GLENN THRUSH: So, Nate, the first question I want to ask you: --How are the Mets going to do?
[Laughter]
NATE SILVER: So we have a forecast about this, actually. I know we have the Cubs slated to win 104 games or something insane like that. We have a lot of Mets fans on staff. We have a whole podcast that's called--what's it called? Like Panic City or something? But it's basically reasons--
THRUSH: That's it.
SILVER: --for Mets fans to see the dark lining of every silver cloud, or whatever.
THRUSH: Because Mets are Brooklyn and Queens and Yankees are Jersey and purgatory.
SILVER: I find--I mean, you know, I moved to New York seven years ago now, and I can't understand why people like the Yankees. I just don't understand the deal.
THRUSH: Thank you. By the way, you just totally won me. All the nasty shit that people have said about you are now out the window.
SILVER: [Laughs]
THRUSH: The Mets--well, I think the Mets--my father used to say that the Mets were the Jews of baseball, right?
SILVER: Yeah.
THRUSH: My dad was, actually--my dad--true story--was a visiting team bat boy for the Dodgers and he knew Durocher. He didn't know him but, like, he was around as a teenager to see Durocher and Robinson. It was a huge‑‑you know, a huge thing. But the Yankees were always‑‑like, the one brief moment I was a Yankees fan was I was a huge Thurman Munson person--
SILVER: Sure.
THRUSH: --who was just like one of my favorite ballrplayers.
SILVER: I mean, individually, you know, I mean, some of the guys--you know, Bernie Williams is a likeable guy. But just the whole kind of brand of the Yankees, and they don't have a lot of likeable players right now. I just don't understand. You know, I don't like the new stadium--
THRUSH: It's awful.
SILVER: --very much. It's totally awful.
THRUSH: It's awful. Cities--you know, look. Shea was a toilet but it was my toilet, you know.
SILVER: Yeah. I grew up with Tiger Stadium where it kind of smelled like urine in like the concourses and stuff like that, and it was kind of falling apart in the latter years, but still just an amazing, beautiful place to watch a game. It had this kind of Roman Colosseum effect where, you know, the standards are sort of stacked on top of one another. The old Yankee Stadium had a little bit of that, too. I only saw a couple of games there.
THRUSH: What would frame those--was it Tiger Stadium that Reggie Jackson hit that epic All-Star Game home run?
SILVER: I think so, yeah, because you literally have the overhanging bleachers.
THRUSH: Right, where it hit the façade.
SILVER: Yeah, and, you know, you can hit light towers. It's kind of very dramatic.
THRUSH: Like The Natural.
SILVER: Yeah. Exactly.
THRUSH: And the perfect player for that stadium was Cecil Fielder--
SILVER: Of course.
THRUSH: --because he would hit these majestic--
SILVER: No. Those early '90s Tigers teams had Cecil Fielder and Mickey Tettleton and Rob Deer. They were kind of pre-Moneyball moneyball teams.
THRUSH: Rob Deer.
SILVER: Rob Deer.
THRUSH: Well, are you--did you invent—-OK. I will stop doing this and hit politics for a second. Are you--one of my favorite elements--and I was a huge--I have to just ‘fess up to being a geek here. I loved, you know, the Prospectus. I loved the write-ups in the Prospectus, too. I liked the wit. But the "Three True Outcomes," I loved the "Three True Outcomes." Are they discredited now? Is that something people don't dig?
SILVER: No. I mean, you know, for better or worse the game's moved more and more in the direction of kind of Three True Outcomes. I think people have realized‑‑
THRUSH: And let's explain what they are.
SILVER: So, a true outcome is a home run, a walk, or a strikeout, right? There's no actual running involved with the three true outcomes.
THRUSH: Right. Or trying to pick someone up in a bar, basically.
SILVER: Yeah, yeah, yeah.
THRUSH: You get the answer.
[Laughter]
SILVER: It's no, yeah--no grace, no finesse.
THRUSH: Right. Right.
SILVER: And, you know, we've moved toward more and more strikeouts. We've had a little bit of a revival of pitching, and so not quite as many home runs, but still, by historical standards, fairly high. But, you know, aesthetically, I don't like that there are so many strikeouts now.
The one blind spot I think kind of mid-'90s, early 2000s sabermetrics had was, like, defense is pretty darn important, too, sometimes, and you can draw analogies to politics if you want, right?
THRUSH: Right.
SILVER: Sometimes we think, oh, if we can't measure something, well, then, therefore it doesn't matter very much. It turns out that when you can measure defense it matters a whole heck of a lot.
THRUSH: Like Jeter's range was always a bit--
SILVER: Jeter's range--again, not being a Yankee fan--
THRUSH: Right.
SILVER: --Jeter's range is always a source of contention, and people are like, "Well, you know, where's that come from? It's just the stats." Well, the stats are actually now doing what humans do, right? They're kind of looking at every play and saying, considering all these factors, where the players were in position, and the speed of the line drive and everything else, you know, how many shortstops would have made this play, and did Jeter make this play? So the defensive metrics have gotten pretty good.
THRUSH: And this is all about--essentially, this is all about the increasing inexpensiveness of memory, right, and the cloud, so people are able to just sort of compound video and data and really make these analyses--
[Overlapping speakers 0:04:47]
SILVER: Yeah. I mean, we're getting to the point now where literally everything that happens on a major league baseball diamond is recorded, and it's just--I mean, I don't know. It's just so kind of light years ahead of even the stuff I was doing 10 years ago, when I was more active in baseball, and so it's pretty remarkable in that way.
THRUSH: So that is why--I guess that's why the Three True Outcomes thing is a little disconcerting, because there's less of a range of variability in sort of the nature of what people are doing, right?
SILVER: Yeah. I mean, I think--although I would say that, you know, the Royals won playing a very different type of baseball.
THRUSH: Well, can I just say, just about--and I know they beat the Mets--
SILVER: [Laughs]
THRUSH: --and it was--by the way, Jeurys Familia--I'm really going off on a tangent here--but, like, I saw Benitez a lot--
SILVER: Yeah.
THRUSH: --and goddamnit if they were protecting a two-run lead in the ninth, Benitez was going to give up that home run, right?
SILVER: Yeah.
THRUSH: And I'm like Familia is a totally different pitcher.
SILVER: [Laughs]
THRUSH: I didn't watch a single damn game of that after the first game.
SILVER: [Laughs]
THRUSH: But the Royals are the funnest--the Royals have always been a fun team. They're like the old St. Louis teams.
