By Mark Mellman - 10/23/12 11:34 PM EDT
Perhaps it’s the seemingly unsettled nature of the race, or perhaps it’s just our blogging culture, but instead of following a single theme through 600 words, allow me to offer some disconnected, though not quite random, observations about the state of the presidential contest.
As I have argued for well over a year, if this race is a referendum on the president, he will have problems, but if it is a choice between him and Romney, or better yet, a referendum on Romney, the president will do better. Until the first debate, the Obama campaign did a masterful job of keeping the focus on the challenger. The race was about Romney — whether it was the decisions he made at Bain to profit on the misery of others, his overseas gaffes or his comment about the 47 percent — and the challenger was the main character in the unfolding drama. And during that period, the president was doing well, maintaining a lead of a couple of points on average nationally.
The second debate put focus back on Romney, but damage had already been done.
I noted it last week, but it bears repeating. One way or another, the horserace always ends up aligning with the fundamentals at some point in the cycle. Ten statistical models that consider only economic and political fundamentals (not head-to-head polls) predict an average two-party vote tied at exactly 50 percent. This was destined to be a close race, and while there are outliers like Gallup, three polling aggregators put the president somewhere between 1 point ahead and half a point behind. The popular vote is, as Dan Rather used to say, “tick-tight.”
Which leads to a core question: Can the president sustain his advantage in the swing states if the popular vote is close? The answer is yes. In prior decades, the Electoral College was biased against Democrats, but that has shifted of late. Speculation about a split between the popular and electoral votes is wildly overblown — while not impossible, the probability of such a split is quite low. Nonetheless, while the reasoning and the math are arcane, the conclusion is that over the last four cycles, the Electoral College has exhibited a bias of just under a point toward the Democrats. Thus, a close popular vote is more likely to produce an Obama victory than a Romney win.
Look at it this way — with all the talk about the centrality of Ohio, Florida and Virginia, the president could lose all three of these states along with Indiana and North Carolina and still emerge victorious from the Electoral College by holding the other states he won in 2008.
This advantage is neither automatic nor somehow independent of the way voters cast their ballots. If the president loses the popular vote by more than 2 points, it will be virtually impossible for him to win. But in an otherwise close contest, his multiple routes to 270 electoral votes give the president an advantage.
But with Obama below 50 percent, can he pick up votes from undecided, or do they all break to the challenger? A few weeks ago I dismantled the myth of the so-called “incumbent rule,” demonstrating it no longer applies in Senate contests. Some argued it still reigns in presidential races. Nate Silver did that math, concluding that since 1968, on average, presidential incumbents and challengers have picked almost equal amounts of the vote over the final polls. Some undecideds are likely to break to the president.
In short, we can expect a very close race, but one in which President Obama retains some advantage.
Mellman is president of The Mellman Group and has worked for Democratic candidates and causes since 1982. Current clients include the Majority Leader of the Senate and the Democratic Whip in the House.