "bitter" America and doesn't have much of a progressive streak. The consquences of pulling out could be to demoralize the troops nationally, not just in one state.