Statistics With Nate Silver: Masters in Business (Audio)

Times and now with ESPN, Nate Silver. We had a wide ranging and lengthy conversation about everything from sports to statistical analysis, to politics and campaigns, and given the current madness in the um campaign season, we spent a lot of time talking about what's going on, much of which is really really fascinating. We we went a little long, and normally I like to give you a little more details in these intros, but I think this, uh, this

podcast stands on its own. We covered everything, so rather than me continually babbling about how fast and in this is, let's just jump right to it. Here's my conversation with five thirty eight Nate Silver. This is Masters in Business with Barry Ridholts on Bloomberg Radio. This week on Masters in Business, my special guest is one Nate Silver. You

probably know him from five thirty eight. A quick bit of background about Mr Silver graduated in two thousand from the University of Chicago bachelor's degree in economics, became frustrated by what he was seeing both in the world of sports statistics and politics. Created we'll talk a little bit about Pacoda, which was the baseball statistical system you had set up, and basically said, while stranded in an airport in New Orleans, the idea for five thirty eight popped

fully formed into your head. Is that true? I'm not sure fully formed right. There was a certain amount of you know, gumbo and stuff have been consumed, and yeah, and on a slash slash thought lots of other things in New Orleans to but the analysis about about elections was frustrating you. And and obviously five thirty eight is a number of total votes in the US Electoral College, of which you need to sixty nine or two and um and created five thirty eight. Just a quick background

about five thirty eight. In two thousand and eight, it correctly predicted the winner in forty nine of fifty states for the presidential election. Uh. It gets licensed by the New York Times in two thousand ten in anticipation of the two thousand twelve election, and he basically Nate runs the table in two thousand twelve predicts every state and the district of Columbia correctly as to the winner of Barack Obama versus Mitt Romney, as well as all thirty

five UM senatorial race. I should be correct. The Senate was in two thousand and eight, we had thirty thirty five. We missed think two in how dare you sir? Well the irony and we'll talk about this morning ironies and like the fact that you know, uh, because we got lucky. Basically, you know, there's some skill, but there's a lot of lucky, and so let's jump right into that. So normally at this point I say, welcome to Boolberg, but I know you've been here before, and I'm I'm really thank you

so much for doing this. I'm really excited. We'll talk a little bit about your relationship UM with various uh statistical approaches, how five thirty eight just blew up at the New York Times, and how life is currently at ESPN, and then we'll take a really close look at the elections.

So I'm sure there's a ton of stuff. I would be remiss if I failed to mention that you were named by Time magazine is one of the hundred most influential people in the world, and that your book, The Signal in the Noise was a New York Times best seller and an Amazon number one pick for Best our Nonfiction in two thousand twelve. Well, thank you. So so that's really um, we're out of time now, but let's let's let's start with a little bit of of baseball.

So you develop something called Pakoda. Tell the listen one, it's exactly what that. So Pakoda is a terrible acronymic. Yeah, it's like picture empirical comparison and optimization tests algorithm. But the idea is that couldn't you've come up with something wonkier than that? Yeah, but Pakoda built Pacoda was a

baseball player. I played for the Royals, like in the eighties, the Tigers fan and so it was a thorn in the side of the Royals, right, So it was meant to be kind of intentionally self parrying and giffy a little a little um part of the punt inside baseball, a little inside baseball. Yeah, And that's the thing I think people don't realize about, you know, kind of what I do. It's always a little bit very serious work.

But we're not taking ourselves too seriously. But anyway, the idea of Pakoda was to use baseball is very rich history to project the future, which is really all you know of Cisco models are really using this history protict the future, right, an extrapolation based on what's the highest

probability relative to what's happened in the past. But in baseball you have so many good years of data and so many players every season where you can say, you know, uh, taked, you know current player, right, Curtis Grander cent or whatnot, age thirty three whatever is for the Mets, and go back in history and see who were the guys who were like Curtis Granderson at age thirty three and similar attributes, similar history, similar attributes, right, and you can say, okay,

so now we can say kind of, here are a hundred different career paths for um, what's the risk of a you very easily could have ended up in a career path where you would have either been in Wall Street analyst or economists because essentially the better ones. That's what they do recognizing the limitations of pretty similar right. And we're spoiled in sports in the sense that, um, so much data is so much data. The data is so reliable, you know, there's not really a lot of

uh uncertainty there in terms of measurement error. Right. Um, so you get really spoiled in sport and you spend a lot of your time actually testing hypotheses, whereas in most fields, you know, to some extent, including politics, you're spending nine saying your time cleaning up the data and scratching your head and saying, you know, does this tell us anything at all? We we look at non farm payrolls, the way it's measured each month has changed and evolved

over time. Everybody forgets, you know, this data doesn't go back forever. Was only after the Great Depression, and FDR essentially created the Commerce and Bureau of Labor Statistics Department in order to assemble this data so we can actually so so let's get into a little more details and and we'll we'll come back to Pacoda because it's it's fascinating.

