campaign desk

Talking Shop: Nate Silver

FiveThirtyEight's founder on polling, politics, and the consolidation of the blogosphere
November 11, 2008

Nate Silver is the founder of, which predicted the presidential election’s popular vote outcome to within tenths of a percentage point (it projected 52.3 percent for Obama, who received 52.7; and it projected 46.2 percent for McCain, who received 46.0). Silver’s unique combination of accuracy with numbers and accessibility with narrative made FiveThirtyEight, founded in March 2008, the first blog ever to be selected as a Notable Narrative by Harvard’s Nieman Foundation.

In its feature about Silver in yesterday’s paper, The New York Times called the thirty-year-old Chicagoan “perhaps the most unlikely media star to emerge” out of “an election season of unlikely outcomes.” Silver has been profiled in the Wall Street Journal, Newsweek, Science News, and New York magazine, and has appeared on CNN, MSNBC, HDNet, WNYC, and Air America, among others. Silver’s newfound celebrity has led Gawker to offer him tailored advice on how he can continue to “rule the world,” Facebook members to found a group entitled “There’s a 97.3 Percent Chance That Nate Silver Is Totally My Boyfriend,” and several media outlets to refer to him, without irony, as a “wunderkind.”

CJR’s Megan Garber spoke with Silver about campaign coverage, celebrity, and his plans for the future.

Megan Garber: How did the media attention evolve—or did it come all at once?

Nate Silver: It was a gradual buildup. We got a lot of attention after the Indiana and North Carolina primaries, where we had said basically that, hey, the polls are wrong here. Obama’s going to win North Carolina definitively—it’s not going be close—and Indiana’s going to be really close, and Obama’s going to win it. And that turned out to be correct in both states, where a lot of the people had anticipated Obama barely squeaking by in North Carolina and getting thumped in Indiana. So that got us a lot of notoriety, because we had kind of stuck our necks out a little bit.

And it all kind of feeds on itself. One thing about media in general, and new media in particular, is that it’s not very linear—by which I mean it’s not that you go from one thousand hits in a day to two thousand—it’s, you know, from one thousand to ten thousand. And, likewise, there’s a parallel track in terms of interview requests and everything else. It kind of went viral, for lack of a better term, and all of the sudden, you’re really busy.

Sign up for CJR's daily email

MG: Are you enjoying it, though?

NS: Yeah, I think so—now that I’ve had the chance to relax a little bit. One thing that’s hard about the way we maintain the blog is that there wasn’t any time off. We had two or three posts going up every day, at a minimum, and some days six or seven or ten, if a lot was going on. And because it was just me and my co-author, Sean Quinn—we have a photographer, too, but he’s not doing writing for us—and a lot of times I was traveling, and whatever else, or just dead tired, or it felt like I’d been saying the same thing six days in a row—like, ‘Oh, it’s the convention bounce’—so there were times when it was a little bit of a chore. But for the most part, I’ve done a lot of fun things. I’ve, at various times, made a living writing about professional baseball, and made some income on the side playing poker. So I’ve done a lot of fun things. And this is definitely the most fun and fulfilling on balance.

MG: Did the TV appearances come naturally to you, or did you have to work at being an “on-air personality”?

NS: I wouldn’t say it came naturally. I think I went from being about a C on TV to being a B+ by the end. It’s definitely something where you need practice—it’s not completely natural in terms of your body language and other stuff. You want to look straight ahead, but you also don’t want to be so stiff. I think sometimes you learn lessons that are good starter rules, but kind of un-learn them eventually. ‘Don’t play with your hands a lot, because the frame’s going to capture your neckline and a little below, and you don’t want hands popping into and out of the frame.’ But if you gesture and gesticulate a lot—or a little bit, I should say—that can add some life and body, I think, to the appearance. So there are little things like that. I sometimes wear glasses on these appearances because my eyes tend to dart around a little bit, and you wouldn’t notice in person, but you’d certainly notice on television. So there are little tricks you learn.

By the end it was kind of fun. But there are also times when you’ve slept for all of three hours, and you have to do a TV hit, and you definitely hope they have a makeup person there, and whatever else. And other times you’re like, “This is fun,” and you’re in the right mood, and everything goes great. It’s definitely not something I anticipated doing when I started the Web site back in March.