SILVER: Oh, absolutely. We have a bunch of Royals fans in the office, too. There's kind of a Kansas City/Mets rivalry. But, you know, and they're kind of zigging when everyone else zags a little. They're also more kind of amicably savvy than people think. You know, they're not using--you know, Ned Yost has a reputation for using--for bunting a lot, because he did in one series where he thought he had an edge, and he doesn't, historically, right? He kind of lets the guys play, but they're looking for a different skill set, and in some ways--I mean, everything is a market as I see it, right, and so people understand the value of on-base percentage and the three true outcomes and everything else, but there's a correction, and therefore maybe you can get guys who put the ball in play a little bit more, or who could play a good defense. Base running still matters. It's not as important as some people might say, but certainly it gives you a few extra opportunities to score runs. And so they're kind of taking advantage of all the undervalued commodities now in baseball today.
THRUSH: Well, it's funny. I had Jeff Rowe on, who was Ted Cruz's guy, who I really respect. I had him on for an hour and a half and he gave me sort of a tutorial on stuff. We'll talk about a couple of those things later. But it turns out Rowe was a--if he hadn't gone into politics, the career he had pursued and he was very good at was, he was a minor league and a high college-level umpire.
SILVER: Oh, wow. So is--
THRUSH: I just thought that was really interesting, and he was all--and the way that he would break down a game was, you know, he gave himself--I think this is sort of standard for high-quality umpires. It's like, I can blow five pitches, and anything below that is considered an excellent game. Anything above that is considered a bad game. I had not realized that they had sort of analytics on that as well.
SILVER: I think, for some reason, baseball and politics, I think, seems to attract certain sorts of brains, because you have to be, like, really patient to follow either one, or to cover either one. You know, it's a long season, it's a long campaign.
THRUSH: And drinking.
SILVER: And drinking. Yeah, definitely. The beer commonality, too.
THRUSH: [Laughs] Let's talk a little bit about where you're from. You have a fascinating background. Your father taught at--did he teach--was it at Michigan State--
SILVER: Michigan State, yeah.
THRUSH: Did he ever have Magic Johnson in any of his classes? [Laughs]
SILVER: He had hockey players. We grew up going to college hockey games at Michigan State, who used to have a very good program, and so he had a lot of hockey players, and he'd always tell me about that.
THRUSH: But you--so your mother--the only thing I could sort of see in the biographies is that your mother‑‑you said your mother was a community activist. Tell me a little bit about your mom.
SILVER: I mean, she was, like, you know, if they needed a new stop sign that was going to be put up or something like that. And, you know, when I was in kindergarten, they moved my elementary school, and that I thought was just down the block. They closed that school and made a community center. So it was all that very local politics stuff that people have to do on the ground, and in some ways kind of affects people's lives at a practical level, much more than--sometimes more than national politics does.
THRUSH: So she--what kind of--I'm fascinated by this, because the accusation, you know, what they were hitting you on in '08 is that you're Spock, right? But it's so interesting that your--so your mother was--was she an organized community organizer? Was she a kvetcher at the meetings? What kind of person was she?
SILVER: Somewhere in between.
THRUSH: [Laughs]
SILVER: I mean, not like--it wasn't like--not like a--excuse me, but not like a Barack Obama-type community organizer, not that organized an organizer, right?
THRUSH: Right.
SILVER: But just someone who--you know, she'd put out a community newsletter and be aware of local politics.
THRUSH: What was it called? Do you recall?
SILVER: I think the Bailey Newsletter or something, right? Bailey Community Newsletter, I think. But, you know, she was--I guess in politics we'd call her like a party elite, right? You know, people cared about her opinion in the neighborhood, and she could probably swing 10 or 12 votes for City Council or whatever else, right, and things like that.
THRUSH: And she--most of the time she was--was she a homemaker? What was her training?
SILVER: Yeah. She was mostly--mostly worked at home.
THRUSH: And she was from where originally?
SILVER: She was from Croton-on-Hudson, New York, technically was born in the Bronx.
THRUSH: Wow.
SILVER: Yeah. So Westchester County.
THRUSH: I always do the--people always accuse me of doing this but I always do kind of the Jewish genealogy crap, so I'll lay off of that. And on your father's side, it's a more, not patrician, but like captains of industry and that sort of thing, right?
SILVER: Yeah. I mean, my mom's side is actually kind of quite WASP, right, and they're all like very--
THRUSH: Oh, I'm sorry. I got it mixed up. Your dad's the--
SILVER: Yeah. And my dad--yeah, but they're all like engineers and stuff like that. He was born in Connecticut, moved to Los Angeles when he was 3, and that's where they grew up. So my grandfather was an engineer. I have an uncle who works for a jet propulsion laboratory, you know, was a scientist. And so they're the more sort of left-brained, disciplined side of the family. My mom's side is more kind of right-brained and creative, like a lot of architectures and, you know, things like that.
THRUSH: So the poli-sci, what kind of--what was your dad's kind of general area of interest as a political science professor?
SILVER: So he was a Sovietologist back when there was a Soviet Union--
THRUSH: No kidding.
SILVER: --which I was always afraid to tell people. I always thought, oh, they'll think, like, does that mean he's a communist or something? But, no, he studied the USSR and he studied official statistics in the USSR, which, of course, were not always accurate, right?
THRUSH: Five-year plans and stuff?
SILVER: Yeah. You know, the Soviet government would claim that, "Oh, we have a very low abortion rate. There are very few abortions in the USSR," and he came up with clever ways to find out that actually, of course, the abortion rate was quite high in the Soviet Union.
THRUSH: Wow. How did he--so, this is obviously‑‑most of this is in the pre-computer age, right?
SILVER: No, there's computers. But I remember like when I was really young he would use, like, punch cards to run regression analyses and stuff like that. But, yeah, I mean--but, yeah, I mean, when you start out and like it's really, literally kind of costly to run one hypothesis or one experiment, right, whereas now you can run 50, 5,000, 5 million versions of the model and kind of optimized stuff like that.
THRUSH: So this was very intuitive to you. So in addition to sort of--because this isn't just about sort of--there are a couple of elements in that that are incredibly fascinating. First of all, he was looking at an opaque system that--where you had to be inventive about figuring out a way that you could make a determination, right?
SILVER: Mm-hmm.
THRUSH: So it wasn't about just sort of glomming data and drawing conclusions. It was about obtaining the data, figuring out which data was crap--
SILVER: Yeah.
THRUSH: --and then--so it's an entire chain of a process, right?
SILVER: That's right, yeah, and that's part of what, you know--and I guess I see a lot of things as being the same way. One thing, actually, that I got a little spoiled by in starting out, working in baseball statistics or in sports statistics, is there you don't have those challenges. There the data is generally very good.
