- Nate Silver is famous for his election predictions.
- Over the last year, he predicted Democrats would lose if Joe Biden was at the top of the ticket.
- Now that Biden is out, he sees the race as a toss-up, with Kamala Harris having a slight edge.
Nate Silver may be the best-known oddsmaker in the world: He first rose to fame for correctly predicting almost all of the 2008 election and repeating the feat again in 2012. And when Hillary Clinton lost the 2016 election, after Silver gave her a 71% chance of winning, he became the subject of much (misplaced) scorn.
Now Silver has a new book — On the Edge: The Art of Risking Everything — which focuses on people like himself, who make a living by calculating odds and placing bets. That includes professional gamblers, as well as entrepreneurs, venture capitalists, and crypto speculators. (Silver, by the way, is a big gambler, too: He's a serious poker player, and also spent $1.8 million over a year betting on NBA games, which netted him a grand profit of $5,242.)
Silver wants his book to teach you about risk-taking and how it applies to the likes of both Sam Bankman-Fried and Sam Altman — and why he thinks people in fields like government, academia, and media are uncomfortable with risk.
But the rest of the world wants to know who Silver thinks is going to win the 2024 election. He scratches that itch by running a Substack that offers detailed models to subscribers and more general updates to freeloaders.
We talked about all of that in a recent conversation. But I'm playing the odds here, too, by focusing these edited excerpts on his election thoughts. Plus a look back, and forward, on his career.
I know you don't want to be known as "the political forecast guy" but I do want to know how you see the election. Your most recent projection has a Trump-Harris toss-up, though Kamala Harris seems to have momentum. You've also written that normally you'd expect her recent surge to stall, but given how close we are to the election, that might not happen. How are you thinking about it right now?
If it were November [Silver's election model] would be really aggressive. You'd say Harris is moving up, and there's not enough time for reversal.
Since it's just August, it's more cautious, because we've already seen big momentum swings.
I think it's reasonably safe to assume that the next several weeks will look good for her in the polls. She's having her convention, and Harris and Tim Walz are relative unknowns, so it might be a little bit more exciting than usual.
There's usually a bounce to the polls after the convention. So she might get up ahead 3, 4, 5 points in national polls. And then September and the debate, or debates, if they're plural, is the reality check.
But yeah. Even I thought she'd be an underdog, just less of an underdog than Biden would have been. Now I think you'd probably rather be in her position.
It's pretty close. It's no longer Trump's election to lose. It's someone's to seize a moment and grab.
Long before the Biden-Trump debate — going back at least a year — you were consistently arguing that Biden wasn't capable of running a real campaign, that his age was dragging down his chances, and that Democrats should have swapped him out.
Did you worry what would happen if you got all of that wrong? That people would be as upset with you as they were after the 2016 campaign, when you said Hilary Clinton was heavily favored to beat Trump?
In 2016, I thought people were in denial about the fact that Trump actually had a decent chance — he was only a couple of points down in these critical states, and Clinton had all types of signs of weaknesses. So then I was trying to emphasize that Trump did have a shot, and [the fact that people didn't understand that] was frustrating.
This time, I was even more skeptical about Biden than the polls. He was down to, like, a 26% chance in our model, but that assumes that he's running a normal campaign and not trying to avoid media appearances, and not having all his fundraising dry up and things like that. And so I thought his real chance might have been 10% or 15%.
As a poker player, it's kind of like getting all-in with aces against kings. It's a really good bet. Biden was probably going to lose.
And yes, I was taking on some risk by being kind of very vocal and insistent about it. A calculated risk.
But it's also kind of how I felt as a citizen, too. Usually these things have more than one motivation, right? I was very frustrated that I didn't feel like I had a responsible alternative to Trump, and I wasn't gonna vote for Biden. I was going to vote for the Libertarian or Working Families Party or something.
By the way: Why don't you want to be known as the election forecast guy? Why not be known as the guy who's really good at this?
To give a self-aggrandizing comparison: It's a little bit like if you're some indie band who has a big breakout hit, then there's always pressure to play the hits. That's part of it.
Part of it is also that we live in a world where people maybe don't understand probability as I would. And so if it's Harris 80/20 and Trump wins, or Trump 80/20 and Harris wins on election day, then that's running some career risk.
I first learned about you in your indie band phase, when you were a nerd idol doing your own blog. Since then you've worked at two really big media companies — The New York Times and Disney — and now you're back on your own. With hindsight, do you make sense at a big company, or do you think you're better as a solo act?
I think I'm constitutionally suited to being a solo practitioner. And also the business model of Substack is very appealing from a cash flow standpoint: All the revenues come in up front, and you get an 80% gross profit margin.
But I definitely hope to be doing some hiring soon. I almost certainly will be — but maybe half a dozen people, not 35 people.
I know friends who have gone through this at all scales, from small to medium to quite large — people who lucked into some idea and became CEO. I don't think their professional lives are terribly happy. You're spending all your time doing internal politics, external politics, compliance, lawsuits, investor relations. Those things are absolutely not my strength and not how I want to spend the rest of my career.
So it's just kind of accepting what can you do with … let's say I'm only allowed to hire six people, and that's a constraint. And let's say that, like, half the bylines have to be mine, right? And with those two constraints in mind, build the best business you can from there. I think sometimes constraints are a good thing.
Have you said how many paid Substack subscribers you have?
It's good. It's a very healthy number.