In two thousand and seven, you're still working for you sell Pacoda to Baseball Perspectives, you're writing for them, and you start posting at the Daily Coast under the name Poblano looking at various um polling data in a way to to, well, why don't I let you describe, yeah,

what you're doing? An oh seven? So uh, So, I had worked for Baseball Perspectives for for several years that point, and this was kind of in an era when Moneyball came out in two thousand three, right, fantastic book, the movie was great, great book, and you kind of that triggered a lot of interests and things that Bill James have running about for decades. Actually, metrics has been around

for yeah, it seems like forever. But you saw how much of an impact that had both on the game itself, the way the game was played, and the way the game was covered by the media, and it seemed like there was very little of that in campaign coverage at all. Um. So I kind of started anonymously a daily coast, uh, kind of writing these little things about about the two

thousand and seven primary between Clinton and Obama. Um and eventually said, you know, I kind of want to be on my own here, and so I launched five dot com in March two eight. You're listening to Masters in Business on Bloomberg Radio. My special guest today Nate Silver of five thirty eight, the man who correctly predicted just about every race that mattered in two thousand and twelve and came pretty darn close, and in two thousand and

eight got just about everything right then. So you launched the side in o A. You do a great job with the election in oh eight, uh, not only determining who's going to win the Democratic primary. So a lot of your insights turned out to be very prescient on that. But then you do the head to head between Obama and McCain and Hillary McCain and tell us what you found. Well, I mean, two thousand and eight was not really among

them more suspenseful general elections. Um, I guess early on in March when I launched, uh, it was a little

bit closer. But historically polls in March don't tell you very much, right, And that's one of the things you talk about very often, which I think a lot of people don't really pay attention to us, which is, hey, here's the polls, but at this stage don't give us at election cycles are really really long thing, right, And we spend this country, in this country, the UK, they don't understand it's five and a half weeks and they spend nine million dollars and They look at us like

what crazy? But here we have it's a two year process, right, um, and we probably spend eighty percent of that time at eight saying you know what, these polls are not very meaningful. You should not take them to literally interpret them with a lot of caution. Right. Then at the end we say, actually, when you get after Labor Day, polls do a pretty good track record, right, and pay more attention to them.

And so but people people don't do that, right, the kind of are too distrustful of the polls in November and much too serious about the polls like now in August of all things, a year before the election. Um, you know, you could go back and look at who was ahead in August of the past nomination campaigns. Well, in two thousand twelve, Rick Perry at this point in time was surging past Mitt Romney. In two thousand and eight,

it was Rudy Giuliani and Hillary Clinton. Two thousand and four it was Joe Lieberman and Howard Dean on the Democratic side. Right, Um, so kind of four in a row. I'm before that two thousand was a more predictable year, right u. Um, but you know ninety two, Bill Clinton hadn't even entered the race yet. Um, So you know, people kind of ignore history at their at their peril. That's fascinating. We'll get to that more in in a

later segment when we talk about this carent election. So, so, oh eight comes and goes and you do a you know, killer job. Who else besides The Times approached you? How did that come about? We had four or five or six conversations with different people, and you know, I try and be respectful of those conversations, but there was there was a lot of interest. I mean, I think people knew how big a story two thousand and eight had been. You know, for me, it was kind of like it

would be useful to have. I mean, I was basically running eight on my own. We had some other part time contributors. I was doing everything from doing the graphics to writing the stories, trying to promote the blog, and it was just a lot of work. And I think, um, that's a nuver used word, but there has to be some synergy between what a contributor might have to say

and a big media company like The Times. For example, Um, was that the M I. T. Sloan Business Conference, which is a sports business conference every year and ran into an editor from The Times on a train platform and we really and we talked, and so it's kind of spontaneous, right, a little serendipity and uh yeah, a little serendipity, and it wound up being a really good fit for a

couple of years. It was a tremendous fit. And if memory serves something like in the height of the election the I'm sure I'm destroying the statistic, but you were twenty or of the total New York Times web traffic. I think, you know a second, it's a little at a peak. At a peak, yeah, um, you know, I think on election week something like twenty percent of unique and users at the Times went by and viewed eight Um. It was really nice, and it definitely got really crazy.

I can imagine towards the end of I mean, it's a little bit like this is really self agruandizing, right, but it's a little bit like, um, I feel some sympathy for like Olympic athletes, where you know that every four years their life becomes like a total crazed swamp of although the difference between you and an Olympic athlete, um is that they're training the entire for you. They train the entire for you. You have I need a lot better than I do. But uh so let me

ask you about training. Food doesn't matter as much as so. You have an undergraduate degree from University of Chicago, great school in economics. I would have assumed it was applied mathematics, statistics, and probability. I know those can be concentrations within economics. How did you find your way into this form of statistical analysis? I mean, you have c is a pretty quantitative school in general, but it was really kind of stuff I did outside of of school and outside of work.

So sports in particular was kind of like a lot of applied statistics. You want to win your fantasy baseball league, you want to win your instate tournament, Brackett. And you know, like I said earlier, the data and sports is so good that it's a good way to train your hypothesis testing skills and logical inference skills and stuff like that.

But it's all, you know, it's all kind of a passion jacked and also you know, I'm motivated by um there's that cartoon about oh someone's wrong on the internet. You can't get to sleep right like I'm motivated by that. Yeah, x K c D one of my favorite come to

bed somebody on the internet. But I'm a little bit like the person in that cartoon, right, he's not going to bed because I'm like, boy, you know, you go and read the campaign coverage and in the mainstream press, and it's blatantly wrong think that because somebody saw some one signs and palm beach, that didn't determine who was No. I mean they rely on an anecdotal evidence or or the you know what's funny, which to an article about how how Hally Clinton's problems are at least as far

as the Democratic primary concerned, a little bit overrated. But I went back and looked at the kind of coverage of past campaigns like two thousand, Al Gore versus Bill Bradley and theme for themes, sometimes almost word for a word. You can see the same stories written, and the fact that people kind of forget that history, right. It's not

it's not forgetting right. I mean, they're really smart people working on on campaign coverage, but there's an insane have to to tell the story and to sell the story in a narrative format, and a narrative format, and that's kind of for me. Campaign journalism's original sin, right, UM is it's like a baseball season where things play out really slowly. It's a really long season. Not that much happens from day to day. Maybe you'll have a few genuine events that are um, upset the apple cart, but

not many. But it's hard to write every day of the story. Like you know, today people campaign and nothing of importance happened. Right. So, in the last minute we have in this segment, how did you end up finding ESPN? I'm going to assume your love of sports has made that a really strong fit. Yeah. Well, there were a

couple of reasons. I mean, afterlve the contract with the New York Times, UM expired and we again UM we the Royal We meaning me and my lawyer and whatnot, talked to six or seven different companies, um, you know, but ESPN offered a couple of things that were unique. One is that they realize that I'm not just about politics that to some extent, I want to hedge and diversify what I'm doing because we did get lucky in um.