THRUSH: Right.
SILVER: It's kind of--every now and then you have to adjust for park effects. It's easier to hit a home run in, you know, Fenway Park than Dodger Stadium, or at least a double. You know, you don't--that's a relatively minor thing, right on the margin, whereas in some many fields, talking about--you spend 90 percent of the time making sure the data is halfway decent and then maybe you offer a few tentative hypotheses from that data. Sports, it's kind of the other way around, and that can spoil you when you get into areas where things are more difficult.
THRUSH: Now your undergraduate training was in‑‑
SILVER: --in economics.
THRUSH: --in economics. And did you learn--obviously that's a formula-based discipline. How did you sort of--when did you first sort of come across computers and when did you start, you know, writing algorithms and stuff?
SILVER: I always, like--so when I was a kid I would do programming in BASIC.
THRUSH: I did, too.
SILVER: Yeah. I kind of wish I'd kept that up. I did that until I was about, like, 13, and in high school I did like the debate team instead. But, you know, if I'd just kept on programming then--
THRUSH: You'd be making more money. [Laughs]
SILVER: Probably. Yeah. But, like--but, yeah, that's why sometimes now, like, actually, I still have to do a little bit of programming, and Stata is now the software I use mostly, and it's kind of very intuitive. All kind of programming languages are kind of the same at some level.
THRUSH: And--but this left brain--this conversation between the left brain and the right brain seems to--I mean, it's pretty--you don't have to do a lot of sleuthing to figure out where the hell you came from, right?
SILVER: Yeah.
THRUSH: I mean, you kind of have it laying out there.
SILVER: No. And if you were to meet my family, I mean, you would kind of see instantly, on my mom's side, where I get certain characteristics from, and on my dad's side, where I get, you know, the other ones from.
THRUSH: Well going from baseball to politics, it's a more--obviously the data is more difficult, it's a more entropic system, things are kind of moving. What made you kind of--you know, obviously, what made you think that you can sort of take the sabermetrics model and move it over to politics? What gave you that confidence?
SILVER: It was partly frustration, right, where I didn't see a lot of analytics-driven coverage of politics, and some of the campaigns do some of it, right?
THRUSH: Right.
SILVER: So it's kind of the opposite.
THRUSH: But more so now than--
SILVER: More so now.
THRUSH: Yeah, yeah.
SILVER: The campaigns do some of it but, you know, I was kind of frustrated by the tone of coverage. I remember in 2008, people were saying, ironically, oh, kind of, Clinton is inevitable, and, you know, I remember looking at the Iowa polls, and the Iowa polls showed a much closer race, actually a three-way race. We forget about John Edwards now.
THRUSH: Oh, he was huge.
SILVER: Yeah. And so it just kind of that. Like I'd done the baseball thing for four years and kind of got the itch to do something different. But it was always a plan to do it on the side and as a hobby, and then even as compared with sports, I mean, as you know, in a campaign, politics can just kind of grab hold--
THRUSH: Right.
SILVER: --of people, and you can get a very large audience. But in some sense it's kind of like, you know, people--we had a bunch of very good forecasts during the primaries--
THRUSH: Right.
SILVER: --but what we did is say "don't pay too much attention to the polls. Demographically this race is very predictable."
THRUSH: Right.
SILVER: And the Clinton-Obama race kind of famously was, right?
THRUSH: Right.
SILVER: So that kind of--
THRUSH: And, by the way, the Sanders--what people I don't think realize, in my conversations with both campaigns, is there's a certain amount of demographic destiny to the Sanders-Clinton stuff too. I mean, that's--
SILVER: Oh, for sure, right?
THRUSH: Yeah.
SILVER: And, you know, we saw that a little bit in California, where we were scratching our heads and saying, you know, I don't quite get why the polls show just a two-point race. Demographically, it looks like it should be Clinton by--I think our demographic model had it Clinton by 9.7, and it's like very close to what it's going to wind up being after all the votes are finally counted.
THRUSH: You mean it wasn't rigged and stolen by‑‑
SILVER: [Laughs] We did not make that many friends with Sanders fans.
THRUSH: I think they're all--judging from, like, the polls in the last two weeks, where there's been this unbelievable consolidation, like much more rapidly, actually, than the Clinton consolidation in '08, right?
SILVER: He lost his moment a little bit, although I would say that, look, Clinton's still at only 43 percent or so, in national polls, on average, which means she has some room to grow, and maybe she has to grow if Trump has a moment or two.
THRUSH: And the third party--well, let's talk a little bit about this. I'm fresh off a conversation--I just did a big magazine piece that involved a lot of talking to Clinton and White House people, and one of them‑‑I won't say who--I said, "I'm going to talk to Nate Silver." This was one of the data people. "What would you ask him?"
And this is the question that they posed: You have two candidates here who have 25 percent rock-solid unfavorability, like a quarter of the population, it's just rock-solid. Obviously it moves around a lot in the intensity or whatever, but there are 25 percent laying out there. That variable hasn't existed in previous elections, and a lot of people on the Democratic side think that that is the ultimate wild card, that is why--and you addressed it--I should just say you had your prediction out.
You have two separate models. One is roughly 80 percent probability of Clinton being elected; the other one is slightly less. Your poll plus one, which includes economic factors--by the way, I think that--we'll talk about that a bit later, but I also think that's going to be a big deal.
So to what extent do you think that that--do you agree that that represents a big variable, and are there any other variable, you think, here that will really come into play?
SILVER: So definitely one thing the models looks at is the number of undecideds plus third-party voters--
THRUSH: Right.
SILVER: --and there's kind of a blur between those sometimes. You know, when you have a high number of those then the model is less confident. So even though if we say Clinton has 80 percent or 75 percent chance, roughly, depending on which version you look at, if there were fewer undecideds that might be 85 percent--
THRUSH: Right.
SILVER: --or 90 percent. And so for that reason, I mean, it's kind of intuitive that--again, Clinton's at only 43 percent, Trump's at 37 percent or whatever, they both have some work left to do. One question is, how many of those 20 percent are going to actually stick with the third-party votes as opposed to being truly undecided.
THRUSH: Right.
SILVER: Right now we have Johnson at about 8 percent. This is another thing where our polls-plus model, which makes more assumptions--it says let's look at history more rigidly. It says, historically you have a third-party candidate and some of those votes will fade down the stretch run--
THRUSH: Right.
SILVER: --and so if that goes down to 5 percent--he'll get something--and Stein is a 2 percent, then you have a lot of votes in play. If he sticks at 10 percent and she's at 4 percent, then maybe there aren't that many undecideds after all. But, you know, that's a big thing that we're looking at.
