In my last post, I noted that the third presidential debate did not move the needle in terms of the popular vote. Before the debate, Romney led Obama by 5 points in the Gallup poll and 2 points in the Rasmussen poll. In the aftermath of the debate he leads by 5 in Gallup and 3 in Rasmussen.
Of course the needle might move yet. There could be a surprise external event that affects perceptions of the candidates. Or one of the candidates might commit a serious gaffe. Or, conceivably, voters might view one or both candidates in a somewhat new light as they make their final assessment.
But let’s assume that the needle doesn’t move. Let’s further assume, to be conservative, that Rasmussen’s number (Romney plus 3) is correct. Finally let’s assume that the undecideds in the Rasmussen poll divide evenly (another conservative assumptions, since undecideds usually break for the challenger).
Can Obama win in the electoral college if he loses the popular vote by 3 percentage points – a raw vote margin of almost 4 million?
This stikes me as an exceedingly difficult feat. As far as I can tell, only Rutherford Hayes, in 1876, ever won enough states to carry the electoral vote, having lost the popular vote by more than 3 points.
Except that Hayes didn’t really accomplish this. His opponent, Samuel Tilden, won the popular vote in more than enough states to carry the electoral vote. Unfortunately for Tilden, more than one state deprived him of his victory by simply overturning the result of its popular vote. For example, Louisiana, still under federal control near the end of Reconstruction, cast its electoral votes for Hayes even though something like 6,000 more votes were cast for Tilden.
George W. Bush, of course, also lost the popular vote but won in the electoral college. However, Al Gore’s popular vote margin was only 0.5 percent (about half a million votes).
In 2004, John Kerry lost the popular vote by about 2.4 percent (about 3 million votes), yet still would have won the electoral vote had he prevailed in Ohio. But, contrary to common perception (including my own), Kerry didn’t come very close in Ohio. He lost the state by almost 120,000 votes. That translated to 2.1 percent, barely better than Kerry’s national performance.
It seems, then, that for Obama to win the electoral college based on the Ohio vote, he probably would need to be within 2 points of Romney in the national voting. And even then, a 2 point national spread might very well give Romney a path to victory that doesn’t require winning Ohio (some combination of Wisconsin, New Hampshire, Colorado, Iowa, and Nevada).
Still, I don’t say it’s beyond the realm of the possible for Obama to lose the popular vote by 3 percent and still be elected. For one thing, Rasmussen, though finding Romney 3 points ahead nationally, says that right now Obama has “the edge in the electoral college.”
Moreover, Obama has targeted Ohio in ways that Kerry could not, by blitzing the Buckeye State’s airwaves with anti-Romney ads during a period when the challenger couldn’t do much advertising. To be sure, Obama did the same thing in Virginia and Florida, where Romney now appears to be ahead. But Romney may encounter an extra level of resistance in Ohio because (a) he opposed Obama’s auto bailout plan and (b) the unemployment picture in Ohio isn’t as bad as it is nationally.
On balance, though, it seems highly unlikely that either candidate could win in the electoral college without coming within 1.5 percentage points of his opponent in the national popular vote.
UPDATE: I should add that this post assumes that Ohio doesn’t end up in the Obama column as the result of fraud. Unfortunately, there is reason to believe that the Democrats are engaged in voter fraud in Ohio as I write this.