Obama’s Lead in the Polls is Not a Historical Anomaly

Ever since Obama established a narrow but clear edge in the polls after the Democratic Convention, conservative commentators such as Niall Ferguson, Andrew McCarthy, John Hinderaker, and Peter Robinson have claimed that this is an unusual historical anomaly explicable only by Romney’s poor campaign strategy or lack of personal appeal, or by some sort of transformation of the electorate in a left-wing direction. They assume, as Hinderaker puts it, that “this election should be a cakewalk for the Republicans” because of the poor state of the economy.

In reality, however, standard economic models of presidential election outcomes all point to a close race, for reasons I explained in this post. Larry Sabato’s recent summary of the predictions of 13 standard election models relying primarily on economic variables also points to a close race with a small edge to Obama. On average, the models predict that he will get 50.2% of the two-party vote. A crude summary of the reasons why the models come out this way is that 1) there has been some improvement in economic conditions since 2008, even if a modest one and 2) things were pretty bad in 2008 to begin with, which leads some swing voters to blame Bush and the GOP for the slow pace of the recovery as much or more than Obama and the Democrats.

Some of the models also take account of foreign policy events. While one can certainly make a case against Obama’s foreign policy, he has not presided over a large and obvious failure that can clearly be laid at his door in a way that swing voters – most of whom have very low levels of political knowledge – can readily grasp.

It’s worth noting that these models also do not account for any boost that Obama is getting for having ordered the operation that killed Osama Bin Laden. If Obama has gotten even a 1 or 2 point boost from that event, that’s enough to account for much of his edge in the polls (currently about 3% in RCP’s average of national polls). The death of Bin Laden is a large and obvious success that even relatively ignorant voters are likely to find out about and appreciate.

The bottom line is that if Obama is overperforming historical expectations, it is only by a slight margin. Similarly, Mitt Romney is doing at most only slightly worse than historical averages would predict – unless you want to argue that Obama is running a highly incompetent campaign that has somehow been offset by even greater incompetence on Romney’s part. We can’t categorically rule that possibility out. But I think it’s pretty obvious that the Obama campaign has displayed at least average levels of competence, if not better.

I am no great fan of Romney, who I think is far from an ideal candidate. A GOP candidate with great bipartisan prestige (e.g. – Eisenhower in 1952) or high charisma (e.g. – Reagan in 1980), would likely have done better. I am also no great admirer of Romney’s record on policy issues, or his technocratic mentality.

So far, however, Romney’s performance in this race is relatively par for the historical course. It is reasonable to argue that he should be doing slightly better, but not to suggest he should be winning in a “cakewalk.”

UPDATE: Some commenters and others question whether Obama really has a 3 point lead, as the RealClearPolitics average of recent polls suggests. Given the margin of error of polls, it’s certainly possible that Obama’s true lead is only, say, 1 or 2 points, though an average of multiple polls is more statistically robust than just one. Regardless, my argument is entirely compatible with a slightly smaller Obama lead than RCP shows, or even a tie race. That too is very close to what standard electoral models would predict about this race. My argument would be undermined if the true situation was that Obama has a really big lead (e.g. – 5-10 points or more). In that event, he really would be clearly outperforming historical expectations. But, so far at least, that has not happened.