Nate Silver

I think 2016 was actually our best election.

Because I think about this as a gambler. And to me, the question is, what are the odds that you have relative to the consensus odds? Elaborate on that. So our model had Trump with a 30% chance of winning. And we spent a lot of time explaining to people that Hillary does not have this election in the bag. It's a much closer race than Obama. All these states, math term, but correlated, right?

Because I think about this as a gambler. And to me, the question is, what are the odds that you have relative to the consensus odds? Elaborate on that. So our model had Trump with a 30% chance of winning. And we spent a lot of time explaining to people that Hillary does not have this election in the bag. It's a much closer race than Obama. All these states, math term, but correlated, right?

Because I think about this as a gambler. And to me, the question is, what are the odds that you have relative to the consensus odds? Elaborate on that. So our model had Trump with a 30% chance of winning. And we spent a lot of time explaining to people that Hillary does not have this election in the bag. It's a much closer race than Obama. All these states, math term, but correlated, right?

Meaning that Michigan and Wisconsin, they're all kind of the same state. If you lose one state, you'll probably lose the others. And so all that has to happen is that Trump overperforms among so-called working class white non-college voters. And all of a sudden, then you can win the electoral college, which is kind of what happened basically.

Meaning that Michigan and Wisconsin, they're all kind of the same state. If you lose one state, you'll probably lose the others. And so all that has to happen is that Trump overperforms among so-called working class white non-college voters. And all of a sudden, then you can win the electoral college, which is kind of what happened basically.

Meaning that Michigan and Wisconsin, they're all kind of the same state. If you lose one state, you'll probably lose the others. And so all that has to happen is that Trump overperforms among so-called working class white non-college voters. And all of a sudden, then you can win the electoral college, which is kind of what happened basically.

So to us, if you were to have invested in Amazon stock on day one, and you said, I think there's a 30% chance that Amazon will boom into something really big. And everyone else says, that's crazy. It's only 1%. And even though you're below 50%, relative to the consensus, we were one of the only people who were saying that Trump really did have a shot. You failed least bad. Yeah.

So to us, if you were to have invested in Amazon stock on day one, and you said, I think there's a 30% chance that Amazon will boom into something really big. And everyone else says, that's crazy. It's only 1%. And even though you're below 50%, relative to the consensus, we were one of the only people who were saying that Trump really did have a shot. You failed least bad. Yeah.

So to us, if you were to have invested in Amazon stock on day one, and you said, I think there's a 30% chance that Amazon will boom into something really big. And everyone else says, that's crazy. It's only 1%. And even though you're below 50%, relative to the consensus, we were one of the only people who were saying that Trump really did have a shot. You failed least bad. Yeah.

Because again, you're getting odds. If you were gambling on the election, which I have no problem with.

Because again, you're getting odds. If you were gambling on the election, which I have no problem with.

Because again, you're getting odds. If you were gambling on the election, which I have no problem with.

You could get five to one odds on Trump, basically. Five to one, six to one. In 2016. Two and a half to one instead. So that means that the expected value poker term, if you're able to make that bet, that would be the best sports bet in the world.

You could get five to one odds on Trump, basically. Five to one, six to one. In 2016. Two and a half to one instead. So that means that the expected value poker term, if you're able to make that bet, that would be the best sports bet in the world.

You could get five to one odds on Trump, basically. Five to one, six to one. In 2016. Two and a half to one instead. So that means that the expected value poker term, if you're able to make that bet, that would be the best sports bet in the world.

Yeah, you're going to lose most of the time, but you get 5x, 6x your money when you're right. Got you. That's not how 99.9% of the political audience thinks. But we did take care, though, to emphasize the uncertainty to the point that we were getting yelled at. People are like, we just think Nate's trying to just get more web traffic.

Yeah, you're going to lose most of the time, but you get 5x, 6x your money when you're right. Got you. That's not how 99.9% of the political audience thinks. But we did take care, though, to emphasize the uncertainty to the point that we were getting yelled at. People are like, we just think Nate's trying to just get more web traffic.

Yeah, you're going to lose most of the time, but you get 5x, 6x your money when you're right. Got you. That's not how 99.9% of the political audience thinks. But we did take care, though, to emphasize the uncertainty to the point that we were getting yelled at. People are like, we just think Nate's trying to just get more web traffic.

The most concise explanation is that people who are more college educated, who read the news more, are more likely to respond to surveys.