Oh, come on John, you are both smarter and more knowledgeable about history than this. Let's put out the facts:
1) The 2-term incumbent President has an approval rating in the 20's.
2) Something like 75% of people think the "country is on the wrong track."
3) Almost twice as many people are self-identifying as Democrats as are self-identifying as Republicans.
4) But somehow, the Democratic nominee is tied or behind in most national polls.
But wait, it gets better:
5) The Democratic party is generally far ahead in the generic congressional ballot question.
Hmmm, I wonder if the nominee has anything to do with this. Oh yes:
6) Something like 20% of self-identified democrats in polls won't committ to the dem nominee.
Could the dems pick up 25 House seats and lose the presidential election? You bet they could! How in the world could that be pinned on anyone but Obama? Conditions have changed so dramatically since 2004, and we're seeing it at level of polling except for the presidency. And before Obama was nominated, we were seeing it at the level of the presidency. How in the world could Obama win fewer states than Kerry? I'm not saying it will happen, but the very idea of it happening shouldn't be crossing our minds. Anyone who argues that the playing field for the GOP has gotten better the last four years is a lunatic. Therefore, how do we ascribe current polling to anything but the candidate?
This election is so counter-historical that it's beyond unprecedented. The general mood of the country this year --- as judged by polling --- is so unfavorable to the President's party that it can only be reasonable compared to three 20th century elections: 1932 (depression), 1974 (watergate), and 1980 (hapless dems; bottoming-out economy). All three of those were bloodbaths beyond belief. With the exception of 2008, public opinion about hte administration has never been lower than those three elections, not even close. By all accounts, nothing can explain 2008 without reference to Obama.
As to your historical numbers, i think you are either ignoring the context or not presenting it:
1) No one in the history of the country has gotten more than about 60% of the popular vote. Getting 50.8 in 1980 was such a magnificent blowout for Reagan because Carter got 40% (Anderson got 10%). A popular vote margin of 55-45 that translates into 400+ electoral college votes is a blowout, period. To say that an election at 52-48 is close or not is not possible; the candidates don't seek 51%; they seek 270. Therefore, the reference point is typically the electoral college.
2) I think you are confusing popular vote with probability of winning. Roosevelt in '32 only won 57% of the vote. But as trained monkey with the democratic label had a 99%+ chance of winning that election. The GOP had been blown out of the water in the 1930 midterms (like the GOP in 2006) and everyone knew that the dem convention was the election: the winner just had to shut his mouth, wait it out until november, and take the presidency. Result: 57% and 42 states. The democrats --- prior to having a candidate --- easily have a monstrous chance of winning this election. But with Obama, they don't.
3) One might argue that the parties are so locked in right now, red and blue states, that an electoral blowout is impossible. But the '32, '74, and '80 elections precisely refute this theory: states that hadn't voted one way in over a generation turned those years.
Look, I still think Obama is going to win. But I think it will be marginal and I think it will be despite himself, not because of himself. Call me crazy, but I think John Kerry easily wins this election. The Bush administration is so thoroughly delegitimized that there is not structural explanation for his party winning the election. It must be candidate-based effects that are driving the polls.
But like I said, I could still see Obama getting 400 electoral votes. But I could see McCain getting 320. Which i literally can't believe I'm saying.
A lot to respond to here. There is no doubt that the environment favors the Democrats, but I will reiterate my point that McCain is not, in the minds of most voters, a standard issue Republican, which is why the Obama campaign has been trying to hard to portray him as such. What if you polled Bush v. Obama? Still think Obama would get ~50%? Of course not, he'd be up over 60%. But McCain is not Bush, so his crappy approval numbers are not terribly relevant here. The last time a two-term incumbant president had a horrible approval rating and wasn't running was LBJ (well, one and a half terms anyway), and in '68 Nixon won with 43.4%. I'll come back to the historical comparisons, I don't want to leave it at that.
Also, the self-identification by party ratio is not even close to 2:1. According to Rasumussen, it is currently 38.7% Dem, 33.6% Repub. In January of 2007, before Obama announced his candidacy, it was 37.5% Democrat, 32.1% Repub., so it is hard to say that Obama has caused Dems to flee the party (all data here).
As for the point about the need to look at electoral college vs. popular vote percentage, I don't think the former is the appropriate barometer for the purposes of this debate. For example, it is hard to say that Clinton in 1992 was a wildly popular candidate, but looking only at electoral results would lead you to believe that he was, because Perot siphoned off a lot of votes from Bush (as many people point out, more people voted against Clinton than for him).
The '32, '74, and '80 elections (presumably '74 was meant to be '76) were cited as examples of game changers in which conditions are somewhat similar to today. Well, in '32, the schlub who had screwed things up, Hoover, was actually running, so people could vote against him. In '76, the vice-president-now-president of the disgraced Republican party was on the ballot, and Carter still only got 50.1%, with no interference from third parties. And finally, in '80, Reagan was running against the actual unpopular guy, just like Hoover in '32 -- people who didn't like Carter could vote against Carter. And, Reagan only won in an electoral college blow-out because of a strong third pary candidate (Reagan got 50.8% of the popular vote). The key here is that in those three elections that are cited, the administration itself was on the ballot, but in this one it is not. McCain is not Bush, and many voters are willing to give him the benefit of the doubt.
Now, if you want to argue that Obama is polling a bit lower than a generic Dem should be because of some latent racism in this country, or anti-Muslim religious bigotry (don't get me started...), well yeah, I'll give that to you. But I don't think that is a terribly large factor.
So here's my question. What popular vote and/or electoral college outcome would be 'acceptable' for Obama and what would constitute him underperforming? Remember, right now, the polls are tied, and McCain just had the best two weeks he's going to have in my opinion. If Obama wins 52-48, he'll likely win with over 300 electoral votes. Seems convincing enough to me.