It seemed for a while that the third-party vote was taking pretty evenly. Lately, though, it seems like there are more polls where, when you include Johnson and Stein, it takes more from Clinton. That's, by the way, the version that we use. We use the polls that do have that third party.
THRUSH: By the way, we had Johnson on here, I want to say a month ago, and this was at the height of the belief--you know, people sort of presuming that he was going to draw more away from Trump, and he sat here and he said he's going for the Bernie voters, and he thinks his largest draw. And really, when you talk to the guy, forget about the libertarian--not the libertarian--the free market stuff, the stuff that really pops is his marijuana stuff--
SILVER: Yeah.
THRUSH: --and the social--so he's much more--I think he's much more generically appealing to--
SILVER: Yeah, and we talked with him a little bit too, and he's kind of very explicit about the--look, I am happy to be the alternative to Trump and Clinton, that there are only three people that will be on the ballot in every state, and he thinks he will be. And I'm happy to soak up those votes. I don't have to be a purist. And obviously the selection of Weld, who is not ideal from a pure libertarian standpoint, is an indication that he is kind of going--he wants to raise some money--
THRUSH: Yeah.
SILVER: --he wants to get as much of the vote as he can with this kind of privilege of being on the ballot in every state.
THRUSH: Now let's talk about your--we've given you credit, or I haven't yet, so let me give you perfunctory credit for 2008. Now let me hit you--and 2012. We'll hit you about the head and neck on Trump. You've done it. You've beaten yourself up pretty good. As my father would often ask me, why did you fuck up?
[Laughter]
SILVER: I mean, I'm tempted to answer the question in two different ways.
THRUSH: Yes, of course you are.
SILVER: One of which is to say it's a crazy world and you--you know, sooner or later a 5 percent likelihood is going to come through. And in some sense I would defend the notion that, you know what? Inherently, Trump, at some point in time, certainly a year ago, was pretty unlikely, and that was a fairly sane and very defensible and kind of, in some cosmic sense, correct forecast.
I think the problem was that we never built a model for this until pretty late, right? When was it? Like late December or early January--
THRUSH: Yeah.
SILVER: --we finally had our state-by-state model that had Trump ahead, not in Iowa but in New Hampshire and South Carolina the whole time, right?
THRUSH: Right.
SILVER: And it said, yeah, if even you kind of account for endorsements we can put that in the model and he has such a big polling lead that, at the very least, he's a plurality favorite, where I never said that he would be, like, a slam dunk to win but, it said, you know what? Trump does have problems but everyone else has problems, too.
So I think getting away from the discipline of having to be more rigorous about your thinking, and I kind of learned how I make all the same mistakes as people I criticize unless I have some organized way of looking at things, right? Like I remember, you know, there was always contradictory evidence, right? So you could always say, well, Trump, sure he's doing OK in national polls, but in Iowa and New Hampshire, not so much. Oh, sure, he's doing OK but they're online polls, right?
THRUSH: Right.
SILVER: You know, it's very, very easy to cherry-pick and to kind of tell yourself a story--
THRUSH: As he does.
SILVER: --as he does, right, that confirms your hypothesis, whereas now, you know, once you build a model it's so much easier. There was a Rasmussen Reports poll that came out the day we were recording this and they had Trump ahead nationally. The last 15 polls don't, right? But instead of ignoring that or telling a new story around it, it just kind of ingests itself into the model--
THRUSH: And factors itself in as it will.
SILVER: --and factors itself in.
THRUSH: Well, you've been very tough on--in the past you've been tough--a lot of people have been tough on Rasmussen in the past.
SILVER: Yeah.
THRUSH: Have they gotten their act together a little bit more, or is it hard to tell?
SILVER: I don't particularly think so. I mean, in 2014, they were pretty far off as far as we could tell. They used to, at least, be prolific about polling a lot of states, and so it frustrates me now that, like, we've had so many national polls, and so if you--state polls--because my whole thing is that, like, there's more information in the state polls, right?
THRUSH: Right.
SILVER: So, you know, so I'm happy to--
THRUSH: And that's the paradox now. She's doing--which I don't think people would have really anticipated because everybody always talks about her built-in Electoral College advantage, and you point out in your forecast that that is the paradox. He's doing better on the meta at the state level than he is doing nationally right now.
SILVER: Yeah, definitely. You know, the state polls in the battleground states sometimes look a little bit more like 2012--
THRUSH: Right.
SILVER: --which, if you remember, in the election Obama won by 4 points, not by 7, with some exceptions. I mean, Florida clearly has moved more Democratic and is--you know, that looks more like a +7 outcome. But certainly the Upper Midwest--
THRUSH: By the way, that's the--a lot of people think that's the whole ballgame. If that is, in fact, true, I mean, look, you can play a lot of models around and throw Pennsylvania in and out, or whatever--which is not going to happen--but a lot of people think Florida is the whole ballgame.
SILVER: Oh, for sure, and the answer could be--I mean, first of all, things have to tighten, right?
THRUSH: Right.
SILVER: You know, if Clinton is ahead by 7 points nationally, then it's kind of an angels on pinheads question, which states she would win. But if things tighten, there definitely is a scenario where, if she holds on to Florida, and if she holds on to Florida, I mean, you could do an amazing amount of damage to her math everywhere else and it becomes very implausible for Trump to win.
THRUSH: Well, some of the--you know, again, talking with a lot of the Democratic folks, because the truth of the matter is you can't really talk to any of the Trump people about analytics or what they're looking at. I mean, he's talking about putting fricking Massachusetts in play--
SILVER: Yeah.
THRUSH: --which is hilarious.
SILVER: Well, this is the other thing too, where--and you talk about kind of--I always think what are factors that we don't account for in our forecasts and which ways could they split. I would worry if I were Trump because if you're relying on having any small edge in swing states, without a ground game, without advertising dollars, and then without necessarily even having the strategy to concentrate there, I mean, that--you know, that seems like more of a stretch.
THRUSH: But that also distorts your modeling, right, because you have to assume, as a baseline, that you're going to get a Romneyish--look, you know, ours have never been as great--I'll tell you just a quick anecdotal. I remember in 2012, going up to the Boston headquarters, and it was one-tenth--I'm exaggerating--one-fifth, one-sixth the size of Obama's operation, and it was because they don't do the same sort of state-by-state ground operation. Republicans don't tend to do it. They tend to do carpet-bombing and media.
SILVER: Yep.
THRUSH: But there's nothing--I mean, what does he have?--at most, 75, 80 national staffers?