MG: And what are your plans for later on? Are you going to stay in politics?

NS: Yeah, I think I’ll be trying to split my time in some reasonably intelligent way. But I’m looking to probably write a book next year (one good thing about the media appearances is you have publishers who are interested in your stuff). And, on the site, we’ll talk about the Congress. I think that, as compared with this point four years ago, there’ll be a lot more real news. Obama seems very ambitious about what he wants to accomplish. He has Democrats in both chambers of Congress, so he doesn’t have very many excuses not to get some stuff accomplished, so I think people will be interested in what he’s doing.

And there’s all this fascination over his chief of staff and stuff like that…and some of that stuff, I think, is kind of boring, but what I’m more interested in is the politics of it—Who are the key swing votes in the Senate? Who does Obama have to maintain good relationships with, and who can he afford to piss off?—and to kind of narrate that play-by-play, and everything else. Now, for example, which senators are up for re-election next year? There are a group of four or five moderate Republicans in tough races in 2010 in the Senate, and if Obama is popular, they’re going to have a difficult time if they look like they’re obstructing what he’s trying to do. Likewise, there are red-state Democratic senators who, if Obama is unpopular, might want to position themselves against him. So you’re going to go from a de-facto filibuster-proof majority of sixty-four or so senators if he’s popular to barely getting fifty on some votes if he’s not. There’s a group of about a dozen swing votes in the Senate between moderate Republicans and moderate Democrats of various stripes.

MG: While it focuses on Congress, will FiveThirtyEight keep its current numbers-and-narrative formula?

NS: Yeah. And we’re going to try and provide more data to people on the Congress, so you can look up someone’s voting record, for example, in a way that we think is more interesting and intelligible than you might be able to find elsewhere right now. Maybe it’d be something where, if you have a vote in the House, you can try to map out and model, ‘Why did people vote for this bill? Are there any people that look like they should have voted for this bill, and didn’t? And, if so, why didn’t they?’ And then maybe you tie that in with, say, lobbying money. So there’s a lot of creative ideas we have. It’ll never be horse race stuff, I don’t think. But we have a midterm in 2010—I think it’s going to be really interesting—and we have some gubernatorial elections next year, and there’ll be special elections, and stuff like that. The news tends to make itself. During the Clinton administration, when the Monica Lewinsky scandal broke, for example: We would have our ! own FiveThirtyEight way to cover that. Probably not talking about the gossip, but looking at his approval ratings and stuff like that. It’s a busy time in the world, and I think there’ll be no lack of things that we can lend our expertise to. Obviously, our bread and butter will probably be election years.

And, hey, there’s going to be a big fight going on in the GOP, as well. Just like you had a year-long Democratic primary this year, I think you’re really going to see a fight for the heart and soul of the Republican party, beginning early in 2011 and people positioning themselves in different ways. It’ll be fun, because, as we say on the Web site…I mean, I hope we have a reputation for being fair and balanced—maybe I shouldn’t use that particular phrase, but—I think it’ll be interesting, really, as a disinterested observer—and not really disinterested, I think it’s really interesting—to be able to cover that primary and say, ‘Who do I really think will win?’ I hope people can really trust my take if I say, ‘You know what? I think Mitt Romney’s really got it this year.’ I hope people can take that as authoritative and interesting when we get to 2011 or so.

We can also look at more everyday economic issues. Every time you pick up a newspaper, you can probably circle two or three items in every section where there’s some piece of quantitative or fiscal information reported that might not be reported all that smartly—so we can do a little of that, too. There was a book a few years ago, A Mathematician Reads the Newspaper, and we can make something like that a semi-regular feature: just things that amuse or annoy us about reporting—reporting in finance and medicine and certainly sports, which is part of my background, and pop culture—how it’s always like, ‘Well, the biggest box office gross…’ (well, but it’s not adjusted for inflation). Little things like that—the result of having a lot of English majors in the newsroom and not as many math majors.

MG: What’s your daily media diet? Do you have papers and sites that you read every day?

