Thanks for following our live blog of Nate Silver’s talk at MIT. We’ll have a full article in tomorrow’s issue of The Tech as well as photos and a video of the talk.
6:56 p.m. — Silver on Snowfall: “I think it’s pretty amazing.”
6:52 p.m. — Silver on digital journalism:
1) “Visual presentation at the Times is really outstanding. It’s super important. They understand that part of what you’re doing is taking complex information and finding a way to convey the sense of the information that is accurate.”
2) Digital first mentality for breaking news: For example, the homepage showed the Twitter feed of the bureau chief in Rome. We’re not going to hold a story to wait on the print edition.
3) People have their own brands on the Times. People follow individual writers more than the brand.
6:50 p.m. — Tips for science journalists — Have respect for the reader. “If you’re not communicating something to the reader, then it’s often your fault. If you have trouble articulating what an argument is, maybe it’s not a good argument to begin with. One thing that helps me a lot is being a sum of a practitioner and being a journalist.” Silver says that people should be skeptical of flashy conclusions.
6:41 p.m. —” To be competitive, you have to know what your competition is doing and what they’re doing wrong. I expect that space to get a lot more competitive.” — Silver
6:39 p.m. — Question: Do you worry about whether other people will catch on where all media outlets will have their own Nate Silver?
Answer: There are already a number of models that are pretty good. They had 98% of the same DNA as FiveThirtyEight. If I’m competing against the Bloomberg model, the Washington Post model, it’s not going to do that much better over the long term. I like the competition against the mainstream pundits who are terrible at what they do.”
6:24 p.m. — “Urban data is another case where you have a dataset that is somewhat rich but isn’t being used all that well. At some point, I’ll have the temptation to dive into a big new dataset.” — Silver
6:22 p.m. — Silver jokes about being against Puerto Rico getting a vote in the electoral college so he doesn’t have to change the name of his blog to FiveThirtyNine, which is already a registered domain name.
6:13 p.m. — “Sometimes news organizations think about how many page views something gets in the short term, and not what it says about your brand in the longer term.” – Silver
“It’s easy to compete with stupid. How can you be differentiating is the question I would ask.” — Silver
6:09 p.m. — John Hawkinson asks about open sourcing his model.
“I’m very sympathetic to the academic viewpoint that it should be open source. But I don’t have a tenure track postion. If you open source something, that value is diminished. I would never ever rule out doing it. I do think we can reach a better level of disclosure when things are more explicit.” — Silver
6:03 p.m. — Ian Condry, CMS professor at MIT, asks about political reporting and media bias. He wonders if there is a danger of “too much accuracy in predictive power.”
Silver says that if he feels that FiveThirtyEight is affecting voting turnout, that would be a dangerous precedent. He gives an example about Tim Pawlenty pulling out of the campaign after press about being behind in polls, less donations, etc… and it became a self-fulfilling prophecy.
6:02 p.m. — Q&A has started.
6:00 p.m. — Silver talks about media bias in stories. “I think that information feedback loops are becoming a bigger part of the story.” Silver says that only 5% of viewers who watch Sean Hannity on a given night is democrat, and only 2% of Rachel Maddow viewers are democrat. “I’d like to go back and see what polls were cited by the media and when.”
5:53 p.m. — “There’s more interest in data science now and big data.” Silver talks about how the amount of data isn’t the constraint in data science; it is the analysis.
5:49 p.m. — “I spend a lot of time on the graphical parts of my presentation.” — Silver
5:48 p.m. — “Everyone comes after the Times. It’s the New York Yankees basically. The less obvious downside of that is that sometimes it’s hard to be kind of casual at the Times.” Silver says that with a blog, you can be “farting around.” But at the Times you can get in more trouble for that kind of thing. People treat it as more authoritative, so it’s harder to find that voice.
5:46 p.m. — Silver’s contract at the Times comes up this summer. He signed a 3 year contract in 2010. “I’m in active discussions with the Times. It’s a great fit in a lot of ways. Jill is a perfect editor. Anything can happen in negotiation. But we’ll see. I’m pretty happy there.” — Silver
5:45 p.m. — Silver says that he wouldn’t have doubted his formula much if things didn’t occur as he predicted. Silver says that all the models converge at the end of the race.
“The fact that Obama won the election didn’t occur to me for two days after the fact.” Silver says he was so involved in the psychodrama of the election that he was “in a bubble” and the results of the elections didn’t dawn on him right away.
5:41 p.m. — “Did you get all 50 states right this time?” “Yeah.”
5:35 p.m. — There was 50% more traffic when he switched FiveThirtyEight to the Times. He said that when he gets on the front page, the peaks are a lot higher, but on the slow days, the traffic is about the same as it was before FiveThirtyEight was on the Times. On his first day at the Times, he got 500 views. After a profile in Newsweek, it went to 5,000. On election day, he got 3 million page views.
5:31 p.m. — Silver says that he had offers from multiple places to put FiveThirtyEight, not just the New York Times.
5:28 p.m. — “When you start writing for the New York Times, you do end up being a bigger target.” — Silver
5:26 p.m. — Silver talks about the beginnings of FiveThirtyEight. He wrote under the name “Poblano” (“I always liked Mexican food.”), and wrote daily for Daily Kos, first anonymously. He went public because he wanted to capitalize on it and also potentially shift careers. He had to “out himself.”
5:23 p.m. — “I get all sorts of spam from campaigns.” — Silver
5:18 p.m. — On the development of FiveThirtyEight: Silver followed the ins and outs of the law that affected online poker in 2006. In 2007, Silver found that politics was still “stuck in the stone age and not data driven at all.”
5:17 p.m. — “I wouldn’t find it all that attractive to work for a [baseball] team, because I’d like to analyze data where you’re your own boss and you can share your conclusions with the entire public.” — Silver, on his work at Baseball Prospectus
5:11 p.m. — Question: Did your parents or friends think you were insane from leaving your consulting job?
Answer: Online, it was possible to make pretty good money for a couple of years. So that let me leave my job.
Question: So you can’t play poker online now because…
Answer: The U.S. government passed a law in 2006 that made it very hard to get money from these sites. The whole ecology/ecosystem collapsed.
5:08 p.m. — Silver explains Pecota and what made it different. Silver explains that one of the innovations was that it could “capture the range of forecasts.” It also had the “world’s best datasets.” One thing Silver tries to do is frame terms of probabilities and show the intermediate steps to get to those probabilities.
5:07 p.m. — “I went to USC, where fun goes to die.” — Silver
5:06 p.m. — “It’s a real honor to be here. It’s a Ground Zero for nerds.” — Silver
5:05 p.m. — “Standing room only! Nerds nerding out”
5:00 p.m. — Director of the Communications Forum David Thorburn introduces Nate Silver and moderator Seth Mnookin.
4:52 p.m. — The organizers just announced that we would start “promptly at 5 p.m.” Seats are filling quickly in Bartos Theater.
4:45 p.m. — Nate Silver and Seth Mnookin just arrived.
The Tech will live blog the talk here starting at 5 p.m.!