Another thing is that you know, they have the resources to invest in a company that can grow at a sustainable rate and UM and evolve a little bit. Right. We have twenty five or thirty people now working for us at at UM. You know it's sustainable. We we think, we hope, I mean, you know, we say, hey, look, our traffic is growing. Therefore can we hire this person who will help us to keep that growing? Um. But you know, that kind of more entrepreneurial attitude, I think

is is a good fit at ESPN. You're listening to Masters in Business on Bloomberg Radio. My special guest this week is statistical wizard Nate Silver On, the man who correctly forecast the O eight and twelve elections. Let's talk a little bit about baseball. So here we are. This is being recorded in middle August. What stands out this year is statistically aberrant or interesting or unusual. I think you actually look at the preseason predictions and thank I

don't have to make these anymore. They've had one of the more inaccurate years in a long time. A lot of the set systems said the Royals um were likely to regress heavily to the mean not so much, and they haven't. Right. Um. You know, the Nationals, I think we're supposed to win nine games and they're struggling, uh to break five d Right now, you've seen the Mets to pretty well, which is um, maybe not that's shocking, given how it is New York. You're waiting for the

late season collapse. But given how much talent they had in their farm system, maybe not that shocking. But it's come maybe maybe the first time that's look at all this talent. Well, look, we'll see how it goes. And you know, finally teams have learned how to manage their young pictures a little bit better and a little bit more cautiously. Maybe, and we'll see if this is sustainable for the Mets or not. But that's a little bit different.

You mean, not ruining them early on. It's not ruining uh, not ruining them early on like the famous like Triple, the van Popol years or the Jason if Ringhausen crop and you know, um, so so far it's gone well for the Mets with their with their young pictures famous last words. I guess we'll we'll see how long that that stands. Earlier, we were talking about Pakoda. Um. I didn't realize that was just pictures. I thought that was players. Is so start out with pictures, the idea being that UM,

pictures are much harder to predict than hitters. UM. And so I thought this is later more. Yeah, what why is that? Uh? For a lot of reasons. One is that kind of measures of pitching are more indirect, Right, like if you look at wins and losses, well, it's conditional on how many runs you give up and how many runs give up? Is it conditional on how many hits you give up and how many walks you give up?

So now we know a lot. But you would think you would think pitch is thrown and strikes versus balls. And but ten years, ten years ago, this is not where the state of the art of thinking was. And also we didn't actually have data like now we can say, oh, you know what, um, Jason Berlander, he's not the picture

he winch was. We can actually say, well, now he doesn't throw as hard as he wins did write you have data on every pitch and what the velocity is, what the speed is, what a picture is doing on different counts. Right. UM, So now the pitching UH, forecasts have come a long way past Pakoda. But you know the other thing too, is that pitching is kind of

an inherently unnatural act on a on a picture's arm. Right, Um, Okay, I I pitched in high school, so i'll uh and you know I have the torn real rotated to prove it. But I'm not gonna argue with you whipping a ball, whipping your shoulder around that way for ninety or so throws a game is is not what normal you could You could throw a spear if you're out in the on the savannah hunting mammoths, but you're not gonna throw

a spear nineties seven times. The pictures, Yeah, pictures. And you know, even though now pitch counts are monitored more. Now every pitch, every at bat is contested so much, right, and the strikeout rates are so high. There are a lot of pitches per plate appearance. Obviously, there's a lot of focus on taking walks and played discipline. Um. The thing about pitching that people realize now, maybe not ten years ago, is that a lot of what people think

of as pitching is really positional defense. And you've seen

phases evolved. Kind of in the early moneyball days, people were like, well, it's hard to measure defense, Therefore, let's just kind of ignore it and put a big lot of big, clunky, chunky out fielders out there, and they'll hit a lot of home runs, right, And then teams realized, um, you know, actually a lot of this is positional defense, Like really the impact of having a good center fielder, a good tes not necessarily the slugging sort of batter as they used to be, but the difference between a

hundred yard blooper over the UH over the second baseman's head, depending on where it goes, is either an out or a double. And the funny thing sasonally the same bad contact with the ball. I mean, if you save uh, you know, twenty hits a year with your defense, I mean, you know, if you've got treen or hits a year with your batting average, and that translates to what forty points of betting average, right, So it's really pretty significant,

and the differences can be that large. But you know, one one thing to remember here is that there's always the peril that people think, oh, if you can't measure something there, for it's not Wharton. And once we got better locational data, we can actually say, oh, now we can actually physically measure how much range uh Lorenzo Caine covers or whatever. Right, we're no longer guessing, and it's like, boy,

