With the election nearly upon us, I've started regularly following Talking Points Memo again. Late last night, Josh Marshall wrote about the early voting results:
Among those who've already voted, it's Obama 57%, McCain 38%. And that number is not inconsistent with numbers coming out of a lot of the early voting states.
CBS says that "about one in five voters" have voted early. I'm sure I'm just missing it, but I can't find it in the polling document. Meanwhile, Gallup says the number is now 27%, which I find astonishing.
Now, there are two ways of looking at these numbers. One possibility is that the big advantage Democrats are having among early voters is just a matter of regular Dems being so hyped up to get out to vote for Obama that they're disproportionately going early. That, or some variant of that argument, is the one being advanced by the McCain campaign. On the other hand, perhaps this is the actual 2008 electorate showing up at the polls and showing a big swing toward the Democrats.
Using these numbers, we can be a good deal more quantitative than this.
If you accept the polling numbers as accurate (a big if, but par for the course with this sort of thing), what would McCain need to do on election day to make up this deficit?
Well, if Obama has 57% of the early vote, then he's already locked up 11.4% of the total vote (57% of the 20% who have already voted) according to CBS, or 15.4% according to Gallup. McCain, on the other hand, has either 7.5% or 10.3%.
To get to a bare majority of 50.1%, McCain needs to get another 42.5% (or 39.8%), out of the remaining 80% (or 73%) of the voters. That means he needs to get 53.1% (or 54.5%) of the vote on election day (0.425/0.8 = 0.531). That's roughly the margin the elder George Bush had over Michael Dukakis, if you'd like a meaningless historical comparison.
Put another way, if we assumed that the real distribution was 50.1% to 49.9% in McCain's favor, then 22.8% (or 30.9%) of Obama's supporters have already voted, compared to only 15.2% (20.6%) of McCain's. This would require Obama supporters to be 1.5 times as likely to vote early as McCain supporters.
So, what does it all mean? Hell if I know. You could argue that a 50% increase in early-voting likelihood is reasonable given the hype about the Obama "ground game." You could probably make an equally plausible argument that it's way too high.
There's really no way to say for sure, until Tuesday night. But if you like numbers with your gun-jumping political commentary, there are some numbers for you.
- Log in to post comments
People like being on the winning side, hence voluntary Media quarantine of Election Day national voting results until West Coast polls close. McCain and Moose Jewel are deeply doomed. A Supreme Court challenge demanding disenfranchisement of early votes is in order. That, or a "lost" Russian nuclear field munition detonated (presumably a fizzle) during its stateless etc. terrorists' transport theough Inner City Detroit in about 12 hours (now 1320 hrs in California).
Save the Homeland!
(We're being robocalled by Republican panic. Goatse to all of them.)
I appreciate the fun of playing with numbers. But this is based on the wrong premise. McCain doesn't need a 50.1% majority. There are plenty of ways to win the electoral college without winning the popular vote, and I think most of the winning strategies left to McCain involve those scenarios.
I'm with becca on this one. In an electoral college system like ours the popular vote only amounts to a decently accurate indicator of who wins. That's not to say that those numbers aren't encouraging, but the proof is really in the pudding and individual states are the ones making the pudding; mostly the swing states at that.
Of course, if we had a national popular vote these numbers would actually matter.
While it's obviously true that the cockamamie Electoral College system means electoral votes are more important than popular vote, at this level of abstraction it's not a bad proxy. The "lose the popular vote, win the electoral vote" thing has happened what, twice?
4 times by my count. Jackson got a plurality of both the popular and electoral votes in 1824, but no one got a majority and it was decided in the house in favor of John Quincy Adams. Rutherford Hayes in 1876 and Benjamin Harrison in 1888 both lost the popular vote, but won the electoral vote. And, of course, Bush in 2000. Doesn't seem like much, but it's around 10% of past presidents, as long as we're slinging around more or less meaningless numbers.
The "lose the popular vote, win the electoral vote" thing has happened often enough that they track that possibility at FiveThirtyEight.com. As of this morning, they estimate a 1.02% probability that Obama will win this way, and a 1.11% probability that McCain will win this way. They also estimate an 0.21% chance of an electoral college tie.
Huh. What happens in the event of an electoral college tie? Anybody know?
Personally, while I think absentee balloting is important, this whole early voting thing (for no good reason) seems almost un-constitutional to me. I mean, why bother with an election day at all? It seems to detract from the importance of the campaign.
What happens in the event of an electoral college tie?
Any time none of the candidates gets an outright majority in the electoral college, the election goes to the House of Representatives, with each state delegation getting one vote to choose between the top two. This has happened twice: in 1800 when Thomas Jefferson and Aaron Burr tied (under the original constitution each elector got two votes, with the winner becoming president and second place becoming vice president; after this fiasco the Twelfth Amendment was adopted to distinguish presidential and vice presidential votes, to prevent this scenario from occurring again), and in 1824, as Eric #5 mentioned.