Thursday, October 25, 2012

Quants Against Pros

We are in the end-zone where, historically, polling has been "pretty accurate." That is to say that if you believe the conspiracy theory all the cheating pollsters get serious in order to save their reputations and the reliable pollsters are close enough to the election to have a pretty clear view. There's that--and the theory that, given 13 days, not much can change so what it looks like now is, kinda, what it'll look like on Nov 6th.

But The Question Is: What Does It Look Like?
According to "the quants"--the guys doing the math based on the polls, Obama is ahead by either a comfortable margin (80-90% likely to win) or a decent one (70% if the election was held today). Looking at the math is moderately boggling--but even the true conservative Election Projection, which puts the candidates a mere 16 or so EV apart has Obama over the win-line.

On the other hand:
The Obama campaign is desperate and scared. Desperate and scared campaigns, like animals. tend to lash out. We’re seeing signs of that now with the “binders full of women” nonsense among other things. The Obama campaign has made a near fatal — quite possibly fatal — mistake. The campaign has been run on what Barack Obama has done for people without barely a hint of what he will do in his second term.
And here:
It’s a razor-close race in Ohio, but if Romney has knocked six points off of Obama’s 2008 gender gap and turned an eight-point deficit among independents into an eight-point advantage in a cycle where Democratic enthusiasm won’t come close to matching 2008, I have to think that the magic has already shifted to Romney.
And here:
Why would Obama be talking about alternate routes to victory not including Ohio?
Obviously, I think, they don't think Ohio is safe. And to talk about losing it... I think they think they're behind.
And so on. James Fallows (the quote from whom the title is taken) notes: They can't both be right!

So where are we? What are the polls showing?

What Do The Polls Say?
Here's the latest polls for Ohio:
Speak To Me Polls ... I'm Listening ...
As far as I can tell here it means that Obama is pretty much winning Ohio. I'm not doing any super math here or inspecting the cross-tabs or any of that nonsense but when I look in the margin I see either Obama or a tie--and for the polls from the 22nd onward, there's only one tie.

If it was really a down-the-middle split, wouldn't you expect some polls to fall to either side of the line? I mean, without having a degree in math or anything I'd expect at least some Romney +1 or +2's in there. Let's look at North Carolina which everyone agrees Romney is winning:
So this looks like a close state that Romney is clearly winning to me. There are some big Romney numbers (R+6) and some small ones (R+1) and a few Obama numbers (O+1) in there--but it's, yeah, mostly Romney.

Let's see Virginia:
Right--so this looks like a swing state that Obama's a little further ahead in than Romney. I mean, it's not all Obama. A lot of those Obama and Romney numbers are small and "close to the line" right? So okay.

What Did We Learn?
Well, we learned that the story that Obama is losing Ohio or that it's tied doesn't seem to be very well born out by the numbers. I'm going to throw some more chum in the water with this note: Rasmussen Has a Two Point Republican Bias Compared To Other Pollsters.
Rasmussen's mean bias is -1.91 points, that is, Rasmussen appears to be making Obama look almost 2 points worse than the other pollsters. ...
Note that this does not necessarily mean Rasmussen is wrong and the others are right. It could be that Rasmussen is right and the others are painting too rosy a picture for Obama. There is no way to know now.
It seems that right now Obama is up--definitively, if not by much--in Ohio and therefore is unlikely to be flailing. If Rasmussen is +2 for Romney (and I have no idea if that's true) then he's pretty much up in VA as well.

What Do I Think?
I think that the conspiratorial vein of thought that has taken hold right now (and there's some of it on the left--but more that I see on the right) is driving the Team-Obama-Is-Desperate-And-Flailing narrative. I suspect that's because Team Obama knows that posture is good for fundraising. I think that if nothing serious changes right now it will be hard for Romney to win and I don't know if he has momentum or not (the polls I see don't suggest a lot of momentum but: I'm not doing the math). However, I wouldn't put Romney's chances at 10% either.

No comments:

Post a Comment