those are some pretty big differences. Actually. So recently on there was an article about tennis gets essentially the Moneyball treatment. What other sports will benefit from this sort of statistical analysis that currently are are escaping its gaze? I mean in terms of kind of what sports are on that nice part of the learning curve. Um, the NBA is one of them, certainly. I mean, you know, teams like the US TO Rockets are run by Darryl Moore, isn't

m I T guy? Right? Um, you know, we're actually introducing a kind of basketball version of pakoda um called Carmelo, which will debut at some point the next Oh gosh, it's a very clever acronym. It's like career arc regression estimate something like that. But a lot of layers in Carmelo were good with very geeky words. Um. You know, soccer is a sport where there's been very little data collected. But now you see teams in the Premiership have their

their stet heads and whatnot. And obviously the magnitude of the economy of soccer is so enormous right where um where you know that might be the next frontier. You're listening to Masters in Business on Bloomberg Radio. My special guest this week is statistical wizard Nate Silver. And here we are towards the end of the summer in and we just had the Republican first debate, and we're looking at Iowa and New Hampshire coming up. This is quite

a fascinating electoral season, isn't it. Yeah, we've never seen so I know, Um, I'm being a little bit anecdotal here, but the type of um traffic that we're getting on our politics articles in August, which is historically this August year before, Yeah, a year before this is historically about the slowest time of the year for politics news or hard news in general. Right. And it's it's kind of like it's September of the election year, right, it's not

the November peak. But like you know, um, people are fascinated by Donald Trump, by Bernie Sanders, um in a way that's way ahead of where things were at twelve. So I kind of conflicted here. On the one hand, it's probably good for business. On their hand, it's a little maddening, right, And I think, you know, so much of what we say at is slow down, right, this is not gonna unfold today or next week or this month. It's gonna take a while. Slow down, take the long view,

don't hyperventilate about this stuff. And so they're conflicting impulses here. I say that about stock markets all the time. It is it is people wanted, like the hairs on fly Apple is down. Part of this that you have this whole industry of commentators who are asked to say, weigh in on what happened today, right, and are very close to the subject matter. Um, and you know, so to a first approximation, you might learn more about the campaign if you went on vacation for for a month right now.

We just love when reports. We'll get back on Twitter and they say, oh my gosh, you know we're still talking about Trump now. It's like, well, yeah, let's let's talk about Trump for first. What does he do to the calculate? Really, the first question is how unusual of

a candidate is he? So I'd say, right now, he's not quite as unusual as people would say, at least if you kind of made him a ciscal data point right in the sense that first of all, we had in UM four or five different Republican candidates from Michelle Bachman to New ging Rich, who surged Herman Kaine in the polls to about where Trump is UM and then

faded UM, sometimes rapidly sometimes slowly after that. But there's also another tradition of kind of gadfly entire sufficient candidates like Pat Buchanan which Trump belongs to, to UM, you know, and some of them were able to hold on to a segment of the Republican base UM through Iowa and New Hampshire. So I don't discount the possibility that Trump could be with us for for a long time. I do discount the idea that he could ever become the nominee.

Or you think it's highly unlikely that Donald Trump will be the Republican nominee. Yeah, I mean I put the chances recently at which is when you kind of go through and say, well, they're basically six different hurdles that he faces, and say he has shot of clearing a hurtle. Then you get to you know, one and sixty four chance, which is, you know, right, that's that's pretty fascinating. So what about on the other side, Bernie Sanders, what does

he do with this calculus? So socialist running for press, socialis running first time in half a century. I mean, the funny thing is, you have all the media hype is about um a Democrat who's not even officially Democrat and Republican who, let's be honest, it's not really even a Republican, right, you know, I never thought that, but that's absolutely true. Um, I think the Bernie Hillary race is a lot more typical in some ways, where you

have someone to the far left of a Democratic candidate. Yeah, and if you kind of mapped out and kind of um actually mapped out in a very mathematical way, kind of where our Democrats preferences, then you know about a third of the electorate would say we're closer to Bernie Sanders. He's not quite there yet in the polls, but it would not be surprising if he got a third of

the vote nationally. And the thing is, though, Um, Iowa, New Hampshire, you have a lot of liberals, and you had a lot of a lot of white liberals, and Bernie Sanders is doing pretty well with white voters, not so well with his manox African Americans. So um, so it's entirely possible he could win, Um, New Hampshire, maybe Iowa and a few other states could win, Wisconsin and

Oregon and states like that Massachusetts. Um, you know, but there was a poll out recently in Alabama which is where Hilly was beating him eighty one to twelve or something like that, like an Alabama football score or something almost. Um, you know. And um, look, what you don't get from kind of reading the press coverage every day is that by every metric she's doing as well as any non

incumbent has the stage of the primary. Ever, whether she's beating Sanders by twenty points or thirty points nationally, she's still a way ahead. He's already been endorsed by half of the Democratic Congress, basically. But also that you know, these things tend to tighten if you go back and look at candidates like Clinton in the past. So al Gore UM came within a couple of points of losing

to Bill Bradley, right. Um, you know, George Bush and was a sitting vice president, which is kind of analogous to Clinton's position. He finished third in Iowa, lost like seven or eight states. Um, George W. Bush in two thousand was as a much of a juggernaut as you could be. But McCain caught him one a few states.