SILVER: Yeah. I mean, even the best Republican campaigns--and we--you know, when we went to Iowa, and one thing we'd do in New Hampshire is kind of visit all the campaign offices, right, and, you know, and certainly Cruz was organized, comparatively speaking. But you go into, like, one Hillary Clinton or Bernie Sanders office and they're like, oh, there are 17 of these all over the state. And so people don't recognize the level of that disparity, but obviously Trump is going to make it worse.
THRUSH: Yeah. So when you do your modeling, because obviously you're making adjustments as you go along in these, do you actually adjust the model as you go along, or is this just--
SILVER: So this is what people don't realize. We don't really adjust the model. I mean, we'll catch bugs, right. There was a bug in kind of how it was handling the third-party vote and we caught that, you know. But, no, we don't make any manual adjustments. And this is part of what I realized, is like, you know, the model isn't the same thing as my opinion.
THRUSH: Right.
SILVER: I mean, they kind of heavily overlap.
THRUSH: Right.
SILVER: Sometimes I'll say, here's what the model says and we understand why it's saying what it does, but, you know, personally, if I were betting, maybe I'd bet on Clinton or Trump relative to the baseline established for this reason.
THRUSH: Yeah.
SILVER: And so, you know, one phrase is it's better to be approximately right than totally wrong--
THRUSH: [Laughs]
SILVER: --and a model is a model. It's not the real world. It's an approximation, right, and to say, you know, approximately we think, through a lot of disciplined analysis, that Clinton's chances are 75 to 80 percent. If you come in and say, "Nate, you're actually 70 percent or 84 percent," I'm not really going to dispute that.
THRUSH: Right.
SILVER: You know, I think they're definitely not 95 percent and I think they're definitely not 60 percent.
THRUSH: This is horseshoes and hand grenades territory here.
SILVER: Yeah.
THRUSH: So one of the things that Rowe--and, by the way, it's one of my favorite podcasts I've done, an hour and a half. It's like a geek festival. But one of the things that he said, he predicted, and I don't know where the hell he pulled this out of--apparently his own model--he thinks that Trump's lack of even a nominally competent data analytics and ground operation, connecting those three together, right, was going to cost him between 2.5 and 5.5 points nationally. That was his assertion, that he just felt that that was, in and of itself, a debilitating--now Cruz obviously has--those people have skin in the game, but, I mean, does that sound nuts to you?
SILVER: It sounds a little bit high, but remember, you know, we run all these experiments, or political scientists--well, I'm not a political scientist--where they'll compare, say, the Obama ground game and the Romney ground game, or the Obama campaign and the Romney campaign, and they're so close you don't really see that much of a difference. Maybe the Obama campaign certainly much more invested in the ground game, but a lot of things are involved in a campaign, including running advertising.
THRUSH: Right.
SILVER: It's kind of hard to separate those effects out. Whereas Trump is such an outlier in this respect that we don't know. You know--
THRUSH: So you're going from, like, stadium effect to no--to one team having a stadium and one team playing sandlot--
SILVER: Playing softball instead of baseball or something.
THRUSH: Yeah.
[Laughter]
SILVER: Or some--yeah.
THRUSH: But then--but doesn't that--getting back to your difficulties earlier, doesn't that kind of create--doesn't that just make it a hell of a lot more difficult for you in the black swan sense?
SILVER: I mean, in some sense.
THRUSH: Yeah.
SILVER: I mean, in some weird sense those kind of two things balance out, right? There's all these reasons to think that, you know, Trump will not get some of the benefit of the doubt that candidates ordinarily do, right, and the race won't necessarily tighten and things could get worse. On the other hand, the dude did win the primary by a pretty convincing margin.
THRUSH: I can't imagine it not tightening, just as--it's the way it always feels.
SILVER: It's hard--well, this is part of why I think, too, you know, and it's kind of--the poll--we have the two models, polls-plus and polls-only. Polls-plus assumes the race will tighten because the economy is mediocre, and so that has becoming a 4-point outcome, which is very familiar because Obama beat Romney by 4 points. And it also makes stronger assumptions that states will eventually revert back to how they voted the last couple of elections.
And so if you kind of are in the mindset that, you know what, this is still kind of the same election again except Trump is making a lot of mistakes and offends too many people around the margin so he's an underdog, but not a sure thing to lose, right?
THRUSH: Right.
SILVER: Whereas polls-only says, you know what, every now and then there is an election that realigns the map every bit--a little bit, rather--and also every now and then there is an election that's more of a landslide, and there are plenty of those throughout American history. And if you look at the polls now, then--with this caveat about the swing state polls not being quite as good for Clinton--it suggests she's pretty far ahead and also might put some red states--
THRUSH: And there's, like, yeah, and actually we're seeing--if the trend line of the past three or four of these, including that whole raft [phonetic] of the online poll--it actually wasn't an online poll. I forgot which one it was. But it just seems like the trend line--the national and states, over the past week or so, seem to be coming closer in alignment.
But let's talk about the map in general, because I think it's somewhere in between. You know, the Obama people, when you talk to David Plouffe, Plouffe will tell you that he thinks 2012 was a better--was a--they ran a fundamentally better campaign than they did in '08, just simply because they entered it with this--with a huge wrong-track number. Obama's approvals were sub-45 for some percentage of the cycle--
SILVER: Mm-hmm.
THRUSH: --blah, blah, blah, blah, blah, but that they got their shit together in an incredible way. And the other thing about it, which I hadn't realized, is, again, getting back to the cloud and data storage, they were just able, in terms of their voter targeting, to exponentially get more data on voters. So, like, I think that's what people don't realize in the Sasha Issenberg sense, is the game has moved radically up the road in the last eight years, like--
SILVER: Yeah.
THRUSH: --campaigns that have access to that stuff are able to really fine-tooth comb things.
SILVER: I mean, I'm very much on board with the idea that the Obama 2012 campaign was more impressive than at least the general election campaign in '08. I mean, in '08, Obama won by 7 points in the middle of a horrible economic crisis, a two-term incumbent who had a 25 percent approval rating, Sarah Palin and everything else, right? So in some sense the question in 2008 is how did this guy only win by--
THRUSH: Yeah.
SILVER: --by 7 points, whereas in 2012, you know, some of the models the political scientists put together that say what can we say without looking at polls [unclear] the economy. They have that election as more of a toss-up, maybe the slimmest Obama edge, and a 4-point win isn't that much but, you know, under the circumstances, like you said, starting out with the mid-40s approval rating and they over-performed a little bit in the swing states, so it was a fairly impressive feat.
