Nate Silver wins, and data is vindicated

The Huffington Post | Salon | Forbes | Newsday | Slate | Time | The Washington Examiner | The New York Times
New York Times polling blogger Nate Silver appears to have called every state correctly Tuesday after being needled for weeks by political commentators who ridiculed his method for forecasting election results.

Salon’s Jacob Sugarman posts two vindicatory maps. (If you prefer, you can view Christoph Niemann’s M&M-based map.) Jeff Bercovici attempted to quantify Silver’s victory: “There’s a 90.9% chance that tonight is going to cap the best three weeks of Nate Silver’s life,” he wrote before returns started coming in Tuesday. Silver “got the race right because he analyzed the data properly and honestly, something his critics were unable, or worse, unwilling, to do,” Lane Filler writes in Newsday.

Silver’s accuracy reflects better on the science of polling, Daniel Engber argues in Slate. Silver’s fans, he writes, are “giving credit to the bearer of good news. In doing so, they’ve made the same mistake that Silver’s critics made last week: They’ve confused his projected odds with hard-and-fast predictions, and underestimated the accuracy of polling.”

Paul Bradshaw writes that Silver’s journalist critics are an “embarrassment to the profession.”

Journalists who professed to be political experts were shown to be well connected, well-informed perhaps, but – on the thing that ultimately decided the result: how people were planning to vote – not well educated. They were left reporting opinions, while Nate Silver and others reported research.

Similarly, in what won’t be the first story about how Obama won, Michael Scherer profiles his campaign’s data operation.

“We ran the election 66,000 times every night,” said a senior official, describing the computer simulations the campaign ran to figure out Obama’s odds of winning each swing state. “And every morning we got the spit-out — here are your chances of winning these states. And that is how we allocated resources.”

In contrast, Byron York writes that Romney’s data initiative, “Project Orca,” was a flop.

Orca, which was headquartered in a giant war room spread across the floor of the Boston Garden, turned out to be problematic at best. Early in the evening, one aide said that, as of 4 p.m., Orca still projected a Romney victory of somewhere between 290 and 300 electoral votes. Obviously that didn’t happen. Later, another aide said Orca had pretty much crashed in the heat of the action. “Somebody said Orca is lying on the beach with a harpoon in it,” said the aide.

Dylan Byers, who crusaded against Silver in various ways recently, tweeted a concession early Wednesday:

Silver didn’t gloat after the race was called exactly as he’d forecast (“we’re going to get some sleep and grab a beer,” he wrote), though he did link to his new book.

Silver talked about Obama’s win last night with Times blogs editor Megan Liberman. The subject of gambling came up and was quickly dismissed.

Related: 71 percent of politics visits to The New York Times last week “included a stop at FiveThirtyEight,” Marc Tracy writes in the New Republic. Don’t read that as saying Silver generated 20 percent of The Times’ traffic, reminds Peter Kafka. Numbers, people!

More Silver (and Silver-ish) stuff: Pundits’ 2012 predictions, visualized as darts on a target (Slate) | Silver was “not the only one who really nailed it” (The Atlantic) | A very good pundit scorecard (The Atlantic Wire) | Michael Barone: “please be assured that I will be on a diet of crow for some time” (The Washington Examiner) | The psychedelic poetry of a Silver parody Twitter account (The Washington Post) | Ezra Klein on Twitter: “The greatest trick Nate Silver ever pulled was becoming the face of aggregated polling

We have made it easy to comment on posts, however we require civility and encourage full names to that end (first initial, last name is OK). Please read our guidelines here before commenting.

  • Lauren Mayer

    Nate Silver is smart, calm, and impressive – he makes statistical analysis cool and has probably inspired scores of kids to pay attention in math class. Plus he’s adorable. This may be the first love song to a psephologist (worth looking up, it’ll come in handy), but it won’t be the last!

  • Robert Knilands

    The real story in this context. And the journalist reference comes straight from the segment about Bradshaw’s report. But diving into semantics is a good way to avoid seeking improvement — another primary goal of today’s JOURNALISTS.

  • Anonymous

    well, that probably isn’t the “real story,” but it is some of the story. romney had gop pros running his campaign, whether they ere outmaneuvered or not. and mutt had a pot full of money, which doesnt even count the incredible resources of his related super pacs. also, i think you are somewhat confusing “commentators/pundits” with “journalists.” there are similarities to be sure, but there ALSO is a difference.

  • Robert Knilands

    To me, this is like being the best buggy-whip maker of 2012. Short of exposing many of today’s journalists as lazy, opinion-based, and fact-myopic — something many people already knew if they were paying attention — what does this “victory” really accomplish?
    The real story is Obama’s campaign knowing where to allocate resources and Romney’s campaign apparently being confused. That concept is far more important than what some blogger predicted.

  • Barry Hollander

    As a number cruncher, I’m glad to see the models of Silver, et al., vindicated. Political pundits take anecdotes and turn them into data. The good news? Pundits have four years to go and take some classes in statistics and probability before they make fools of themselves again. Oh, and here’s a neat list of how the pundits called the election:

  • Andrew Beaujon

    I have written about that! But also the headline uses the term as a collective noun, which AP says to go singular on.

  • Anonymous

    data = plural. Just one example of many of how journos fail at using data.

  • Anonymous

    data = plural. Just one example of many of how journos fail at using data.