So that kind of modal outcome was that is that she does lose a few states, New Hampshire being one of the better candidates, right um, and then by um by February March of next year, we're like, you know, what was all the what was all the fuss about it? She's also out raising him monetarily and the money and the money helps too. It might stay with Jeb. Right,

Jeb is the big winner on the Republicans. Jeb has although on the GOP side, Uh, everyone has so much access to capital that that's one of the things that is, I think a little bit different now it used to be that. So look at um Tim plenty uh in two thousand and eleven, right, perfectly plausible Candy Midwestern governor right um, kind of down the middle positions for Republican UM, but he ran out of money, uh, you know, even four years ago, and so he's like, well, I have

to put the plug. Um. You know. Now a candidate like Rick Perry who's raised almost no public funds, like a million dollars, which is Um, you know Donald Trumper called that pathetic. I think, uh, you can, you can tap into a superpack. The superpacks no longer are maintaining any pretense of separating their operations from the campaigns. I think there's no um enforceable legal risk there and so

his campaign can can continue. But that's why I think, you know, we are a little bit in New Territory with a goop slide, just because you have seventeen candidates running. And as much as Democrats might not want to admit it, these are seventeen candidates, a lot of whom have some really impressive credentials, right like governors and centers from big

large swing states as our typically pretty strong candidates. Um, of course, like let's stall go Hio then yeah, uh well, um Ka Sick is a candidate who maybe is a little bit of a of a dark horse. Um, he's done something interesting, which is he's invested a lot of money in advertising in New Hampshire. Now, um, the reason

interesting forging Iowa and going straight to New Hampshire. Well, I think the goal is actually to get some media buzz, right, because the advertising says that hey, um or the empirical literature says advertising has really short lived effects. Right, you see an ad, you might remember it for a week or two, So you should save your money towards the end and then advertise. But I think he wants to say, look,

it's a field of seventeen candidates. I want to stay on the first debate stage, so to speak, get some positive attention, build some momentum. And so he's kind of investing in trying to make a name for himself now. But yeah, he is a case and he can say, look, I'm really popular in Ohio. Boy, that happens to be a pretty important state in the general election. Um, And I'll what it indicates that I can I can win

over abroad and diverse coalition of voters to um. So you know, I mean there are questions about uh, you know, so I'm skeptical about Trump's chances. There's no one Republican you can point to though and say, boy, they've had a great start to their campaign so far either right, Um, Jeb has kind of faltered on the Iraq question a few other things. I think people were generally unimpressed with

Walker's performance. Walker, although I would say, um, Rubio didn't hurt himself yeah, Rubio is Rubio was in win candid where you know I talked about on the Democratic side, come March, we might be saying, yeah, that was kind of obviously I was going to play out right. I think we could say that March of next year. You know, boy, it was clear that Rubio was a really smart consensus choice. Right. He's a little bit more relatable than Bush. He's a

little bit more electable than Walker. Um, you know, he wasn't panicked about trying to boost his poll numbers in August. He waited till towards the end. Um you know, Ruby O, I think I don't know. I mean, you know, Bush has the money, he's a little bit ahead of Ruody on the polls. And no, one can't have a Bush. You can't have a Bush Rubio ticket. You can't. Well, there's some machinations about I guess there are some end arounds maybe, but we you know, I don't know all

the constitution of details there. But basically the answer is basically no, I think you would not have you could have you know, Rubio and Walker, or Rubio and Kasick. You do have a lot of all the way around Kasi Rubio, the senior potentially. So one thing that happened, by the way, is that the GOP had UM a very good election in and then another one in UM, and they won lots of Senate and governor seats and

swing states. So you know, that's the benefit now as they have a whole suite of caanics pick for them that are are popular or at least somewhat popular in in big swing states like Florida and Ohio and whatnot. Quite quite fascinating. So I just said, Kasi Rubio, how important are the VP choices to the electability or the polling note numbers of either candidate? So history really, VP

choices are are not all that important. UM. As best as we can tell, you get about a two or three percent boost in the state where the VP candidate is from sometimes, so even that seems to be fading. Paul Ryan didn't help hit Romney very much in Wisconsin apparently, UM. But there's downside of the VP choice, you know, I think UM, although she was heralded at first, Sarah Palin probably UM wound up further injuring any chance, but came

muld have had in two thousand and eight. Uh, you know, we go back to seventy two and Eagleton and Shriver and that whole mess. Um, so you know, some extent making the um the way Obama played it, where you picked this boring, old, safe, safe white guy basically right, Um, but I know there are different ways to do it. We've been speaking with Nate Silver. If you enjoy this conversation, be sure to check out our podcast, where the tapes keep rolling and we keep recording until the guests pass

out from exhaustion. If you enjoy this conversation, be sure and check out all of our other interviews. Follow my daily column on Bloomberg View dot com, or follow me on Twitter at Ridhults. Your your Twitter handle is at Nate Silver Silver, Nate Silver Fight. What did somebody else grib Nates at Nate Silver? I think so, yeah, that's that's a shame. I'm sure you're you have the blue checkmark, so you've very got the blue check mark and um, but you know so our website is actually five spells

spilled out, not the number. I'm Barry, Ridhults. You're listening to Masters in Business on Bloomberg Radio. All right, welcome to the podcast portion of our show. Nates, thank you so much for doing this. I'm really I'm excited about this and I've been looking forward to for a while for having me my pleasure. So there's so many different questions we we didn't get to that I really want to talk to you. This is the big, big one

will save. So the one quiet Mike bat Nick in my office is, uh said you have to ask him this question, and it's how important are managers to team's successful failure? And how do we measure that statistically? So in in baseball in particular, I think you can talk football also because that that that comes up quite often. I think in baseball, First of all, the strategic choices

of manager makes are not very consequential, right. Really you can debate like, oh, how MU should you sack bunt or you know, um, some guys now have the picture hitting a The impact of that stuff is probably not more than a couple of wins a year over a hundred two games season. Um. What I think is understudied is, uh, how much influence a manager has on getting the best out of his players? As simple as that sounds, right, Um, you know there's been some studies done in the NBA.

