Hindsight Bias and Obama's Victory
Brendan Nyhan writes today about the metamorphosis of campaign coverage:
The media has undergone a strange change of mindset. Immediately before last Tuesday’s election, many reporters and commentators ignored or dismissed the consensus among forecasters and betting markets that President Obama was very likely to defeat Mitt Romney and acted instead as if the candidates were neck and neck or Romney was ahead. Afterward, however, coverage of how Obama won betrayed far less uncertainty about the outcome of the election, which was frequently portrayed as a fait accompli—an inevitable consequence of how Romney’s image was defined by Obama’s early ads or overwhelmed by the President’s superior ground game.
Brendan says this is a result of hindsight bias, and I suppose that's true, sort of by definition. But there's nothing unique here. Before the Super Bowl, sports talkers chatter about how well the two teams are matched up. After the game, they talk about how the winner managed to win. Why? Because the game is over. They now know what worked and what didn't.
I'm not sure it's really fair to call this "bias." Once a contest is over, and you know who won, you also have a better idea of which tactics won. In the case of the 2012 election, reporters have concluded that defining Romney early worked and that Obama's ground game made a difference. If he had lost, they would have concluded the opposite. They might be wrong in those conclusions—hell, historians are still arguing about why the South lost Gettysburg even after 150 years of study—but there's nothing irrational about it. I happen to agree that reporters tend to overdo this, paying too little attention to things like economic fundamentals and the power of incumbency, but still, once you know how something turns out, it's perfectly sensible to conclude that the winner's tactics were effective and the loser's tactics weren't.