5

Did any political campaign in USA ever officially acknowledge practically using Nate Silver's (of 538 blog) models and/or numbers in some way, at least to check their own polls/predictons?

If it was just a campaign member who did it (as opposed to a policy of the whole campaign), it's in-scope as long as the practical effect was meaningful (e.g., it informed the decisions of a fairly-important campaign member).

6
  • Just to be clear, merely "I read 538" is not sufficient to be an answer. Admitting to letting what Nate's #s/model say influence campaign decisions is the cutoff. – user4012 Jan 30 '13 at 13:01
  • I'm unfamiliar with your use of "/#" etc. as punctuation - is it equivalent to e.g. "model(s)" (in the first use), or are "#s" and "model" two concepts you're separating with a slash? (if so, then I don't know what "#s" represents). – user97 Jan 30 '13 at 13:42
  • @ZeroPiraeus - meant "Models and/or numbers" :) – user4012 Jan 30 '13 at 15:13
  • BTW, the answer for "models" is likely "NO" since it's a proprietary product that Nate doesn't actually give access to, based on his reddit interview. – user4012 Jan 30 '13 at 15:15
  • 1
    @DA. - given how 2012 went, the answer - at least on R side - is most emphathically "No bloody way they use proper data". – user4012 May 20 '13 at 17:23
1

The best example I'm aware of of a campaign using Silver's numbers to validate their own polling is described in The Center Holds: Obama and his Enemies by journalist Jonathan Alter. It relates to Obama's 2012 re-election campaign, at the beginning of which, the New York Times Magazine published Silver's infamous Is Obama Toast? article which described the President as "a slight but not overwhelming underdog".

This article seems to have served as a reality check for Obama's campaign managers. Alter describes the reactions of Jim Messina & David Axelrod:

In Chicago Jim Messina practically choked on his breakfast. Obama was trailing the front-runner, Romney, by 3 or 4 points in national polls but looking better in battleground states. Messina fired off a rebuttal to the Times Magazine and vowed that after winning the election, his trophy would be a framed copy of the “17 percent chance” cover with the president’s signature. David Axelrod happened to be in St. Louis when he read Silver’s story. He noted that in September Silver had predicted that the St. Louis Cardinals had only a 1 percent chance of winning the World Series. In October, one strike away from elimination, they won it in seven.

While this example doesn't mention Silver's analysis informing specific campaign decisions, it seems likely that this event almost a year before polling day had an effect on Messina in particular, possibly influencing his choices over the campaign. Towards the end of the book, Alter mentions:

After the election Jim Messina had one favor to ask the president, which was easily granted. In mid-November he walked into the Oval Office, where Obama presented him with a signed and framed cover of the New York Times Magazine with Nate Silver’s “Is Obama Toast?” story.

The effect of this analysis on at least the emotions behind the campaign is backed up by Dan Balz in his book Collision 2012: The Future of Election Politics in a Divided America, in which he quotes Axelrod after Obama's victory speech:

“I felt the weight of the burden on [Obama] more than I felt the elation four years ago,” he said. “I know he believes this was a more satisfying win even than that one because of all the obstacles. A year ago the sainted Nate Silver was writing a magazine piece the headline of which was, ‘Is Obama Toast?’ To be written off and degraded and all that and come back and win . . .” He didn’t need to finish the sentence.

On a more concrete basis, it seems Silver's numbers were also used to validate private polling after Obama's less than stellar performance in the first presidential debate on October 3rd. Although the debate went poorly for Obama, his campaign was hopeful that it wouldn't swing votes away from Obama, and the effect would be limited to Romney winning back votes he had lost since the Democratic Convention. Alter describes the campaign being advised to "do the hardest thing in a tough situation: nothing". This decision was later validated by Silver's model.

When public polls and Nate Silver’s model showed Obama in trouble, Chicago continued to fret. Then, less than a week after Denver, David Shor, the former child prodigy in the Cave, dove into the latest numbers and told Kriegel, “Tomorrow, Nate Silver goes up.” Sure enough, the New York Times polling analyst, a security blanket for millions of Obama supporters, reported the next day what Chicago had already heard from the Cave and from Benenson: No big change.

The Golden Reports from Analytics found the race stable, with one exception: Numbers coming out of the Green Bay area showed Romney expanding his lead there from 2 points to between 6 and 9. This didn’t sound right, but the Cave’s models couldn’t be dismissed. The campaign bought more ad time and eventually sent both Obama and Bill Clinton into the region, even though it may not have been necessary.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .