Given the latest polls, it is highly unlikely for McCain to win. But not impossible.
I’m trying to enumerate all the scenarios which might lead to a McCain victory, and try to assign the likelihood of each
The polls are wrong because their methodology is wrong (e.g. they are not selecting “likely voters” correctly)
Likelihood: low. It has happened before. I seem to recall a case in England in the late eighties, where the polls got it completely wrong, and it was very embarrassing for the major pollsters
The polls are wrong because of the “Bradley effect”, i.e. people are lying about supporting Obama so as not to appear racist.
Likelihood: low. I don’t see why someone can’t just say they are voting for McCain for his policies, even if the real reason were racism. Why the need to say Obama?
The polls are right, but people have a last minute shift in preferences
Likelihood: very low. Even if some change their minds, the gap is large enough that it would take a large chunk of people to change their minds to make a difference in who wins.
What do you guys think of the above scenarios?
Any other scenarios not mentioned above?
If the polls are wrong they are likely wrong in favor of Obama.
Young people disappoint and stay home on the couch playing Nintendo instead of turning out in force. Conservatives, scared at what they think will be a mountain of Obama supporter turnout come out in droves themselves and tip the balance.
Likelihood: low. Obama’s GOTV machine is unprecedented in American elections. The young vote has been increasing (IIRC) the last three elections. The young vote seems more engaged than ever in this one. I suspect they will be (more) significant this time.
Early 90’s. The 1992 General Election in fact. A friend of mine who worked for the BBC political department went around in his BBC Election 1992 t-shirt with the word “Oooops” written in underneath the logo. There was a lot of talk at the time that the Conservative vote was underestimated because people were ashamed of admitting they supported the Conservatives.
Anyway, it does seem likely the likely voter models are off this time. But that won’t necessarily help Mccain. Cell phone only voters are underpolled, as are newly registered voters.
It’s odd to me that newly registered voters are considered unlikely to vote. It seems more intuitive to believe that they are enthusiastic enough about this election to bother registering in the first place, and would therefor be more likely to vote, not less. But I guess historically that isn’t the case. Just seems strange to me.
Fraud
Likelihood: low. Hard to quantify, but the existence of electronic voting without a paper trail is prime ground for tampering. Having one of the main companies, Diebold, being run by a Republican is not reassuring
I think vote suppression’s going to have a significant effect. Stories are already emerging about Republican scare tactics, registrations not taking, and other funny business. I think OBama may do shockingly poorly in states with a strong GOP presence.
I still pick him to win 306-232, but that’s a tighter margin than most predictions.
I don’t think ‘people lying’ is necessarily what is meant by the Bradley Effect. It also might mean that some people stated a genuine preference for the black candidate when polled (perhaps due to altruism), but changed their mind at the last minute (perhaps out of fear or because they hadn’t given it serious thought before). I could see this being a big factor tomorrow.
My boss is. He’s convinced that Obama is going to bring terrorists into the cabinet, get rid of all guns, tax the hell out of everyone, and institute a welfare state.
Me…I’m so nervous about this election I’m giving myself an ulcer. I think it’s because Obama’s lead is so large that I’m afraid of that last minute hail Mary upset.
I just feel that Obama has the potential to be one of those truly great Presidents…one of the rare breed that only come along every half century or so. And I am REALLY sick of 8 years of Republican rule. I’m sick of ‘liberal’ being a bad word. Go Obama Go!
On the other hand, the Obama ground game is doing a very impressive job of defending against such sleaze tactics. In 2004, folks found out about “Democrats vote on Wednesday” schemes when they found the fliers on their doorsteps in the morning, but this year, we’re finding out about them when Obama canvassers are seeing (and removing) them on their regularly-scheduled rounds.
I don’t think it is that they are necessarily less likely to vote. They are, however, less likely to be polled, because they might not show up on the list that the polling company is using.
It’s seemed to me for a few weeks that barring shenanigans, McCain’s only route to victory was to win all of the following: NV, CO, MO, OH, VA, NC, and FL. None of those is an impossible win for McCain individually; Bush the younger won them all twice. And if he carries them all and doesn’t make any ridiculous fumbles elsewhere (Arizona, snicker) then he’ll get to 274.
Winning all of them, however…that’s the trick. He’ll probably get some of them, MO for sure, and I still have a hard time believing that Obama is leading the polls in VA and NC. But all of them? That’s a tough row to hoe.
By my math, McCain needs EVERY toss-up state (not so unlikely…he leads in most), Virginia especially, and PA. PA’s going to be the big one, and he can’t get it. Perhaps MN is more likely, but it’s a stretch. To be clear, he needs to get AZ, MT, ND, MO, GA, NC, IN, OH, FL, and either MN or PA, in order of increasing difficulty. I think he’ll stop at NC or IN. OH, FL, PA, and MN are going the other way.
ETA: If he gets OH, IN, and FL, 2 out of NV-CO-NM would work too. That puts him at 270 or 274.
