My Prediction for Tomorrow

Here's the problem I have with the polls.

The problem here is like the Clarence Thomas/Anita Hill thing. I struggled to imagine a scenario in which both people were telling the truth, where the disagreements were just of interpretation.

I couldn't. It was impossible. Someone was lying. It was very frustrating for me, because my first, and second, inclination is to imagine some scenario in which there are simply two versions of the truth, subject to the normal amount of fudging and perspective bias, but no one's flat-out lying.

Or, in the case of polls, simply flat-out wrong.

In this case, I've been trying to reconcile a national vote tie with Obama's persistent lead in Ohio and other swing states, which should in fact mirror the country. Ohio should be more Republican-leaning than the average.

As Brit Hume says, something's wrong here, and we don't know what.

What's wrong here are the huge partisan splits in the Democrats' advantage.

How can Obama be tied with Romney, in CNN's final poll, with Independents going +22 for Romney?

Oh, right: Because, CNN projects, Democrats will enjoy an unheard-of +11 advantage in votes.

I feel odd finally rejecting the state polls, and some national polls (D+11, indeed!) -- giving up on finding some way they could be telling a piece of the truth.

They're just wrong. I'm uncomfortable just saying data is wrong but that's where I come down.

One problem I have, though, is that many conservatives still seem to be making arguments based on numbers from 14 days ago, talking up a big Romney national vote lead. In fact, now, thanks to Obama doing a photo op (man, that guy is good at photo ops!), it's now tied.

Still, the assumptions for this tie are very strange. Apparently 2010 never happened, the country never rose up to reject socialism and failure, and 2008's Democratic plurality grew by leaps and bounds.

Apparently the continue majority support for repeal of ObamaCare somehow managed to increase the Democrats' popularity. Apparently the $6 trillion in new debt Obama signed into existence boosted his party's support.

Apparently presiding over a higher unemployment rate than any president ever re-elected (since FDR) has made the Democratic Party the dominant political force in the country.

Bill Clinton, apparently, had it all backwards. He presided over a furiously growing economy with (for a couple of quarters) a sub-4% unemployment rate but didn't manage to realign the country in the Democrats' favor.

But Barack Obama, by keeping unemployment at the 8% level (higher than the very high unemployment he started his term with) has won the hearts and imaginations of the nation for the Democratic Party.

That Clinton. What an idiot. He tried success. Didn't he know catastrophic failure in nearly every detail was the right approach?

Now, Team Obama explains that a growing, rather than shrinking, Democratic advantage in the electorate is possible because they've signed up 1.8 new voters. That's very wonderful.

Trouble is, a WaPo/ABCNews poll found that Obama had lost 9.1 million of his 2008 voters directly to Romney, with 3% more undecided (and they will probably break 1.5 to Romney, 1 stay home, .5 to Obama).

Now, Obama didn't just lose these voters. They're not just staying home. 9.1 million are flipping to Romney -- that's an 18.2 million net swing. Ten times the size of the new voters Obama brags he's signed up. Plus, using my back of the envelope guess as to the remaining 3% of undecided 2008 Obama voters-- that adds another net +3 to the shift to Romney.

So, in total, on election day, we'll find that of Obama's 2008 voters, there's been a net shift of 21 million to Mitt Romney. 10.5 million subtracted from Obama's column, and 10.5 million added to Team Red's.

But you know -- 1.8 million new voters registered.

So how on earth could the nation now be more Democratic than it was in 2008?

It can't be. And it isn't. I simply cannot believe that the 2010 repudiation never happened, that the parties-at-parity that year has not only reversed itself but rebounded +6, +8, and even +11 in the Democrats' direction.

So I have decided the state polls are simply wrong, and the national polls understate Romney's support by 3-4 points.

@benk84 presented this map in an email string; it turns out it was the same map I had made earlier today.

So here's my prediction: Romney 348, Obama 190.

Some other predictions I saw are fuzzing it, predicting a Romney squeaker by 271-267.

I find that scenario unlikely. If the polls are right, then Obama wins. Romney doesn't somehow squeak a hair's-breadth electoral win if the polls are right.

The only way Romney wins is if the toplines in the polls is wrong for whatever reason -- overweighting young and minority respondents, too few people responding to pollsters at all (now it's around 9-10%), and too many of the people responding to pollsters happy to offer the Socially Preferred answer.

The socially preferred answer is "I'm voting for Obama." After all, voting for Obama in 2008 didn't prove you weren't a racist. In 2008, Obama seemed like a reasonably good candidate (for those without any savvy or history or ideological underpinnings).

No, voting for the catastrophically disastrous Obama 2012 proves, beyond any doubt, that you're not a racist.

Even racists will hire a highly competent black man. But only the most anti-racist people in the world will re-hire the incompetent one, the one who seems to spend the bulk of his time golfing and chatting up Sir Paul McCartney.

So, once you've made that determination, that there's something simply wrong with these 2008 or higher Democratic splits and the toplines are just plain wrong, you have to just look at deeper numbers like Independents, Republican enthusiasm and solidarity, and Democratic defection rate.

As well as Democrats lagging their 2008 early-vote pace, and Republicans boosting their own.

And if you want to talk about ground game, let's talk about ORCA.

And all that forecasts a big Romney win.

Dixville Notch: I completely forgot about Dixville Notch, that New Hampshire sub-hamlet that always votes at 12:01 am on Election Day.

In 2008, Dixville projected Obama the winner, 15 for Obama, 6 for McCain.

This year? 5-5.

Ten Obama voters stayed home. Apparently one McCain voter did too.

(I'm now being told Dixville Notch only has 10 registered voters now. Well, still going from three to one Obama to a tie.)

Wisconsin: I've been thinking about Wisconsin. I think like this:

The Scott Walker recall was effectively a presidential-year turnout mobilization effort on both sides.

Now, Walker's opponent was not the most attractive candidate. Then, Barack Obama's not the most attractive candidate. He has a thing called a record, and it is woeful.

So if Wisconsin delivered a seven point win for a Republican candidate just this past June, why do people think Wisconsin "leans Democratic"?

It is true that some Democrats voted for Walker on the theory that the recall efforts just weren't fair.

Still, if you don't like a candidate, you don't really get into second-order questions like "what's fair?" for that candidate.

So Wisconsin just delivered a huge Republican victory to Walker (and before that, to Supreme Court Justice Prosser), but now, five months later, the state is so Democratic that it won't elect a strong and attractive Republican candidate (with a favorite son as the Veep candidate) over a failed Democratic one?

Of all the polls I don't believe, Wisconsin is the one I don't believe the hardest. This is a state which has gotten used to voting Republican and, I might say, gotten rather good at it.

But they're going to vote for Obama?

I just don't believe it. Honestly, I can see Obama winning Ohio. I have bought into The Narrative on Ohio -- autobailout, Obama's ads killed Romney for six months, blue-collar white males flipping over to the Socialist view of the world, etc.

But Wisconsin? If Ohio has been trending Democratic, certainly Wisconsin has been trending Republican.

What was Walker's plan in Wisconsin, by the way? Cut spending, balance the budget, reform the processes by which so much money goes out the door to schools and such, cut taxes.

And... it's working.

But they're going to look at Romney and Ryan and say, "Well, sure, we like when Walker does that on the state level, but on a national level, that's just crazy"?

I don't see it.

Oh, and in 2000, with the Democrats running on Clinton's record (albeit with Clinton's sex scandals), the state barely went Gore, and was only called for Gore days after the election.

But now they're like, "Oh yeah, we gots to have us more Obama"?

posted by Ace at 11:58 PM

| Access Comments