Karl Roveâ€™s incredulous response when Fox News called Ohio for the president on Tuesday night has attracted considerable scrutiny. Most pundits saw it as a garden variety case of wishful thinking. If someone wants something to be true strongly enough, he can believe it even in the face of overwhelming objective evidence to the contrary. But this explanation leaves some troubling questions unanswered.
Rove is a numbers man. Heâ€™s been a close student of political polling for three presidential election cycles, and as even his most vehement detractors concede, he is extremely intelligent and a consummate pragmatist. Ohio was by far the most heavily polled swing state during this election cycle. From the beginning of the campaign, the overwhelming majority of polls showed the president ahead. In the closing days his margins appeared to be increasing, with an average lead of roughly three percentage points. If the polls were correct, they foretold an almost certain Obama win in Ohio. Rove was surely following them closely.
He was also surely aware of well-documented biases that caused many polls to understate the Presidentâ€™s support. Some, for example, call only landlines, missing many Democratic leaning younger voters who have only cellphones. Most polls query respondents only in English, a practice that understates the likely participation of Hispanic voters. Many employ automated robocalls with limited repeat calls to non-respondents, which tilts samples toward Republican older voters, who are more likely to pick up their phones.
And then there was the vaunted Obama ground game. Long before the election, even Republican strategists acknowledged its significant advantages over its counterpart in the Romney campaign.
In an earlier post, Mark speculated hopefully that these factors might boost the Presidentâ€™s national popular vote margin by several percentage points relative to election-eve polls. Of course, many conservative pundits insisted that published polls were biased in the opposite direction, arguing that Romney supporters were more enthusiastic. But none offered persuasive objective evidence for that claim.
Rove might have hoped that Republicans would turn out in unexpectedly large numbers in Ohio, but it strains credulity to insist that he felt sure that Romney would prevail in an honest count of the votes there. Having witnessed Roveâ€™s Fox meltdown live, and after having reviewed clips of it several times since, I find it hard to believe that Roveâ€™s astonishment at the Fox announcement was feigned.
So if Rove REALLY thought Ohio was in the bag, we seem forced to choose between an implausible claim and a disturbing one: Either he is much less competent than anyone has reason to believe; or else he knew of some secret advantage that would tip the vote count in Romneyâ€™s favor by several points. Ohio, in any event, was the only swing state in which the presidentâ€™s final marginâ€”1.9 percentâ€”was smaller than forecast by the final round of polls.