THRUSH: I think the voter targeting stuff is, like, it's revolution--I mean, it has already revolutionized and it's going to continue to do that.
Let's talk about--just a little bit--I'll give you a little--some insights that I've had in talking with people over the past couple of months on states. There does seem to be some Dems, again, and I've talked with a couple of the Republican campaigns about this as well, but the sense is--and again, this is very early in the game--the feeling is Nevada is trending more red than it did in 2012--people are feeling that--whereas North Carolina is trending significantly more blue. And Florida, unless--I mean, things could change but the influx of Latinos in the Orlando corridor, Florida feels considerably more blue to sort of the Dem professional class, and Iowa, a little more red.
So there are--I mean, I think this is a little bit around the margins, but the margins are the whole ballgame in the Electoral College. So are you seeing--we talk about sort of a larger political realignment based on the candidate. Are you seeing--what's the impact of these demographic alignments? You talked about Arizona maybe being a little redder--a little bluer.
SILVER: Yeah.
THRUSH: Georgia being a little bluer. To what extent are the demographics moving some of this?
SILVER: I mean, you know, one of the clearest things our model finds is that Clinton seems to be making larger gains in Hispanic-heavy states than other states, and that includes Florida, it includes Arizona, it includes Texas, where some polls have it within high single digits. I don't think she'll win it, but it's an indication of where the vote might be moving.
But yeah, also the new Census Bureau numbers came out last week. In Florida--usually when a state already has a large Hispanic population, it doesn't grow as much. It's actually the states that were more white, where you see agricultural communities--
THRUSH: That's interesting.
SILVER: --that where the population is growing. But Florida, the Hispanic population is still growing quite a bit, and that growth is not coming from Cuban--
THRUSH: Puerto Rico.
SILVER: --coming from Puerto Ricans, right. And, by the way--
THRUSH: We'll get to a--
SILVER: --the Cubans aren't as pro-Republican as they used to be anyway and they're definitionally not younger--
THRUSH: It's a generational thing.
SILVER: Yeah.
THRUSH: And the Puerto Ricans--the thing about being Puerto Rican, you--you know, you go from disenfranchisement to enfranchisement the second you step on the soil.
SILVER: Yep.
THRUSH: So it becomes a totally different dynamic.
SILVER: Yeah. And so it's not that hard to see Florida shifting. I mean, Nevada's becoming more diverse, too, but it's not been particularly blue in midterm elections, and people say it's partly a turnout effect. But, you know, the short answer is, I will trust anyone's opinion who has any poll in Nevada, because there hasn't been one since November, publicly.
THRUSH: It is nuts. It is nuts, and I don't know of any of the private stuff, either. I don't know why.
SILVER: It's really weird.
THRUSH: Yeah.
SILVER: Colorado, Virginia, we've had very little polling also.
THRUSH: Colorado in particular--I think there's a ton of private polling going on in Colorado. I admit, it's got more money than God--
SILVER: Sure.
THRUSH: --but it's like--well, that's the other thing. Colorado--but then we see in some polls recently pushing it--making it a little bit tighter. But the other feeling is that Colorado is bluing up a little bit more too.
SILVER: I mean, you would think that Colorado is not particularly a Trump state. You know, I do think it's maybe not a great Clinton state either. I mean, it's kind of--even the liberals there are fairly anti-establishment sometimes.
THRUSH: Right.
SILVER: And Clinton's obviously ultimate kind of insider. And so we'll see.
You know, the one state that we haven't had a lot of polling, where I have to figure it's a bad Trump state, is Virginia. The polling we have had has not been particularly bad for Trump. I just wonder if we got three or four more polls there--it just seems like, you know, pro-establishment, very high income--
THRUSH: Barack Obama, dude, is going to live in Henrico County.
SILVER: [Laughs]
THRUSH: I mean, that is where--I think Virginia and North Carolina--I think if you're going to talk about where Obama's going to campaign, he's going to live--I was with him when he did these--like, Emporia, Virginia, is going to get a lot of Secret Service traffic.
SILVER: Yeah.
THRUSH: I mean, I just think, like, that is where that shit's going to go down.
Pennsylvania--you know, the joke on Pennsylvania is it's blue until the last three weeks and then Ed Rendell freaks out.
SILVER: [Laughs]
THRUSH: Yeah. Do you see anything--in terms of Pennsylvania, do you see anything? Because that's really a sine qua non for Trump, as far as I'm concerned.
SILVER: Yep.
THRUSH: That's the bellwether. Ohio, to some extent. Ohio's always tight as a tick anyway. But Pennsylvania, from what you have seen--again, there's not been a ton of polling. There's a lot of races going on, but do you see anything that would indicate to you that it would move into the blue column?
SILVER: I mean, so the polling in Pennsylvania has been weird, where there's a lot of polling show, like, a 2-point race, and occasionally they'll have one showing a 15-point lead or something, and you kind of average that together.
You know, look, our model says that Pennsylvania has moved closer to the national trend line, both overall and with Trump, specifically, where it is struggling a bit economically. I mean, Pennsylvania is basically two states, too--people have to remember. You know, you do have a lot of the--
THRUSH: Carville had the line on it, that it's like you've got Philadelphia attached to Alabama.
SILVER: Yeah.
THRUSH: Right?
SILVER: Yeah. And it's kind of half Northeastern and half Midwestern, or something, and a little bit Southern in places. But there is a large white working class vote there, and so, you know--and whereas, you know, to me, the states that Republicans will dream out winning and have more trouble in are Wisconsin and Michigan.
THRUSH: Right.
SILVER: Those seem like fleeting dreams. But Pennsylvania, absolutely, and one thing we calculate in our models, what we call the tipping point state calculation--
THRUSH: Yep.
SILVER: --if the election is close, which three states or which states will matter the most, and it's Florida, Ohio, and Pennsylvania. So kind of, everything old is new again. And then after that, the kind of list gets more diverse, and it says Arizona. Every now and then Trump makes a big play in New Jersey. He had a bad poll there yesterday, but before that there had been some polls showing it close.
THRUSH: He's not going to come close in Jersey. I mean, that's my feeling, is he doesn't close in Jersey. He just doesn't have that--
SILVER: Well, see, that's where--you know, that's the thing, too, is that, because, as we've talked about before, you know, because you have 20 percent of this vote that's sitting out there saying third-party or undecided, I have to think that some of that's going to come home after the conventions and that some of these polls, the occasional polls showing Trump only down 8 in New Jersey, or the poll showing Clinton competitive in Kansas, I have to think that some of that will look silly by September.
