Six Questions With Nate Silver
From Kenyon News - May 15, 2018
The 190th Commencement was held Saturday, May 19, at 10:30 a.m. Click here to watch video and read the transcript of Nate Silver’s Commencement address.
Acclaimed statistician and writer Nate Silver will address the Class of 2018 on Saturday, May 19. Silver, the founder and editor of the website FiveThirtyEight, has gained fame for his exceptionally accurate and occasionally counterintuitive political prognostications, especially his predictions for the 2008 and 2012 presidential elections.
Prior to his work with FiveThirtyEight, Silver was known for his prowess with baseball analytics. He is the author of a series of books on baseball statistics as well as the 2012 best-seller “The Signal and the Noise: Why So Many Predictions Fail — But Some Don’t.” Silver, who earned his bachelor’s in economics from the University of Chicago, will receive an honorary doctorate in humane letters from Kenyon.
The New York Times recently described Ohio as “a purple state shading red.” Do you think Ohio will remain an important swing state, or will the battlegrounds move elsewhere?
I do think Trump had a particular appeal, perhaps, in Ohio, and that Clinton had a particular lack of appeal, perhaps, and maybe [the outcome in 2016] was a little bit exaggerated. But certainly, if you look at the demographic shifts in the country, where Democrats are increasingly concentrated among college-educated people on the coasts, and not among working-class white voters, I think some of that’s going to hold. Ohio might be one of these states that Democrats can still win when they’re having a good year but that, in a year where things are truly 50/50, is not necessarily the swingiest of swing states.
Kenyon is located in the Ohio 7th District, represented by Republican Bob Gibbs. Do you think that seat is safe, or could Democratic challenger Ken Harbaugh conceivably win it this fall?
It’s one of those districts that would ordinarily be safe, but it’s not that far removed in terms of what the voting patterns have been from, for example, the Pennsylvania 18th District, where Conor Lamb won. So, in a nightmarish year for Republicans, that district could become competitive, potentially. But if they were to lose the Ohio 7th, then that’s the kind of scenario where they probably are losing 60 or 70 seats. Which is not impossible, but because of the way that districts are drawn and gerrymandered, including in Ohio, there’s a big cliff that Democrats have to overcome. [Losing the seat is] plausible enough that [Gibbs] should be running a campaign, but Republicans probably have many other things that are more immediate concerns.
You’ve applied your skills as a statistician not just to politics and election forecasting, but also to baseball analysis and even online poker. How do you balance these interests?
I have fun with the sports stuff, and then I feel obligated to cover the election stuff, is one way to put it [laughs]. If I’m sitting home by myself or with a friend at 9 p.m. on a Tuesday, I’m watching sports, I’m not watching MSNBC or CNN or Fox News.
I’m not by nature a politics guy. I don’t necessarily like all the ritual around politics and I don’t particularly like how politics are covered in the media. That’s part of the reason for FiveThirtyEight. But, it does happen to be a time when it’s a fascinating field to cover. I’ve been blessed in the sense that we’ve had a lot of really interesting elections in a row: 2016 was fascinating, 2008 was fascinating, 2018 is going to be fascinating, who knows what’s in store for 2020.
Do you not like it when people call you a pundit, because you don’t want to be associated with those political rituals?
Yeah, we think of pundits as the enemy. We think pundits are lazy and don’t actually do any analysis. The whole point is, you should show your work — you should not rely on gut instinct, you should rely on evidence. I think of punditry as an evidence-free zone.
What skills have you found necessary to achieve your work, from analyzing baseball stats to predicting political outcomes?
Certainly you need some technical training, although I think people can often train themselves. Secondly, you need some domain knowledge; if you don’t know anything about how baseball works, then you’re going to make a lot of dumb assumptions when you build models and formulas. And third, you need a lot of practice, because this stuff is actually hard and you will get stuff wrong, especially if you’re inexperienced. Just kind of having that attitude to build models and experiment with things, and tinker with things, and fix things, and having a lot of curiosity about the problem you’re trying to study.
I was fortunate enough to go to the University of Chicago, and having that liberal arts background helped a lot because it’s about how do you ask good questions, how do you scrutinize evidence, how do you develop a thesis. That stuff is really key, because there are a lot of people who are technically trained who have no idea what they’re doing, and that frankly can be a little bit, or a lot, dangerous.
Do you have any words of comfort for fans of the Cincinnati Reds, who are currently at the bottom of the National League Central standings?
Well, if the Cubs can win a World Series ... in baseball, things do turn around. It can take five years, but it’s not hopeless. A team begins to rebuild its farm system and it gets better draft picks, and five years from now, the Reds are as likely to be good as anyone else, roughly speaking. This year, we already have them projected for 100 losses, which is rare this early in the year. But if you’re not going to make the playoffs, you might as well lose by as much as possible and get the high draft picks.
This interview has been edited for length and clarity.