Franken to Win Recount by 27 Votes

Nate Silver uses a combination of regression analysis and wild ass guesses drawn from limited information on challenged ballots to project that Al Franken will come out ahead by a mere 27 votes.  Kevin Drum, for one, is convinced.

While I’m less enthralled by Silver’s savant status than most (many came closer to the results by informed guessing) I agree that a narrow Franken win is the most likely outcome.  Even though Norm Coleman was ahead by 600 votes with 100 percent of the precincts in, every action since then has resulted in Franken picking up votes.  I have little doubt that he’ll manage to come out ahead in the end.

The amusing thing one finds in reading the comments thread at Silver’s place is that, in addition to confusing the ability to apply basic statistical analysis with “genius,” is the number of Democrats who think that, if Coleman does manage to win, it’ll somehow be a “stolen” election — just like 2000!  This, despite the fact that Coleman and Bush both won the initial count.

For reasons I don’t quite understand, Democrats seem to fare much better on recounts than Republicans.  Indeed, I can’t recall a statewide election that ended closely and required either a recount or going to the absentee ballots in which the Democrat didn’t come out ahead.

via Memeorandum

FILED UNDER: Campaign 2008, US Politics, , , , , ,
James Joyner
About James Joyner
James Joyner is a Security Studies professor at Marine Corps University's Command and Staff College and a nonresident senior fellow at the Scowcroft Center for Strategy and Security at the Atlantic Council. He's a former Army officer and Desert Storm vet. Views expressed here are his own. Follow James on Twitter @DrJJoyner.

Comments

  1. odograph says:

    I want Franken to win for the comedy (intentional and non) … and it’s not like he’s going to be my Senator.

  2. sam says:

    For reasons I don’t quite understand, Democrats seem to fare much better on recounts than Republicans.

    FWIW, here’s one explanation:

    Recounts favor Democrats
    Scott Rafferty tells me one more thing I didn’t know:

    Since 2000, seven out of seven peer-reviewed academic studies confirm that Democrats tend to make more mistakes than Republicans. The populations that are, in academic jargon, “vulnerable to error” are low-income, low-educated and minority, all of which disproportionately vote Democratic.

    So Scott Rafferty concludes this makes recounts favor Democrats, particularly in states with systems like Minnesota’s:

    The largest source of error is going to be human error — marks that the machine cannot be expected to recognize. The voter crosses out one oval and fills in another. The machine sees an overvote, but the hand canvass counts the vote correctly. When the voter circles the candidate, or uses a check mark, or doesn’t fill in the entire oval, the machine sees no vote at all, but again, the hand canvass registers the vote.

  3. James Joyner says:

    Sam: I think that’s a large part of it. Unfortunately, that leads to a human decision having to be made as to the voter’s intent. And those humans have preferred outcomes. It’s a process fraught with peril.

  4. LaurenceB says:

    I call “nutpicking”. Fifteen yard penalty.

  5. SavageView says:

    A posting like this from a site that trumpeted the “attack” on a McCain “volunteer” and decried the non-existent “proposal” to appoint Beau Biden to his father’s seat after Beau had rejected the idea.

    ‘Tis to laugh. But what else do the Bush dead-enders have to talk about?

  6. DL says:

    27? remarkable being as how that’s the same number as his IQ

  7. sam says:

    It’s a process fraught with peril.

    There is truth to that, James (but I suppose one could say that democracy itself is a process fraught with peril–depending on your view of an election’s outcome). But what is the alternative to a hand recount? And if there is none, what can be done to attenuate the peril?

  8. sam says:

    And, Jeez, SV, dial it a back a little, you’re sounding like Bithead’s Evil Twin (evil from his point of view).

  9. Tano says:

    “I can’t recall a statewide election that ended closely and required either a recount or going to the absentee ballots in which the Democrat didn’t come out ahead.”

    Huh? hello….
    Didn’t you just mention Florida 2000?

  10. James Joyner says:

    Didn’t you just mention Florida 2000?

    It ultimately didn’t go to statewide recount. I have little doubt that Gore would have prevailed had it done so.

    But what is the alternative to a hand recount? And if there is none, what can be done to attenuate the peril?

    Leaving the initial results stand unless a strong case is made that fraud occurred? And then counting only ballots in which the instructions were followed?

    And, of course, making ballots uniform and simple to begin with.

  11. tom p says:

    Unfortunately, that leads to a human decision having to be made as to the voter’s intent. And those humans have preferred outcomes. It’s a process fraught with peril.

    Seeing as both candidates have representatives there to challenge any ballot where intent is unclear, and that those humans have preferred competing outcomes… and lacking any evidence to the contrary (your vague misgivings do not count, James), I suspect the recount process is fair,

    Leaving the initial results stand unless a strong case is made that fraud occurred?

    In this day of electronic voting, it is even more important, especially in close elections. We have the scan machines here in Crawford Co and if it were not for that paper trail, I would have had no faith in the system. We have been using these machines for several election cycles, and each time the pens have changed. Why? I suspect that they do an audit of the votes after the elctions, find problems, and then try to correct them, as a good non-partisan election board should.

    Fraud is the least of the problems that can occur. For instance, why did it take so long to declare MO for McCain? The rest of the country was weeks ahead of us. I would really like to know the answer to that question.

  12. Floyd says:

    “”For reasons I don’t quite understand, Democrats seem to fare much better on recounts than Republicans.””
    “”””””””””””””””””””””””””””””””””””””””””””””””””
    Denial?? Naivete’??
    Or is it like Illinois, where they always keep a sufficient quantity of “contingency”votes on hand?

  13. odograph says:

    “”For reasons I don’t quite understand, Democrats seem to fare much better on recounts than Republicans.””

    My serious question would be to ask if you actually have a significant sample size?

  14. James Joyner says:

    lacking any evidence to the contrary (your vague misgivings do not count, James), I suspect the recount process is fair

    The problem is that there’s no way to know. Even those of us who follow American politics fervently basically have to take their word for it. When the process consistently overturns Election Night result, it’s simply natural to question the legitimacy of the process. My contention is that the perception of legitimacy is at least as important as accuracy in elections decided by a fraction of a percent of the vote.

    My serious question would be to ask if you actually have a significant sample size?

    Statewide elections that are so close as to require a recount are rare. But we have the 2000 presidential election, the 2004 Washington governor’s race, the 2002? North Dakota Senate contest, the 2008 Alaska Senate race, and the 2008 Minnesota Senate race off the top of my head.

  15. tom p says:

    My contention is that the perception of legitimacy is at least as important as accuracy in elections decided by a fraction of a percent of the vote.

    On that, I whole heartedly agree with you.

  16. Steve Verdon says:

    Ha ha ha ha.

    Shorter Nate Silver:

    I’ll fit a regression model with 8 variables, all of which are statistically significant. But when I want to do my analysis I’ll use a reduced form and count only 2 of the 8 variables and pronounce Franken the winner.

    Of course he could still turn out to be right, unfortunately his logic and model are wrong.

    To use the model to predict the winner you’d have to have the total number of challenges initiated by each campaign and then do the crunching. There is no justification for tossing the variables he’s included simply because in December there wont be anymore challenges. That is true, but we don’t know how many there will be in each precinct so his projection/prediction is based on faulty logic.

  17. Zelsdorf Ragshaft III says:

    So the Democrats get to steal another election by cheating. When do we start the 10 million man march on Washington to take back our country?