More Evidence of a Coleman Win in Minnesota Senate Race
Despite favoring Al Franken, stat geek Nate Silver of Five Thirty Eight.com crunched the numbers on the recount and finds only one scenario out of 8 where Franken might win, using various assumptions.
Now before you get too excited, this does NOT mean Coleman has a 7 in 8 chance of prevailing. Far from it.
First, all the models necessarily use assumptions, reasonable ones, but still the data still is subject to large errors compared to the Coleman lead.
As Silver notes, "Given the high degrees of uncertainty and ambiguity implied by the models, they would suggest that Franken has roughly speaking somewhere between a 25% chance and a 50% chance of overtaking Coleman depending on which model is selected."
Second, the model makes no assumption about the eventual outcome of the fight over absentee ballots. Silver writes that "If Franken is able to get such ballots counted -- and there is a strong chance that he will -- they will likely be worth a net of somewhere between 25 and 100 votes to him. In this eventuality, the race should probably be considered a toss-up."
Finally, this model does not project what the US Senate might do, as the final judge of elections. On this front, the news is not good as Franken's lawyer has been making a lot of noises about eventually going to the Senate. The Hill newspaper today reported that "Franken attorney Marc Elias made the case to reporters Monday that as many as 1,000 absentee ballots were improperly disqualified and that the Senate or the courts may need to step in to resolve the issue."
Personally, I don't think the Senate would reverse a final decision in Minnesota, though I admit it could attempt to do so. I think a more likely result, should Franken attempt such a maneuver, is that Barack Obama would say he thinks the Senate should respect the final decision of the State of Minnesota. It would avoid a bitter partisan dispute that would detract from his efforts to govern.
A footnote: As Sam Wang points out in his post "Ties, damned ties and statistics," the election was effectively tied. He takes two scenarios where a tie is assumed in his statistical model of running an election with about the same number of voters as the Minnesota Senate race, and crunches the numbers. In both cases, the model with a tie "assumed" would produce a larger margin of victory than any recounted number we've yet seen from Minnesota.
