There's all sorts of speculation and back of the envelope math being done right now on the Massachusetts Senate special election and whether Republican Scott Brown might be able to pull off a surprise there. Dave Weigel reports that Rasmussen is going to be polling there tonight so that should give us a better idea but I thought it would be worth taking a look at what would happen if things played out in Masschusetts similarly to Virginia, which is a sort of Democratic worst case scenario.
The Virginia exit poll showed the folks who turned out there voted for John McCain by an eight point margin, in contrast to the actual six point victory in the state for Barack Obama. A similar 14 point drop on the Obama margin in Massachusetts would mean the people who come out for the Senate election voted for him by a 56-44 margin.
In Virginia Creigh Deeds won 88% of the Obama vote and Bob McDonnell won 95% of the McCain vote. Give those percentages to Martha Coakley and Scott Brown and you have Coakley at 51.5% and Brown at 48.5%. That finding incidentally is similar to what Sean Trende of Real Clear Politics came up with last week using some different metrics.
So how does Brown win? One path would be for Democratic turnout in Massachusetts to be even worse relative to the Republicans than it was in Virginia- that seems unlikely to me as what happened there really does seem about rock bottom. A more plausible (though still highly unlikely) scenario would be for Brown to pick off a higher percentage of Obama voters than McDonnell did in Virginia. Democratic turnout there may have been awful but the ones who did turn out voted for Deeds pretty much across the board- Brown would have to do a better job picking off Obama voters than McDonnell did to pull the shocker.
If Rasmussen shows a relatively close race we'll poll it this weekend.