On Fox News Sunday, Brit Hume highlighted Mitt Romney’s clear advantage in Gallup, Rasmussen, and other national polling, and said, “Now…if those polls are generally correct, it is difficult to imagine that Ohio would be all that different. Ohio has pretty closely tracked the national outcome…since about 1960.”
Indeed, in nearly half (6 of 13) of all presidential elections starting with 1960, the margin in Ohio has been within 2 percentage points of the national margin. And all but once over that span, Ohio’s margin has been within 4 points of the national margin (the exception being in 1960, when Richard Nixon beat John F. Kennedy by 7 points in Ohio during an election in which the popular vote remains in dispute to this day).
But while Ohio has closely tracked the national outcome, its specific path has generally been to the right of the national average. Otherwise stated, the Republican nominee usually does a little better in Ohio than nationally. In fact, Ohio has been to the right of the United States as a whole in 10 of the past 13 presidential elections — or 77 percent of the time. (That includes 2008, when Barack Obama won in Ohio by 2.7 fewer points than he won by nationally.)
The only three times since 1960 that the Democratic nominee has done better in Ohio than nationwide were in 2004 (when John Kerry did just a smidgen better in Ohio — losing by 2.1 points — than nationwide — losing by 2.5 points) and in the landslide elections of 1972 (when George McGovern lost by 23 points nationally and 21.5 points in Ohio) and 1964 (when Lyndon Johnson won by 23 points nationally and 26 points in Ohio.) The other ten times, including in every close election except for 2004, the Democratic nominee has done worse in Ohio than nationally.
In other words, it would defy a fair amount of historical precedent for President Obama to lose the national popular vote — especially by more than a point or two — and still win in Ohio. That’s not to say it can’t happen, just that it’s not how things normally happen.