When you say take a preseason prediction about a guy's ciscal line that doesn't know who the manager is. And then how often does uh, Gregg Popovitch, for example, get his players to outperform that prediction, either by by improving their skills or I fit up feeling a all better than the answer is, you know, routinely the Spurs and way that's disically significant beat their kind of raw um projections from name, not scars from Vegas. Vegas can account

for that um, but from like a naive algorithm. Right, and so I've not seen that done for baseball, and I'm sure someone has, but kind of saying, you know, who really motivates their guys to do the best, that's still kind of a black box a little bit. That's fascinating. I have a mixed um relationship with sports books. Some are better, some are worse. The one that I really liked, which is right up your alley. Did you read Tom Coughlin's book on the Right to Win? Former coach? I

did not know. So the one statistical thing that stayed with me from that book that was so fascinating. I don't remember which player was complaining about it, um, uh center linebacker. That was you know, they would run a whole bunch of numbers for every opponent on second and long. Here's what they like to do on third and one, here's what I like to do on in the red zone late in the game. Here's what they like to do. And they they would just bury these guys who just

weren't used to this sort of mathematical stuff. And he always, I wish I can remember who it was, which giant um that one a championship with him, and complained about this constantly, and then gets traded somewhere else and it's you know, third and one. He goes, all, right, what are they gonna do? And I don't know? And it dawns on him that, oh, this conkling cockling guy was fantastic, that that data was really helpful to know what the Certainly it's not a sure thing, but hey, they like

to do this at at third and one. And I got to think that that how do you measure? Yeah? I mean football is a different story, right. I think everyone from stat heads to UM coaches themselves would say that NFL coaches are pretty darn significant. Yeah, look at

get the Seattle UM the Super Bowl. This year. Yeah, you know, well, so this is you know, we would say that there was some very long announced we did that the choice that Pete Carroll may was fine, you defend that, right, I see, I looked at it as he was playing a U a high risk, lower probability game. Consistently, if you look at a lot of the choices that were made, you start to veer into the anecdotal Well

what about this and why did they run here? And why did they But you know, something like one percent of passes get intercepted in this situation, so it's kind of a a rounding error really, and there's a lot of kind of you know, post facto sure analysis. Right, no one would ever have complained about that play call for a second, if he caught the past, if it was if it wasn't intercepted, it's a whole different Well, to the to the victors not only goes to spoils,

but the opportunity to write. Um, So we we mentioned intangibles, how do you measure the intangibles of a player? And and again mentorship, attitude, locker room behavior. Can you ignore

these things? What do they actually actually mean? So I've been thinking about ways to kind of set up a um uh verifiable experiment for this, and one might be if people identify in advance, say twenty baseball players they think have strong or weak leadership, and then see what happens when they change teams and as there's some residual value you can identify, but you know sort of that can you really another ways unless you really run a

control group, it's pretty hard to. In baseball performance is pretty individualized. Like the NBA is different the NBA. The whole challenge the NBA is you have um you know, one possession per possession, right, one shot per possession. You have to get the guys to cooperate. They have selfish incentives to each take the shot and boost their stats or whatnot, right, And so that makes coaching important, It makes um chemistry important, and you can to measure this.

You can kind of say, uh, what residual value is there when a guy is on the court or off the court. And there's some better and better methods for doing that. Now it's some of the guys who UM who play twenty five minutes a game, they're like, oh, they're defensive players. It might have been ten years ago that the stats were like, oh, this guy can't shoot right, you can't really do anything. And now we're seeing actually they often do have a big impact on the game.

But baseball I'm a little bit more more skeptical, in part because I think sometimes, um, sometimes the guys who um have a rep for being good clubhouse guys are just guys who are friendly to the media. And that's

a slightly different thing necessarily. Sure. So in basketball, you know, I always as a nick fan who always felt thwarted by the bulls under Michael Jordan's I always a picture him as a guy who doesn't put up with any nonsense in the locker room, and he demands top performance, and he drives the rest of his team the way that Patrick Ewing never did. It's arguable whether anybody since Michael Um, since um, anybody in the Lakers ever did this,

since Magic Johnson. Um. You know, you don't really know, you don't really see Kobe Bryant is that sort of player? Is that just anecdote? And and after the fact, or did he? Is he the sort of guy that really had that impact? And is there any way after the fact we can we can determine that. I mean, one weird thing about and I grew up a Pistons fans, so, um, you know, everyone hated Michael Jordan for reason or not.

I remember in the in the final Championship E one was it I kind of involuntarily found myself rooting for him, right, And I'm like, I can't root for Utah, come on, right, Um? But those contents and letters to Nate Silver at ESPN

dot Com. One thing that I think helped Jordan's reputation, apart from being maybe the best best player of all time, that he didn't have a lot of near misses, right you know, Uh, the Bulls were probably the favorites and five out of the six UH championship seasons where they won the championship, right, um, but they were blown up in a hurry, right Um. The year before they won their first one one, it was a really strong time for Eastern Conference kind of sat out the year and

a half, right Um. So for some reason, it almost feels to me like if Jordan had UH won five championships and ten tries, he would look worse than six out of six. But that's a pretty petty complaint. I mean, you know, Jordan had an amazing career. One reason I kind of like the NBA and contrast to other sports where there's so much randomness, is that in the n b A there are not very many undeserving NBA champions right, we saw in the not a lucky call or a

