Obama is in the lead, both nationally and in some very key battleground states. This election is his to lose. But elections can be surprising affairs. Let's ask a question: What combination of factors would need to occur for Obama to lose? And how realistic is that scenario?
Here are some important factors to keep in mind.
1. You can't trust the polls. We have seen too many examples in the past when the polls were misleading. When you have "national" polls purporting to show that Obama has a lead in one poll of 12 points and in another poll of two points, that tells you something is off. So much for the "science" of polling. When I see a poll that tells me they have a margin of error of 3.5%, I double that. A poll with a seven percent margin of error is only good for giving you some indication of general trends, nothing more.
2. Undecided voters. In recent presidential elections, the undecided voters have broken for the Republican candidate. And a recent poll from the Associated Press found that 1 in 7 voters are still undecided. Assuming this poll is accurate, that's a lot of undecided voters, about 18 million. If the undecided voters break for McCain 60% to 40%, that's a pick up of 3.6 million voters for McCain. If spread across the right battleground states, that could translate into a McCain victory.
3. "Invisible" voters. On both sides, we keep hearing about all these "invisible" voters who will turn out for them. On the Obama side, a lot of young voters supposedly are going to vote this time -- but we have heard that before. Then supposedly there are a lot of newly registered minority voters. But African-American voters have voted at nearly the same percentage as whites, so there might not be much potential there for an increase. Latino and Asian voters might increase some, but those populations don't necessarily identify closely with a black candidate. So let's make a modest assumption that there will be some increase, but not the tsunami that some are predicting.
On the Republican side, the invisible voters would be not only those undecided voters who break late for McCain, but also those who are part of the notorious "Bradley effect," i.e. white voters who won't vote for a black candidate yet don't show up in the polls that way. Polls have found that a fair number of Americans will not vote for a black candidate, so we can assume some degree of a Bradley effect, but how much? No one really knows.
So let's say that, on the whole, these two pools of invisible voters cancel each other out. We don't know ultimately how these factors will play, but erring on the side of caution, let's say it's a wash.
4. The winning Republican playbook. The McCain campaign is using the same Republican campaign plan that has been successful in winning presidential elections since Nixon. That involves attacking the Democratic candidate as a tax-and-spend, big government liberal who is un-American, is a socialist and a pal of terrorists who is not part of the cultural fabric of mainstream America.
The common wisdom is that the Republican template is not working this year due to the economic crisis, which is widely blamed on the incumbent party. On the other hand, if so many voters are still undecided at this point in the election, then clearly there is something about Obama and the Democrats that make them uncomfortable. Something about Obama/Democrats has not allowed them to close the deal on this election. So the McCain strategy of slowly chipping away at Obama's lead using the same old Republican playbook may pay off yet in these last few days, as it has worked previously with the undecideds.
5. Disqualified voters. Around the country, tens of thousands of registered voters have been purged, many of them through no fault of their own but due to clerical errors which have caused a mismatch in their name in various state databases. The evidence indicates that most of these voters are probably Obama supporters, so the Obama advantage in new registrants may be muted.
Also, like in past elections the Republicans are doing everything they can to disqualify Democratic voters. They are preparing to challenge certain types of voters at the polls, even possibly challenging voters who have lost their home mortgages and had to change addresses, making their voter registration invalid. There are reports of letters being mailed to minority voters warning them of possible arrest if they try to vote and have any outstanding traffic violations. No one knows how effective or widespread these efforts will be.
6. Defective voting equipment and poor election administration. We already are hearing stories from around the country about problems with voting equipment. One problem has been "vote flipping" on touchscreen voting equipment that lacks a voter verified paper trail -- a voter selects one candidate and the machine picks another candidate for them. This probably results from lousy voting equipment rather than some conspiracy to steal an election, yet it's really odd that all the anecdotes seem to indicate that it's usually a Democratic vote that gets flipped to the Republican candidate.
Also, remember the "butterfly ballot" in Florida in 2000, which caused thousands of voters to select the wrong candidate? In 2004, no one anticipated there would be so few pieces of voting equipment in university precincts and black precincts in Ohio, causing extremely long lines and people to go home frustrated without voting. What election administration snafus await us in this election? Nobody knows.
OK, so let's add up our six factors. All the polls show Obama with a lead, but with polls having a mixed track record, we don't really know how sizable. Meanwhile, there seem to be a fair number of undecided voters who historically have broken for the Republican candidate. The Democratic candidate is African-American with a known Bradley effect out there, though the impact is difficult to quantify. There are unknown questions lingering about defective voting equipment, faulty election administration, purged voters, and Republican attempts to disqualify Democratic voters. What will be their impact, ultimately, on vote totals?
Given all these factors, is there a plausible scenario in which McCain could win? I would say yes, absolutely. Perhaps not likely, but certainly plausible. There is also a plausible scenario in which Obama wins handily. The truth is, there are too many moving parts in this election to know for sure. But if all the above factors line up just the right way, the nation could be in for a shock on Election Night.
As former President Bill Clinton once said, "The paranoid win elections." And now we know why.
Steven Hill is director of the Political Reform Program of the New America Foundation, and author of "10 Steps to Repair American Democracy" (www.10Steps.net)