A Post-Mortem on Post-Mortems
With 95 percent of gasbags reporting, an unofficial tally shows that there are 7,936,682 lessons that we must learn from Mitt Romney’s defeat. My favorite comes from David Axelrod, President Obama’s suavely mustachioed major domo. On a conference call with reporters after Election Day, Axelrod spoke darkly of the vast amounts of money Republicans had spent on Romney’s campaign.
“The heartening news,” he said, “is that you can’t buy the White House.”
It was an unintentionally amusing thing for a man to say who had just spent more than one billion dollars so his candidate could live in the White House for another four years. Yet no one on the conference call snorted or hooted in response. Journalists aren’t prepared for hilarity when political hacks speak on the record. More to the point, the business of lesson-drawing, in the hours and days immediately after a big election, is assumed by everyone to be deadly serious.
It certainly is oppressive. Compulsively I read about a week’s worth of election analysis before I was able to resume what in my case passes for normal life. The post mortems began instantaneously; the body politic hadn’t stopped twitching on election night before the coroners were digging in with their rib shears and skull crackers. NBC News declared Obama the winner at 11:12 p.m. National Journal published its story “How Obama Won” at 11:31 p.m. The answer, said the article, was a “multi-million dollar media blitz casting aspersions” on his rival “before Romney had even…claim[ed] the nomination.” A couple hours later, morticians at the Wall Street Journal credited the victory to Obama’s “late spring offensive” of “ads blasting away” at his opponent before Romney could respond. Almost simultaneously, the forensics team at the New York Times reached the same conclusion.
I was satisfied. The sun wasn’t even up yet and the mainstream press had already offered me a highly plausible and elegantly uncomplicated reason for Obama’s victory. But that instant explanation was too meager to satisfy the gluttonous political class. An election that is being observed by thousands of professional observers can’t be packed away so easily; the observers will find larger significances or die trying. Among Republicans, of course, this involved a great deal of doom-saying and blame-shifting. Responsibility for the humiliating and (to most of them) surprising defeat was tossed around like a hot brick: It was the fault of Romney, or his staff, or the conservative press, or the “Republican establishment,” or the Tea Party, or most originally the weather.
The Fox commentator Dick Morris, who had predicted a Romney landslide, wasn’t alone in fingering the weather. Hurricane Sandy stopped Romney’s surge in the polls, Morris said: “It made all the difference.” Except it didn’t. Romney’s surge had stalled while Sandy was just a twinkle in a forecaster’s eye. But there was more, Morris said: “I derided the media polls for their assumption…that blacks, Latinos, and young people would show in the same numbers as they had in 2008. I was wrong. They did.”
Except, again, they didn’t. It takes a special kind of pundit to be wrong even as he’s confessing how wrong he’s been. Exit polls showed that Obama’s share of the Latino vote inched up from 2008, but his share of the black vote dropped two percentage points. In 2008 he won 66 percent of the under-30 voters; in 2012 it slipped to 60 percent. Indeed, Obama won several million fewer votes than he had four years earlier (the precise number won’t be known for weeks). Nevertheless, lots of Republicans saw the election as proof that the USA had irrevocably changed, and not for the better. By his own account, when Rush Limbaugh scanned the early exit polls on election night, he let out a cry: “How in the world can you deal with this?” It was a glum El Rushbo who tucked himself in that night. “I went to bed,” he told his listeners, “thinking, ‘We’ve lost the country.’” By the next morning he had discovered a “nation of children” enthralled by Obama’s Santa Claus.
Perhaps Limbaugh is showing the strain of being a champion of the common people at a time when half the common people disagree with him. But the mantle of doomsayer fits him no better. It is in the nature of partisans to swing wildly from triumphalist (Republicans circa 2010, Democrats 2012) to apocalyptic (Republicans 2012, Democrats 2010) without any stops in between. My friend Jonathan Last recalled the Times columnist Thomas Friedman softly weeping ’neath the mourning veil in 2004. “This election,” Friedman wrote then, “was tipped because of an outpouring of support for George Bush by people who don’t just favor different policies than I do—they favor a whole different kind of America.”
The Republican gloom, like Friedman’s, was rooted in the old superstition that demography is destiny—the historically unsupported belief that while certain groups of people will grow dynamically, their political sympathies will remain static. But the predictable triumphalism among Democrats had superstitions of its own. These were rooted in the consuming fad of “quantitative analysis” that has swamped everything from baseball to gardening. Generations of campaign professionals have sifted data, but the numerical faith reached a climax this year. “I love numbers,” said Jim Messina, Obama’s campaign manager. He boasted of his campaign’s teams of behavioral scientists. “We were going to demand data on everything.”
And it worked spectacularly. The campaign identified pockets of voters that hadn’t been exploited before, goosing the margin of victory. Indeed, quantitative analysis always works spectacularly—until it doesn’t, as the “quants” who manipulated the mortgage-derivatives market in 2007 and 2008 can testify. Even Messina’s operation, expensive and impressive as it was, made errors here and there. Messina said his number-crunching and behavioral analysis would actually increase Obama’s turnout from 2008. It didn’t come close. And miscalculations in several swing states led the campaign to spend money where it didn’t have to and underspend elsewhere.
The election’s quant hero was an unprepossessing Times polling analyst named Nate Silver. Like aborigines greeting the Great White Hunter in an old Hollywood movie, credulous Democrats simply assumed Silver’s godlike powers. A besotted Jon Stewart interviewed him after the election. “Don’t you want to stand up and say, ‘I am Nate Silver, bow down to me!’” Bwana! Any roomful of Democrats would have been happy to oblige, and reading the blogs and comments on the Web you got the creepy feeling that Obama’s followers had begun to think that the quantitative analysis, including Silver’s polling model, actually caused the victory. At a minimum, as one Huffington Post blogger put it, “hard political science” had vanquished Republican backwardness.
Here again professional Republicans and Democrats are not so far apart as they want to believe. The demography-is-destiny gloom of one and the techno-euphoria of the other are alike in their determinism. Both views assume skin color or ethnicity or age or marital status will overwhelm the vagaries of willful human individuals, who nevertheless tend, over time, to frustrate the most sophisticated calculations. In their fumbling fashion people respond instead to words and images, to good advertising and bad, to appeals to interest and even (wonder to behold) to ideas, especially if presented appealingly.
On that assumption, my own nonstatistical, non-computer-generated probability model has developed an algorithm of its own. In four or six or eight years, the triumphalist of today will trade places with the doomsayer. The ex-triumphalist will be stunned by the election result, and the former doomsayer will say he saw it coming all along.