bad bounce or something like that in tennis. Really, if you want us opens coming right, um, but you have to earn the NBA championship. We saw in the NBA Finals last year the Calves playing the absolute best of their ability and they can win. They can win two close games, right, It's really hard to win a seven game series against the team when you know your second base player, best players j R. Smith or something more

power to j R. Smith. But but so in other words, Lebron doesn't have the same supporting cast that Jordan's did, and the Calves need to do some offseason. Well they've done some stuff, they need to do some more. I mean everyone's if everyone's healthy, they're a good team. I mean it should be really fanned. That's a great era

for the NBA. By the way, between the Spurs and the Warriers and the Calves, um, you're gonna have three great teams are all very different, looks different stylistically, so you know, Um, you know, my boss just signed this big new NBA contract. Should have a great season coming ahead. Um, and you said you mentioned tennis. Let's talk a little bit about some of the the male players who I think. It's been fascinating watching this go back and forth between uh Na Doll and you know, go down the list

of the top five players. There's been this sort of rotating how fast do the skills deteriorate? At age thirty? It seems like these guys peak nine and then they start, you know, just losing it a little bit. I think twenty eight or twenty I might even be a little bit early. Obviously, you have guys like Samprus that held an agacy that held on for for longer, but maybe more like like twenty five. I mean, I go to

the US Open every year. I'm not a huge tennis fan, but I think people who are watching on TV don't realize how physically demanding it is. A five set match, right Um, in the heat in in New York in September, or in Australia and in the you know what summer down there, right Um, these guys are hitting at a hundred thirty d and it's insane. And where that there the guys now actually play a you know, really good defensive game. Um, you know, very very challenging if you're

not absolutely in the top of shape. Right, And so in general happens is that for you look at baseball where the data is good or something, Um, you know the average player peaks at twenty seven. What that really is is at your fiscal attributes probably peak actually at

at age twenty three or something. Your mental attitudes keep growing. Um, that makes up for the physics you have the best overall combination for And if three of these guys peak at twenty if you look at like an NFL running back, and um, guys were just kind of purely about brute strength and how beat up or not your body is. Right, Very often in NFL running back is as best as

we'll ever be in his rookie season. Right. There's a quarterback conversely, where obviously the arm streak matters, but the mental parts important to a quarterback. Of course, those guys can be perfectly good, sometimes even better in their thirties. You could, you could deflate football's way into your thirties. You can keep it. You could. I don't think there's in any age limitation. But I love the stories about Jerry Rice about how he used to train in San Francisco,

and you talk about the mental out of it. He used to invite people from the opposing teams to train with him, and he would just run up and down the hills of San Francisco and he would destroy these guys. They like could not keep up with him, and then they seem on the field and they would be horrified. Stop and think about that sort of head game. I just find that, uh, fascinating. But he's a guy whose career went you know, quite quite deep into his thirties.

That's um. Is that unusual for a receiver for receiver is? Yeah, receivers are not quite as early peaked as running backs. But um, but you know they're definitely mid to late twenties for for the most part. Well, that that's amazing. Um. So tennis we mentioned is the is one of the next things. And soccer, how are they going to change? How are you going to change the statistical analysis of of World Cup football, of of soccer? Well, how do

you how do you look at that? Um? I mean historically the only data that was collected, we're just goals and bookings and yellow cards and and red cards. Right, that's it. That's changing now though, Right So now, um, you know in the big four or five leagues in Europe, paying what you think of the French League? Um, you

know you have lots of real time data collectives. And now we ran a big article and we know Messy I remember that where it's like he actually is saying, now, okay, now we can actually measure all these things that we thought of as intangibles before, right, like how well does he set up his teammates, how much space does he cover? Some things are actually a little bit kinter intuitive, like

I think Messi doesn't cover that much territory. He's very much kind of thinking about how can I have the most impact and being in the center of of the pitch more or less to kind of be the focal point for for the offense. Um. But yeah, I didn't play in the World today Messi for sure. Yeah. I mean, and people people don't realize how good he's amazing he is. Yeah. I'm a World Cup fan for a long time and I find it fascinating and I wish we had the

World Cup every uh every two years. I actually went uh to the Women's World Cup and in Canada, which was a ton of fun. Right, Um, to see Americans actually win something at soccer was spectacular. So, UM, all right, so let's see what else we've forgotten or I skipped over in the prior segment. Um, So we'll save the election stuff for a little later. And I'm gonna I'm gonna touch that. We did that, we did this. Wow, there's so much stuff I could I could stay on

sports for a long time, but I want to. I want to bring it back to um, bring it back to to what you're doing with five thirty eight now and a little bit of the history with that. So your name to one of Time magazines most influential people in two thousand and nine. Then comes the Times, then comes the the election. Personally, what's it like blowing up like that? I mean, I used to be disorienting and

kind of like what's going on? So one thing is that, um two twelve, you're so busy that you kind of don't have time to process it almost all really, so he doesn't, Well, my book came out in September two thousand twelve. Um, and so you know the combination of that, and of course the you know Penguin. My PR people had me doing shows like this like like every day pretty much, right, and then there was random stuff. Hurricane Uh Sandy hit at that time, so also's going on.

