Campaign Diaries

Less than 48 hours from polls closing in most swing states, Barack Obama is the clear favorite to win the presidency. Renowned polling outlets (Pew, Gallup, CNN/Opinion Research, CBS News) are now releasing their final national surveys, and they, alongside nearly all polling firms, stand being discredited if McCain prevails. Obama is such a clear favorite, in fact, that it is hard to find McCain a path to victory. How could he close such a large national lead? Even if he were to somehow win Pennsylvania, could he also save the myriad of red states that are leaning towards his opponent?

It is worth looking at possible factors that could account for an unexpectedly strong McCain showing on Tuesday, so I have put together a list of items that keep some Democrats awake at night. Needless to say, for McCain to win nearly all polls must be consistently understating his support, perhaps because of some flaw in their turnout model or because of racial factors; McCain would also need to convince most if not all voters who remained undecided heading in the final week-end. Unlikely, yes, but there are still plausible scenarios.

(The same exact exercise could be done in reverse, as there is a long list of factors that might be leading to Obama’s support being understated by some pollsters, such as cell phone users not being taken into account or the continuation of the surge in African-American turnout we have been seeing in early voting. But given that Obama is already favored, there isn’t much of a point in finding him more paths to victory than he already has. I am also giving less air than I would typically to counter-arguments against these scenarios, as I am here trying to test the plausibility of scenarios that would allow McCain to win.)

1) Race: Will there be a Bradley effect? This is a question that has been keeping everyone occupied ever since Obama wrapped up his party’s nomination. There has been little to no instances of a gap between a black candidate’s polling performance and final results over the past decade (see Harold Ford’s strong closing, for instance), and I argued in late September that Obama should have little reason to worry about a drop in his performance on Election Day because it is important not to confuse voters who are supporting McCain because of racial factors and voters who are lying to pollsters about supporting Obama because of shame about their racial prejudice. The former factor certainly exists, but it has nothing to do with the Bradley effect; it is the latter we are concerned with.

However, it is unlikely that polls could as off as they would need to be for McCain to actually have a fighting chance if at least some racial factor was not in play. And there are scenarios under which a Bradley effect could matter: There has never been an African-American candidate running for this high an office; so do historical patterns really matter? Furthermore, those areas in which a Bradley effect would be most powerful (if it manifests itself) are precisely those areas McCain has to do very well in on Tuesday (starting with Western Pennsylvania, places like Scranton, Southeast Ohio, Southern Virginia) while the GOP has more or less conceded states with less racial polarization in which a Bradley effect is less likely to occur (Colorado).

2) Undecided voters are whites reluctant to vote for a black candidate: This is a variation of the previous scenario, but it does not concern Obama’s support being overstated as much as McCain’s support being understated. Instead of lying to pollsters, some voters who should be voting for Obama but realize that they are not doing so for racial reasons might be saying they are undecided. That some polls are now insisting that undecideds are disproportionately white (for instance Mason Dixon’s Virginia poll) opens the door to this possibility.

3) Undecided voters are disgruntled Republicans who will end up coming home: This is the more plausible scenario according to which undecided voters will heavily break towards the Republican. And it is a somewhat convincing one: Most of the focus groups that were being conducted after the debates featured a disproportionate number of Bush voters or of conservative-leaning voters who were disgruntled with the incumbent Administration and looking for a change but not quite sure they can go all the way and vote for Obama. Democrats say that if they haven’t gone home until now, they are unlikely to suddenly feel comfortable enough to vote for McCain and they will at worst stay home. Republicans hope that the recent focus on Obama’s character, his tax policies and alleged socialism will get those voters to stick with McCain (see #8). One problem with this scenario, however, is that some polls (for instance the latest CBS News survey) suggest that there as many undecided Democrats as they are undecided Republicans.

4) Persuadable voters are likely to be weary of unified government: Voters that remain truly persuadable in the final days of an election are unlikely to be partisans, and they are more likely to like divided government in Washington, DC. As reports of an Obama juggernaut and expanding congressional majorities for Democrats pile up, could these voters grow weary of giving Obama too much of a mandate and move towards McCain? (The Denver Post appears to find at least one such voter.) I have argued before that the reflex of a voter who wants to vote Democratic but is worried about unified government would be to vote Republican at the congressional level rather than in the presidential election, but predictions of an Obama triumph have reached such fever pitch that such an effect cannot be ruled out.

