Running up against the clock

The McCain campaign has an extremely tough row to hoe this year.
The polls show them down well more than the margin of error in enough of the country that the real question is not 'will Obama win on Tuesday,' but 'will Obama win a landslide on Tuesday?' But the McCain campaign soldiers on, and that got me thinking over the weekend.

What would have to happen for this election to break for McCain? How could the Republicans win? I read fivethirtyeight closely, I've seen the pictures of empty McCain offices across all of the states that they absolutely must win if they have a shot at the Presidency. A moment of disclosure: I have both canvassed for, and donated money to, the Obama campaign this year. The McCain voters I talked to have been talking to their family about backing McCain. The Obama voters I talked to have been talking to everyone about backing Obama. Also, if anything I say below makes you disinclined to bother voting, whether because you feel that the race is already won or already lost, please vote anyway. Democracy works better when people vote. A huge part of what made the 2004 election break for Bush was that Bush knows how to run a get-out-the-vote campaign. The Obama organization is amazing. Not only are they enthusiastic and energetic, which is what I would expect political volunteers to be, but they know exactly what they need to do and are busy doing it.
But back to McCain. It looks like his get-out-the-vote effort is in nonexistent. Worse, enough states have had a significant period of early voting for this year, that a lot of the votes have already been cast. In some states, more votes have been cast already than were cast in 2004. So, the question is not only What could change now, but What could change now that would be sufficient to turn the Election?
And furthermore, What could be happening that we wouldn't be able to see in the polling?

• There could be massive, top down vote fraud going on. I say this first because someone will mention it if I don't. I think it's so unlikely as to be irrelevant. Even if you believe that somehow one campaign or the other culd do it while keeping it secret, to reverse the polls would require not only turning one or two states, but turning a half dozen. It's too hard. Even if there is a fair bit of retail voting irregularity, McCain needs to win all of the current battleground states. They need too many things to go their way for them to win by serious, widespread vote fraud. And on the Obama side of the aisle, the state attorney's general who've looked into allegations about people like Acorn are finding hundreds of bogus registrations, not the tens of thousands that would be required to tip the scales. And, to be fair, not a single instance of actual voting fraud. So let's leave fraud out of the conversation. Yes, there will be voting errors. We're collecting votes from north of a hundred million people, that's an awful lot of times to roll the dice. But there won't be so many of them that the outcome will change.
  
• People could be lying to the pollsters, either saying that they will back Obama (because they think that's the socially acceptable thing to say), or that they won't (because they think that that's the socially acceptable thing to say). The hitch there is, again, that the spread is so wide. The Bradley Effect has been considered as much as 3%. For the Election to go the other way, every pollster has to be being lied to by 3% of the population, and the polls have to have all been at the very limit of their margin of error, systematically under-reporting support for McCain, for the last month. This is unlikely as well.
  
• Obama supporters don't actually vote. This is a real potential problem. Although the early voting turnout certainly looks like a lot of Obama support turning out. But at this point, if McCain gets his voters out tomorrow, and the Obama supporters stay home because they think the race is already won, McCain could win.
  
• Which gets me to my final possibility. The McCain get-out-the-vote effort looks like it's not being very effective. But is there something they could be doing that they could see, but that people who aren't McCain supporters couldn't, that would move enough people to make pull out a McCain win? They seem to need almost all of the Republican voters in the battleground states to vote McCain, and almost all of the 'undecided' voters in the current polls to also vote McCain... doing that without contacting a fair number of Obama supporters along the way seems... difficult to believe. However, if the McCain campaign were running it's GOTV effort entirely by phone and on-line, then it might be possible for them to be doing really effective work without their campaign offices being open. [Ok, this is improbable - volunteers who all show up in an office, even if they're using their own cellphones to make calls, will play off of each other a lot more effectively than if they're sitting in the comfort of their own home calling at their own pace. But it's possible, and a lot of other things aren't]
  

This morning, I saw this Salon article. McCain is definitely doing the online phone banking thing. I don't know how well it's working, although from that article, it doesn't sound like it's working that well. That goose is looking awfully done.
That said, there's still one way that Obama can lose this election: If you're an Obama supporter, and you decide not to vote before the polls close tomorrow. This looks like it's going to set historical records in voting in the US. But at the end of the day, the outcome of the race will depend on who actually goes to the polls. So whoever you're planning to vote for tomorrow, get out there and do it.