Even though every political and statistical indication points to an Obama victory tonight -- and a healthy one at that -- a certain brand of liberal paranoia persists. This is too good to be true, Democrats declare, fingers grasping at their hair. McCain is tightening the race in key states. The youth vote won't come out.
And so it goes.
But if in fact McCain were to win this election it would be, one of the nation's foremost pollster says, almost historically unprecedented.
"There is no reason in history to suggest [Obama won't win]," said Frank Newport of Gallup. "All you can go by is history and compare our last polling that we have done before the election and the actual outcome in the presidential election... We have most polls showing Obama with a statistically significant lead nationally and also in these states. If he were to lose, it will be the first time since World War II something like this has happened. Now, keep in mind. It's a small sample, less than 20 elections, but it would be very unusual, in fact, exceptional... improbable."
Indeed, the last time that Gallup's final poll before the election did not accurately determine the winning candidate was 1948, when they stopped polling a week before Harry Truman's comeback victory against Thomas Dewey. Even in 1980, when Ronald Reagan staged a late comeback that turned into an electoral rout, Gallup caught glimmers of this trend just in time, showing the Gipper up three points in its last poll.
When it comes to the current election, the firm has Obama up eleven points in its final survey. But what should make Democrats more assured, said Newport, is that the Illinois Democrat has maintained a steady margin throughout the past month.
"Since September 15, Obama has been ahead in every poll we have conducted or any other polling I have seen and often by substantial margins," he said. "It is not like it is race in which McCain was leading and we are seeing some kind of shift for Obama, it has been Obama ahead pretty dominantly."
Moreover, other polling firms are documenting similar trends -- a confluence of data that validates the larger picture.
"We are all using a measuring instrument to estimate a big population," said Newport. "It is like we have a giant lake and we are trying to estimate the bacteria percentage. So we take a sample and test it and that is what we are doing. But yes... if you have 15 scientists and they are all showing the same thing, that does give you more assurance that the lake has some bacteria."
There are, of course, Obama supporters who will remain unconvinced. And as evidence they could cite the polls leading up to the New Hampshire primary, which showed the Illinois Democrat in a similarly comfortable lead only to lose to Hillary Clinton by two points. Newport acknowledged that the primary fight in the Granite State gives him and others in the business pause -- he has yet to find a smoking gun to explain what happened, though he hinted that massive late-stage change in voter preference moved too quickly for polls to pick up.
But that was, for better or worse, an aberration. Pressed to quantify just how big a failure for the polling industry a McCain victory would represent, he didn't feel comfortable even following the hypothetical.
"Call me tomorrow," he replied. "Obviously when Gallup and other scientific polling organizations do our best... and if for some reason the actual voting out there didn't mirror, internally, what we were showing, it certainly would be a time where we would have to say, 'What are we doing wrong?'... But we will cross that bridge if we get there. Right now, we aren't crossing that bridge... It is improbable. But like I said, call me tomorrow."