NS: A couple places. If you go to Daily Kos, you’ll just see things based on what people have linked to—sites that kind of play traffic cop, in a way. That’s some of what I do—or Huffington Post, or The Atlantic, or National Review. There’s another one called Memeorandum, where it’s all automated. So you look at those, and you see what people are talking about, and you don’t necessarily have to go to a million different places. I actually buy the paper version of The New York Times maybe once or twice a week, and the Trib or the Sun-Times maybe once a week. I try and read them over lunch. But for the most part, this stuff moves so fast, that you just kind of figure out what everyone’s talking about, and you just go there. And maybe at night, I’ll search over my larger list of links and see if there’s something that’s gone undiscovered. But, for the most part, by the time you hear about something, even if you’re constantly online, it’s kind of old news already.

MG: Is there anything that stands out in your mind as some the media did particularly well—or particularly badly—in covering the campaign overall?

NS: I wish the media had been a little bit less obsessed with race. I think the Bradley Effect got more attention than it deserved, probably. Sometimes they really jumped the gun. During the primaries, for example, some said that Hispanic people wouldn’t vote for Obama because of some race-based thing, and it turned out to be totally false. It was more about, number one, the Clintons are thought of very affectionately in the Latino community, and number two, it’s probably about economic class, and about whoever is capturing that working-class vote—no matter what their race—during the primaries. And during the general election, Obama won those voters over, in every region except Appalachia, basically. So I think people were too quick to reduce that to the race narrative when they didn’t have anything else to say, necessarily.

But in general, I think campaign reporting is a process that got a lot better, I think in part because of Web sites like ours, and people like Chuck Todd at MSNBC who are very good. Part of it is you have this whole big, long Democratic primary process where people realize, ‘Hey, it’s not about the popular vote,’ and different states have different rules for how they apportion their delegates. And by the way, the Obama campaign—their language is delegates; they’re not concerned about the popular vote. Their language during the general election was electoral votes and not the top-line popular vote number. So I think they kind of forced people to think about things in that way, eventually.

MG: You mention Chuck Todd. I’d love to be a fly on the wall during a conversation between you two—do you know each other personally?

NS: No, not really. I’d guess that we have a lot of respect for one another. At some point I should get his phone number from one of my contacts at MSNBC and say, ‘Hey, Chuck, let’s get a beer.’ But I think we have people that are implicitly—Andrew Sullivan, for example, at The Atlantic, has linked to us a ton of times. I’ve never had a conversation with him, but I think there people where it’s just kind of implicit: ‘Hey, we respect what you’re doing, we’re going to help you guys out.’ And vice versa. I think one other trend we saw this year is, to some extent, the consolidation of the blogosphere. I think you saw more traffic going to a smaller number of sites, whether it’s a Daily Kos or a FiveThirtyEight—or, on the other side of the spectrum, a site like the National Review or something like that. There are a certain number of go-to destinations for political coverage, and those sites, I think, have a lot of influence. Some of them are old, some of them are new—like us, or Talking Points Memo or something like that—but we’re seeing what I’d call the maturation of it, where it’s not just a million monkeys with a keyboard, it’s a hundred monkeys with a keyboard. And there’s kind of a selection process for who’s the most reliable in terms of timeliness and everything.

MG: Do you see that kind of consolidation as a simple meritocracy, or is it more complex than that?

NS: It’s mostly a meritocracy. Sometimes there are sites that take a while to get noticed. There’s one right-leaning site, The Next Right, which I think is terrific, though it’s not my political point of view. And they’re a site that, once people start to notice them, I think will be taken very seriously. Every now and then you catch a diamond in the rough, where you know it’s just a matter of time before people discover them. So there are a few inefficiencies at first. But I think it’s pretty meritocratic relative to other things, by and large—though maybe not perfectly so. Running a site, there may be a couple days when you have a million things going on, and you’re stressed about some real-life thing, or you’re sick, or something. But you have to maintain the quality: don’t dilute the brand.

Brand is really important in Internet media. Because there’s so much competition, and the barriers to entry are so low, all that you are, really, is your brand. And if you spread yourself too thin, then you’re making a real mistake, I think. Our model, certainly, is to cover things in depth, and not to try and do everything. Maybe we’ll try and have someone in Washington covering the White House. But if we do, we’ll want to make sure we do it really well.

Update: The popular vote breakdown listed in the introduction to this article has been amended to reflect the AP’s current numbers (as of November 12).

Megan Garber is an assistant editor at the Nieman Journalism Lab at Harvard University. She was formerly a CJR staff writer.