When a Democrat loses a federal race in Massachusetts, the default assumption ought to be that several factors are to blame.
Clearly the national environment has gotten worse for the Democrats since Barack Obama's inauguration one year ago. This has been obvious from Congressional generic ballot polling, Presidential approval polling, early polling of 2010 senate races, the number of Democratic retirements, the outcomes of New Jersey and Virginia, the tenor of the political discourse in the country, and so forth. But perhaps it is somewhat more bad than we had previously realized.
Clearly also, the quality of the candidates and the campaign matters a lot, especially in open seat races. Although it might seem strange to have a Republican Senator from Massachusetts, it is not dramatically more strange than having a Democratic Senator from Alaska or Nebraska, or a Republican Representative from New Orleans, all of which our Congress already had before tonight. Martha Coakley, needless to say, was not a good candidate and did not run a good campaign.
Finally, there is a third category: contingencies specific to Massachusetts, but not specific to Coakley. This was a state in which Democrats had twice changed the rules governing Senate succession, first in 2004 to prevent then-governor Mitt Romney from appointing a Republican to take John Kerry's seat (should he have been elected President), and then again last year to allow Deval Patrick to appoint an interim appointee. Moreover, because it was a special election, the time frame of the campaign was dramatically compressed, making it harder to define the Republican opponent or to recover from any initial missteps in the campaign. Lastly, Massachusetts is unusual in that it already has universal health care and the Democratic health care plan would not do it much good, which allowed the Republican to promise to oppose it without looking like a typical partisan hack.
Overall, we have a 31-point swing in the vote to explain: from Barack Obama's roughly 26-point victory in November 2008, to Martha Coakley's roughly 5-point loss today.
At a bare minimum, 10 of those points must be assigned to the national environment. Generic ballot polling suggests that the Democrats' position has worsened by a net of 10 points since November 2008, from winning the House popular vote by 10 points in 2008 to being dead even with Republicans today.
Also at a bare minimum, 11 points of blame should be assigned to Coakley. That represents the difference between the 58 percent of vote that she received at her high-water mark in the polls to the 47 percent she received on Election Day. A fairly large number of voters, it appears, actually turned away from Coakley; it was not just a matter of undecided ones turning toward Brown.
That leaves us with 10 more points of blame to assign; let's just dole those out as evenly as possible, giving 3 more points to Coakley, 3 more points to the national environment, and 4 to Massachusetts-specific special contingencies -- it gets the extra point because it hadn't received any yet.
That would make the final score: national environment 13, Coakley 14, special circumstances 4.
If you follow through on the math, this would suggest that Coakley would have won by about 8 points, rather than losing by 5, had the national environment not deteriorated so significantly for Democrats. It suggests that the Democrats would have won by 9 points, rather than losing by 5, had the candidate been someone other than Coakley. And it suggests that the race would have been a 1-point loss (that is, basically too close to call), rather than a 5-point loss, even if Coakley had run such a bad campaign and even if the national environment had deteriorated as much as it has, but had there not been the unusual circumstances associated with this particular election.
Obviously, this is a rather imprecise and unsophisticated exercise. But each of those implications feels about right to me. Maybe you'd do the math a little differently. But don't be sparing with your blame; there's plenty of it to go around.