Two Theories on the Polls

Among nearly all demographic groups, the shift from Obama to Romney is significant Last weekend’s polls could not be closer. The Real Clear Politics average shows President Obama and Governor Romney at virtually identical percentages, separated by only half a point. However, two respected polls emphatically diverge: As of October 28, the National Journal poll showed Obama ahead by five percentage points, and Gallup showed Romney up by five.

There are two theories about this, one a center-left theory, and the other a center-right theory.

Everyone agrees that polls tally no more than public opinion (influenced by mood on a particular day) on the day they are taken. But the only thing that counts on Election Day is the actual vote, not the opinions of yesterday or last week. And actually going to vote, or applying for an absentee ballot and sending it in, takes far more effort than merely having an opinion. This is where the two theories about the voting public come into play.

The center-left theory is that the composition of the vote (by age, race and ethnicity, religion, gender, political party, and even the relative proportions of each) this election will be very similar to the composition of the vote in 2008.

The center-right theory is that in 2012, the vote among whites, women, and other demographic groups — especially Republicans and independents who actually come out to vote — will be a couple of percentage points lower for Obama and a couple of points higher for the Republican than in the 2008 vote.

It is not clear on Sunday who will be correct on Tuesday. The tabulation of the vote on November 6 will tell us that. Then it will no longer be a matter of theory, but a fact.

Of course, there are bound to be disputes and contentions over certain ballots, and the final count could be delayed by some days. On Election Day, any ballots challenged by one side or the other will be set aside for later examination (as in Florida in the 2000 Bush–Gore election).

For now, partisan passions sway each side, each claiming certain victory.

There are, I believe, powerfully credible reasons in favor of the center-right theory’s coming closer to the actual final vote. The main reason is that among virtually every demographic group, support for Obama is lower in 2012 than in 2008, which the center-left pollsters are not picking up in their template for the composition of the electorate. They are counting on the same number of independents, women, the young, Catholics, and other blocs voting for Obama in 2012. Almost certainly not.

Well, theory is one thing. Actual results are another. Results will show who was correct all along.

Today and tomorrow, in the American tradition many prayers to Providence will be said. These will be prayers in the darkness, for God’s actual will is not known to us in advance of the event.

Published in National Review Online on November 5, 2012