THRUSH: Well, I don't know. You know, on some of the stuff, I mean, Kansas has local atmospherics with the Brownback stuff. Weird thing--the other thing about it is there's weird local weather in some of these states, you know. Like I don't think it's going to be that weird, but, I mean, I do feel--one does feel like the demographics are moving--in general, you know, as you're looking--you know, one of the things that all campaigns over the last, I would say, six years have been telling me, the velocity--and again, this has to do with social media but also demographic change--the velocity of issues, the change on issues--and gay marriage, of course, is the big one that people talk about--I think trade is moving rapidly, running away from the establishment, right?
SILVER: Oh yeah.
THRUSH: Let's talk a little bit about that, the velocity of these issues. People are really moving around faster, right? Is that a perception or is that empirically supported?
SILVER: I agree with that perception.
THRUSH: Yeah. [Laughs]
SILVER: You know, I'm thinking about how you would or wouldn't prove it. I mean, you know, to get very meta, we've had politics that are kind of in stasis for a long time, or at least at the presidential level kind of the math looks more or less the same in 2000 and 2012 and every year in between, and that's unusual. Historically, politics is quite tumultuous. I mean, for some reason people think of, like, oh, the '50s and the '90s are normal, right--
THRUSH: Right.
SILVER: --and everything else is abnormal. Well, that's two decades out of like the last eight--
THRUSH: That's right.
SILVER: --and we're not totally crazy.
THRUSH: That's right.
SILVER: The others decades are suddenly crazy. And so, you know, it feels like you have--and, I mean, in a literal sense--but you could have fairly dramatic and violent change, right? And, so, oh, if Arizona and Maine's congressional districts flipped, I mean, that's not particularly--
THRUSH: Right.
SILVER: --you know, that's still pretty glacial, right? So maybe there is a chance of the states realigning themselves quite a bit and we'll see some initial inklings of that this year.
THRUSH: Is this--OK, so you jumped from baseball to politics. How long have you been doing politics now? It's been--
SILVER: I guess eight years.
THRUSH: So it's like twice as long as you were doing--no, no. How long did you do baseball?
SILVER: So baseball, I mean, I was at Baseball Prospectus for four or five years--
THRUSH: Right.
SILVER: --and kind of studying it on my own before then. But it kind of--about the same length of time, roughly.
THRUSH: Okay. So basically you're at the end of your second term. Are you going to get bored with this and jump on something else? [Laughs]
SILVER: I mean, well, part of what we're trying to do at FiveThirtyEight is diversify, and we cover sports and we cover, you know, we have a science section, we have a whole big feature on gun violence coming out next month that we're really excited about. We do some reporting, more than people realize. We believe in the value of reporting stories like Flint and Ferguson.
THRUSH: Right.
SILVER: So, you know, I'm interested in this whole kind of range of things, but the election is just so‑‑it has so much, kind of, scale.
THRUSH: Yeah.
SILVER: I mean, I'm thinking, I guess, kind of as a publisher, right, but like, you know, there's so much interest around, at least the last couple of elections--maybe that's abnormal. If we were in the '96 election or '84, it would be pretty boring, I suppose. But people are so darned interested in politics, it's kind of like, it's hard to get away from.
THRUSH: Right.
SILVER: You know what I mean?
THRUSH: Right.
SILVER: And, you know--
THRUSH: Well, the consequences are--well, you know, what's really interesting about it is--and I've covered the White House in addition to politics, which gives me a different perspective because I've actually fucking covered events, right?
The thing about it is, the less control--and this is a paradox--the less control a president has over his own destiny, and increasingly in every area, in every arena, the presidents have less power than they used to--
SILVER: Mm-hmm.
THRUSH: --entitlements are chewing into discretionary spending, world events are running apace, China is eating a bigger part of the pie, right?
As we have become--as the presidency itself has become less in control of itself, the competition, internally, over this job has become more and more ferocious. I just think that that's a fascinating--like the politics have become more violent and active. I don't mean violent in the literal sense--
SILVER: Yeah.
THRUSH: --where the presidency has become more reactive, and to some extent--
SILVER: Yep.
THRUSH: --under Obama, more passive. That's just an observation.
SILVER: No. I think people will--you know, everyone kind of thinks this whenever they're in the middle of something, right--
THRUSH: Yeah.
SILVER: --but I think people will look back historically on Obama's presidency as a turning point in a lot of ways, including the style of how the presidency is managed.
THRUSH: How so? In a good way or a bad way?
SILVER: Well, what you just mentioned, kind of Obama being more kind of reactive--
THRUSH: Yeah.
SILVER: --letting things play out and then making kind of the final decision, being more analytically kind of oriented. You know, the critics would say that means leading from behind, right, but it's kind of the same version of this idea which is kind of collect information and then make a careful reasoned decision, right, versus kind of trying to dictate the course of the country. I mean, every presidency is kind of an overreaction to the previous one, and so things kind of--an overreaction to Bush, of course, was very alpha male and trying to kind of rewrite the narrative constantly. And so, you know, for Obama to be so, kind of, cool-headed and Spock-like, I guess, I mean, that's--you know, I think that's unusual for--
THRUSH: I think that's why his numbers are spiking, I think, in a direct response--look, it happens in general at this time anyway, if you don't have a huge scandal.
OK, here's a stupid question, or a couple of stupid questions, and you could punt on it if you want. You know, one of the really difficult--if I remember, in the Prospectus, and correct me if I'm wrong, you guys have real trouble judging the performance of managers.
SILVER: Mm-hmm.
THRUSH: Like it's the toughest--
SILVER: Yeah.
THRUSH: --apart from certain defense, it's like one of the really tough things.
SILVER: Yeah.
THRUSH: If you were to liken Obama--this is a cheap-ass question; you can punt.
SILVER: [Laughs]
THRUSH: If you were to liken Obama, Clinton, Trump to managers, who would you--
SILVER: Oh, my gosh. I mean, I think Trump is Pete Rose or something.
[Laughter]
SILVER: It's the obvious answer.
THRUSH: No Hall of Fame for him, huh?
SILVER: Yeah.
THRUSH: [Laughs] [Inaudible comment 0:44:02]
[Laughter]
THRUSH: Who would Obama be like as a--I was thinking Madden, maybe.
SILVER: Joe Madden, yeah, quite detached--
THRUSH: Yeah.
SILVER: --like at least kind of outwardly quite successful in the long run.
THRUSH: Yeah.
SILVER: I don't know who Clinton is. Do you have ideas here?
THRUSH: I don't know. Well, she's both kind of like--she tends to hit the panic button--
SILVER: Yeah.
