The Washington Post

My election prediction: The polls will be right and Obama will win with 290 electoral votes

"Watching the conservative pushback on the polls reminds me of this slender volume I bought in Oct 1972," e-mailed a friend today. "The triumph of hope over analysis is not a new phenomenon."

The "slender volume" is "How McGovern Won the Presidency and Why the Polls Were Wrong." The book, published in advance of the 1972 election, purported to explain why George McGovern, then trailing by a huge margin in the polls, was nevertheless going to win the election. In "Nixonland," his history of the period, Rick Perlstein summarizes the argument:

The shallow attachment voters had with the president would be overwhelmed by the passion of the McGovernites. The reports on the souring of the youth vote would turn out to be wrong. The polls didn't know how to adequately sample these new, young first-time voters. They wouldn't reflect the millions of Americas too poor to have phone service. They didn't factor in the love for an honest man, Maybe people wouldn't say they were for McGovern, but that was only testament to the president's success turning the name McGovern into something you were supposed to be ashamed of. America still had a secret ballot. In their hearts, they would know who was right. Pat Caddell reported that when they gave homes they were polling sealed ballots, McGovern did 9 percent better.

You can hear shades of these arguments even today. The Romney team argues that the polls don't know how to correctly count young voters, and so are oversampling them because so many turned out in 2008. Obama's supporters argue that the polls haven't caught up with the shift from landlines to cell phones, and when you adjust for that, Obama's lead is far larger than the polls suggest. The Romney team argues that Americans don't want to tell pollsters they're going to vote against the first African-American president after one term, but when they get into the ballot box, they're going to do it anyway.

I also heard these arguments in 2004. Then it was Democrats, centered around the Donkey Rising blog, trying to convince themselves (and the media) that the polls were wrong and Kerry would win the election. In the end, the polls were, as usual, right, and the hopeful partisans trying to convince themselves that the polls happened to be biased against them were wrong.

I have a simple rule when predicting presidential elections: The polls, taken together, are typically pretty accurate. Systemic problems, while possible, aren't likely. There are a lot of pollsters producing a lot of polls and each and every one of them has every incentive to try and get it right. When they converge, it's typically with good reason. And right now, they have converged. The 3-4 percentage point error necessary for Romney to be the real favorite in this race is extremely unlikely.

Critics of the polls, meanwhile, tend to be self-serving. You don't hear Romney supporters arguing that the cell phone users really are undersampled, and their candidate's position is even more dire than it seems. Similarly, no prominent Obama supporters have argued that the polls are assuming an electorate that looks too much like 2008, and as such, are undercounting Romney's likely support. Frankly, I'd be much more likely to take a critique of the polls seriously if it cut against the critic's self-interest. But somehow, it never does. 

Self-serving critiques can be correct, of course, but because they're motivated by what the critic wishes to see happen, or what the flack needs the media to think is happening, they deserve to be treated with suspicion. And because the pollsters themselves have clear incentives, more expertise, and a deeper understanding of the data, I tend to give them the benefit of the doubt.

So here's my prediction for tomorrow: The polls will prove to be right. President Obama will win with 290 electoral votes. I'm not extremely confident in the precision of that estimate: Some swing states are close enough that it's entirely possible for a good ground game to tip, say, Florida into Obama's column, or Colorado into Romney's. Virginia is basically tied, and I'm giving it to Romney based on the assumption that challenger wins in a tie, but it could easily go the other way. So if Obama ends up winning with 303, I won't be surprised.

That said, 290 is what a conservative read of the polls says, as of this moment. And I trust the polls more than I trust my intuition, or the fragmented, impressionistic reporting on the two GOTV efforts. So I'm going with that.



Success! Check your inbox for details. You might also like:

Please enter a valid email address

See all newsletters

Show Comments
Most Read



Success! Check your inbox for details.

See all newsletters

Your Three. Videos curated for you.
Play Videos
How to make Sean Brock's 'Heritage' cornbread
New limbs for Pakistani soldiers
The signature dish of Charleston, S.C.
Play Videos
Why seasonal allergies make you miserable
John Lewis, 'Marv the Barb' and the politics of barber shops
What you need to know about filming the police
Play Videos
The Post taste tests Pizza Hut's new hot dog pizza
5 tips for using your thermostat
Michael Bolton's cinematic serenade to Detroit
Play Videos
Full disclosure: 3 bedrooms, 2 baths, 1 ghoul
Pandas, from birth to milk to mom
The signature drink of New Orleans
Next Story
Dylan Matthews · November 5, 2012

To keep reading, please enter your email address.

You’ll also receive from The Washington Post:
  • A free 6-week digital subscription
  • Our daily newsletter in your inbox

Please enter a valid email address

I have read and agree to the Terms of Service and Privacy Policy.

Please indicate agreement.

Thank you.

Check your inbox. We’ve sent an email explaining how to set up an account and activate your free digital subscription.