So the thing I can't figure out about the left-leaning skeptics' and pessimists' position, meritorious as it may be: if McCain isn't losing, why does everyone involved in his campaign seem to think he is? The McCain/Palin folks are less a campaign than a gangbang of finger pointing, back-stabbing, and self-interest these days. If they were confident about pulling off some early November surprise, Palin wouldn't be campaigning for 2012 and the hired help wouldn't be pre-emptively spinning the loss with an eye toward future employment.
It is popular among political scientists to discuss the impact of large-scale turnout among young voters and African-Americans on our elections. This is akin to discussing what would happen if a comet hit the Earth – we make guesses based on fragmentary evidence, but no one has actually witnessed it.
It is an unquestioned fact that if voters between the ages of 18 and 24 are least likely to vote (or even be registered). In fact, the relationship between age and turnout is positive and persists until very old age.** Mountains of evidence also exist to show that black voters lag their white counterparts in turnout, although increased mobilization efforts may be closing the gap. This creates a vicious cycle in which politicians talk more about issues relevant to people who vote while ignoring issues relevant to young or black voters. They're playing the percentages. This is why you hear so goddamn much about Medicare and prescription drug prices and almost nothing about student loans, urban blight, or the decaying market for careers as opposed to entry-level jobs.
It also happens to be true, however, that young or black voters lean left. It is standard operating procedure for Democrats to attempt to increase turnout among these groups. Some have even based entire electoral strategies on it, and history is littered with their failed campaigns. The reasons for failure are numerous and conjectural – young voters have less life experience, may not consider politics important, may be ignorant of registration/voting procedures, or simply don't hear anything interesting out of the candidates (see above). Lower black turnout is speculated to be a function of cynicism, scapegoating by mainstream (white) politicians, socioeconomic deficits, etc.
Obama is not the first candidate to invest significant resources into turning out voting-eligible black or college-aged Americans. Several have hoped that it would put them over the top, only to be sorely disappointed. The thing is, young voters get real excited, swear they will vote, and then…..they don't. Likewise, large numbers of new black registrants are added each election season with a negligible increase in turnout. So this strategy, although common, has yet to produce a demonstrable victory.
We may see a meaningful increase in black- and young-voter participation in 2008, but a careful analysis of demographic splits suggests that the result will not be as impressive as many observers expect. Black voters are about 12-14% of the electorate, a disproportionate number of whom (compared to other ethnic groups) are ineligible to vote. Increasing black turnout by 10% (an ambitious goal) would only increase overall turnout among all eligible voters by about 1%. Since black voters choose the Democrat about 93% of the time – maybe 99% this time around – the majority of that increase will benefit Obama. So the 1% increase is significant enough to matter in really close states (note that some, like Iowa, Nevada, and NH have negligible black populations) but most of it will be insignificant, falling in states like Mississippi, New York, South Carolina, Illinois, and so on. It may help a tight race in Ohio, NC, or PA, although they would have to be very close for this to matter.
The effect of young voters is even more dubious. Let's say that they really jack up turnout rates from 30% to 60%. Well, 18-to-24s only comprise 8% of the electorate. And the split in their allegiances, according to available polling data, is something like 65/35 Obama. So even doubling 18-to-24 turnout is unlikely to have a statistically significant impact on electoral outcomes unless a particular state is extremely close.
In short, if Obama wins big it is going to be on account of his appeal to middle-aged and older white voters. I don't think this is the case because said voters are "more important" – they are simply the most numerous by far.
I am a fairly committed anti-skeptic at this point in the race. Turnout among college-aged and black voters, however, will remain firmly in the "believe it when I see it" camp. My feeling is that the makeup of the electorate will change while overall turnout increases only slightly. The reason is simple: for every person who would not ordinarily vote but will turn out for Obama, there is a Republican in Illinois or New York who is dangerously close to thinking "Um, fuck it." Right or wrong, Republicans have been demoralized by nine months of a bad candidate, a worse running mate, and incessant messages about the impending bloodshed in Congress. In areas not broadly considered competitive, apathy (or overconfidence) might suppress turnout as much as other circumstances promote it.
**Turnout increases with age because voters become "stakeholders" in their 30s/40s, buying homes and having kids. They have more to gain or lose. Another bump occurs at age 65, as the elderly have time on their hands. But at very old ages (80+) the relationship between age and turnout becomes inverse, as mobility, the ability to drive, and mental faculties tend to decline rapidly and take political participation with them.