5) Complacency: The Obama campaign is worried enough that reports of an inevitable victory could deflate Democratic turnout that they are insisting that the race is tightening, going as far as to send an e-mail to Pennsylvania volunteers warning that the campaign is not meeting its phone-banking targets. The high level of early voting Democratic turnout suggests that Obama’s base is hyper-motivated, however, and that the burden is on Republican voters to meet the Democrats’ enthusiasm rather than rely on lower-than-expected Democratic turnout.

6) Electorate’s breakdown is not as favorable to Democrats as some predict: Not all pollsters agree on what turnout model should be used to predict the election. While some surveys contain a wide gap in their partisan breakdown (suggesting that the electorate will be more Democratic than it was in 2004), others weigh their results with a breakdown far more favorable to Republicans (this is the case of Zogby, for instance). While Mason-Dixon provides very little information about its internal numbers, it seems safe to say that this trustworthy pollster believes that the electorate will favor Democrats by the extent some are predicting - leading to the Republican lean of most of its state polls. Whose turnout model proves right is, of course, the $1,000,000 question.

7) First-time, sporadic or young voters do not show up: This is a complement to factor #5, as it also concerns differing turnout models. The more first-time and sporadic voters go to the polls, the more the electorate resembles a registered voter model rather than a likely voter model, the better it is for Obama. But these are voters who historically tend to not vote (which is why they are defined as sporadic…), making Obama’s electoral coalition somewhat more unstable. As for young voters, they have broken the hearts of candidates who have relied on them time and time again. Will 2008 be different? Democrats say it will be, but we have heard that tune before.

8) The spotlight has been on Obama: We always knew that McCain’s hope of winning rested in disqualifying Obama and focusing the spotlight on the Illinois Senator - and this is exactly what happened in October. As Obama took a commanding lead, the spotlight shifted exclusively on him, his tax policies, his socialist tendencies, William Ayers. There has been very little comparable probing of John McCain over the past few weeks; sure, Obama has been running attacking his opponent on health care on TV airwaves, but the constant Democratic offensive of August and September has died down. Combined with the sense that Obama is heavily favored, this means that voters will be thinking about Obama in the final days rather than about McCain, increasing the latter’s prospects of coming out as the lesser of two evils. This could especially be the case among Republicans who are tempted to vote Obama.

9) Voting problems, malfunctioning machines: Some type of election fraud is obviously the Democrats’ nightmare after the 2000 Florida debacle and the controversies surrounding Ohio and former Secretary of State Blackwell in 2004. Electronic machines are particularly high on the list of Democratic worries.

Beyond the question of fraud, however, there is a whole array of problems that could arise: machines could malfunction, polling places could run out of paper, and precincts could be so swamped by a surge in turnout as to be unable to accommodate everyone. What happens on Tuesday if there is a 10+ hour line in some precincts, as was the case in Georgia during early voting? I noted this morning that the Obama campaign was recruiting volunteer magicians or musicians to entertain voters while they stand in line and make sure that no one leaves the line.

Such problems would disproportionately affect Democrats, as the most deficient machines and the most under-prepared precincts tend to be those in urban, low-income and African-American neighborhoods - all constituents among which Obama needs to perform well.

One key difference from 2004, however, is that Democrats have gained the Secretary of State position in a number of crucial swing states - first and foremost in Ohio, where a number of Jennifer Brunner’s rulings drew criticism from Republicans. In particular, Brunner’s refusal to supply counties with a list of new registrants whose information did not match with state databases prevented thousands of names from being purged from the list and will make it more difficult for Republicans to challenge voters at the polls. (This went all the way to the U.S. Supreme Court, which ruled against the state GOP.) However, Democrats are angry at decisions taken by Colorado and Georgia’s Secretary of States.

What else is keeping Obama supporters up at night? What else is giving Republicans hope?