THRUSH: --but she has a lot of--maybe Ned Yost. [Laughs]
SILVER: Maybe Ned Yost, or maybe there's like a kind of a--I mean, she's [unclear 0:44:31]. Maybe like a Joe Torre, like part of a dynasty--
THRUSH: Yeah.
SILVER: --and, you know, kind of does the basic blocking and tackling stuff well--
THRUSH: Yeah.
SILVER: --kind of a subject of frequent criticism but ultimately pretty good in her job and isn't going to--is not going to screw up in new ways.
THRUSH: And resilient.
SILVER: And resilient.
THRUSH: Yeah, resilience seems to be--
A couple of more questions. Thanks again for taking the time. This has been a blast.
OK, so this was--you know, back in '08 I hated your guts, just completely hated you, because at that point in time the differentiation between what you were doing and what we were doing as sort of experiential reporters--
SILVER: Yeah.
THRUSH: I just coined that term--was much darker, and like the camps were much more clearly defined, like "fuck you, man, don't tell us we're not getting the story right."
SILVER: Yeah.
THRUSH: Right? It seems to me, judging from hour site, which I read and enjoy, that you have--I wouldn't say you've--I think you've changed a bit. I think you've become more interested in the classical modalities of what we do. Right?
SILVER: Mm-hmm.
THRUSH: And you're out there more, and you sort of seem to be engaging in it on our level.
Tell me a little bit about that transition. Do you feel like--am I identifying something that's happened with you?
SILVER: Yeah, and I think--I mean, I think, likewise, Politico, by the way--and I'm not just saying that because I don't say thing to flatter people--
THRUSH: Yeah.
SILVER: --I think the coverage Politico has done is very good, and, like, you know, I was kind of so hoping that we would have a contested convention, as every reporter hopes, because, you know, you guys did such terrific reporting on the delegate situation, and I think we had a lot of value there, too.
THRUSH: We've got great people. Thank you.
SILVER: But, look, I think the beef that I had, in some sense, with the way that the campaign is covered is both very narrow and very broad. It's very narrow in the sense that, you know, the things that irk me‑‑
[Cell phone rings.]
SILVER: Excuse me.
THRUSH: That's me.
SILVER: The things that irk me are relatively narrow range, where it's like the kind of throwaway assertions about, oh, this race is a toss-up, right, or Pennsylvania is going to turn Republican this year, or whatever else, when you could use polls on the types of evidence to make claims that are more factual. But that doesn't mean we don't believe in the value of reporting. I mean, as you said, you know, we have people who we've hired who basically act as reporters and have that reporting background.
You know, I see models being a good approximation. Maybe it gets you 80 percent of the way there.
THRUSH: Was there anything--
SILVER: Twenty percent's really fascinating, though, and often, you know, there are great stories in the 20 percent, and, by the way, the model is as good at telling you what. It doesn't tell you very much about why.
THRUSH: Right, and that's--it's when you're out there, and you did some traveling this year. I ran into you in--at that really kind of fun day, the day after Marco flopped the robot debate in New Hampshire was just a crazy day. You really felt like the worm was turning that day, right?
SILVER: Yeah. [Laughs]
THRUSH: And, what do you--was there anything that you have learned from being out on the road or doing sort of human intelligence kind of reporting, that has altered the way that you view--the way you do the other stuff?
SILVER: I mean, you know, you realize there are humans behind the campaigns and kind of a lot of smart people thinking about stuff, and that's--you know, that's always good to see, I think. You know, the flip side, I think you also see you're not the flip side. Maybe it's part of the same equation. You see the candidates are more human sometimes.
THRUSH: Yeah.
SILVER: Whenever a candidate does something, particularly Trump, people are kind of trying to, like, kind of think through, oh, you know, what's the strategy behind this decision? Well, maybe the guy just woke up and felt like doing a certain thing--
THRUSH: Yeah.
SILVER: --like most of us do. You know, we can't kind of rationally explain everything that we do. So that's part of it. You know, I saw a couple of Trump rallies in Iowa, and I did not get the feeling that, oh, there's this hidden Trump movement, but then again, Iowa was not particularly a good state for--
THRUSH: Right.
SILVER: --for Trump, and so I'm not sure if that was a confirming or disconfirming--
THRUSH: What did it feel like--and you are--I mean, culturally, you come from a--you were raised in a college town, you lived in a coastal city.
SILVER: Yeah.
THRUSH: Your sexual orientation is not, you know, acceptable to a certain percentage of the Republican electorate. What did it feel like being at a Trump rally? I've been at a couple and I found them fascinating.
SILVER: So I'm from Michigan, and, you know, my partner's from the middle of Kansas, and so, you know, to some extent, like, it still feels--Iowa feels quite familiar to me--
THRUSH: Right.
SILVER: --and so it's not that weird. But, you know, we kind of went in and one of us got press credentials before we started being denied credentials.
THRUSH: Oh, that's right. You guys--
SILVER: We're on the list. We don't brag about it as much as, you know--
THRUSH: Wait. Were you--were you the one who got the credentials or no?
SILVER: No. Claire Malone, who's our more traditional reporter.
THRUSH: You wore a hoodie, and--
SILVER: No, I just went into the--
THRUSH: [Laughs]
SILVER: I mean, almost all these events--
THRUSH: You go into them.
SILVER: --you just go into them, right?
THRUSH: Yeah, yeah.
SILVER: I mean, who cares?
THRUSH: People are nice.
SILVER: Yeah. But I don't know. I mean, to me, in this--in these Iowa rallies, Trump was more, kind of, there was--people were very excited at the start, and by the end I wasn't sure if he had sold people. He spent‑‑this was in the phase where he was spending so much time talking about his polls, and I'm not sure if that appealed to people as much. But, you know, my conclusion was would he have a little bit of trouble closing the sale. But this is also a time in the campaign when it was not one of the more violent--and this time I mean it in the literal sense--it was not one of the more violent periods.
THRUSH: I mean, they were fighting over eminent domain.
SILVER: Yeah.
THRUSH: Like, you know--but it must have been--come on—Nate Silver is sitting at a Trump rally, listening to Trump read off a piece of paper on polls.
SILVER: [Laughs] I think he bragged about, like, this rally he bragged about how, you know, I made polling famous. Right?
THRUSH: Oh, you were there for that?
SILVER: Yeah, yeah.
THRUSH: [Laughs] So what--you're sitting there. Come on, man. You're sitting there. It's like the cosmic streams--
SILVER: [Laughs]
THRUSH: Like Ghostbusters.
SILVER: I mean, it was kind of--he was, like, praising the local wresting team. I mean, it's--I mean, it felt like being at, like, a con