I was like kind of commuting in the dark from my uh apartment then in Brooklyn to to the New York Times office sort of do a media hit or whatnot. It was just kind of crazy time and then you emerge from it. And I was like at some conference in Chicago the weekend after the election, and it was like like every time I got the elevator, someone would like recognize me and be like, hey, are you Nate Silver? Right? And that was that was that's pretty weird. Yeah, so

I do less TV than I used to. But the one place that I got to ask, if you have this experience, if you ever check a bag when you fly and you're waiting by the carousel, you're just fair game. You're you're fair game. For some reason, it happens a lot in airports. I think a lot of people watch serious quote unquote TV and and airfats on the planes. But um, but it definitely there's a half life based

on how recently you've been on TV. It probably happens, you know, a couple of times a month still, but not daily it used to be used to be. Yeah, yeah, and certainly we may we may see that pick up again towards next year, towards we'll see that although I'm trying not to be uh, I don't know, I'm trying not to overexpose myself too much. That makes sense, And part of it is that you know, you've probably learned

this too. But if um, if you have your own platform right and we have the opportunity to reach a very large audience at five eight every day with the writing and the stuff we produce in house podcasts and videos, um, you get very finicky about how you're presented. You know.

It's really easy to kind of come across as oh, here's this kind of you know, NERD is a know it all and you know, and we trying to have a little bit more subtlety and how we're presenting our view of how campaigns and elections work and so you know, um, so I'm trying not to uh go to overboard with the with the media stuff. So what I certainly wouldn't argue with Nerd. However, you never struck me as someone who presents himself as a know it all. In fact,

your whole approaches. We don't know what the future can bring. We don't understand which of these polls is gonna actually be right. However, we can look at statistically taking the average of all. Do you see how that didn't did in the best adjusted for some minor modifications, and come fairly close to the actual outcome. That's a humble approach, not an arrogant. Appreciate that, thank you, But right, was

that part of the plan? So I think it's you know, it's kind of how do you wait, what's your baseline? Prior's right? Um? You know in a general election, the prior in a campaign where the incumbents running is that the incumbent will probably win unless things are pretty bad. In an election with the incumbent, it's probably this is going to be really close, right, Um. You know in a primary it's maybe a little bit more difficult, but you know, the history is that the establishment usually wins

the primary. In the primary, Um, you know, the candid who it's a consensus building process, a nomination process. Literally, it is a democratic party's nomination and the gops to bestow they kind of set the rules for how delicates are allegated and what happens. Right, So they have a lot of influence on the a lot of influence, right.

You know, if the GOP doesn't want to nominate Donald Trump, it's actually not a popularity contest, Right, they can make it very difficult for Now and Crump to do it. And there are maybe one or two cases. So McGovern in seventy two, UM was very smart about getting UM a lot of grassroots support from people who would be delegate at the Democratic National Convention and kind of UM he had designed the system for how delicates re alligated and so he really knew allocated, so he knew how

to UM how to work that system really well. But seventy two is kind of the only example of where, uh, the inmates started burning the asylum. What about oh eight, Um, wasn't Hillary sort of ordained in advance and sort of that? So I'm upset the apple card a little bit. So Clinton in two thousand and eight had ah, you know, at this point in the race, like a ten or fifteen point lead over Obama. This year it's kind of

a thirty point lead over Bernie Sanders. But it has more let's do with Hillary and more to do with uh with Barack Obama versus versus Bernie senators where Barack Obama represented a whole, huge, kind of mainstream part of the Democratic coalition. Um, obviously he did very well with African American voters, which is a huge voting block in the coalition. So you know, um, you can basically split the Democratic Coalition into the third so you can have uh,

white liberals, white moderates, and non white voters. Actually it's more of it really, right, and um, you know he had better support across coalitions. But also he about the support from Harry Reid and whatnot. Right, lots of influential people in the dipretic party said said, we have an amazing cadate in Obama, right, and we want to make

it a fair fight. Whereas this year there is not a single sitting Democratic legislator excuse me, senator or representative or governor who has endorsed Bernie Sanders, in part because you know, he's not a Democrat. Um, he's seventy four years old. He probably lose general election by by ten points. Right, Um, you know, uh, what about Buden, what about al Gore. Is anybody else got a credible gonna throw that hat

in the ring? Or is that just early chatter? I mean I think, you know, let's say that Hillary Clinton, I think it's not likely. Maybe it's a ten percent chance. Right, Let's say these scandals are more serious than they appear to be the email or something else, maybe something else. I'm not I'm not totally persuaded. So you know, baseball, there's just set value over replacement player. Right, It's like

for me, it's like valuable replacement scandal. Right, We're gonna you see over the Clinton's that there's something like slightly funny going on, Right, So maybe something that's a given in Otherwise, Yeah, the placement scandal, I'm not sure that this is that remarkable proved to be, you know, the Republican Control Committee. But I have Republican friends who are in d C who insist the email scandal is totally different than it's. Sorry, this is more serious. You've seen

a long decline in Clinton favorability ratings. It's really hard to determine cause and effect and say has that been accelerated by the email scandal or it certainly isn't helping um but overall, you look at markets and their betting markets, their perception of where Hilly Clinton stands hasn't changed. They've had her with about an eight chance of winning the nomination for a long time. It's not been affected except

at the margin by by Bernie or by the scandals. UM. You know, Biden, UM or someone like Biden would at least be a little bit different in that UM in that let's say that Clinton is not doing well, Democrats would still, i think, want to intervene to not nominate Bernie Sanders. Whether they could say, well, Biden, he's fine, right, you know, Biden will give us a shot at least, and so you know, that would be one of the more significant developments so far. From what I've read, it

all seems pretty speculative. I think part of what Biden wants to do is say, you know what, UM, it's tricky because it's August right now. It takes a long time to mount presentational campaign. What happens if there is maybe it's the email thing, maybe something new, maybe it's

a health problem. What happens if Hiller Clinton, if that happens in November, and then UM and then only Bernie Sanders is on on the ballot, right, then you have a deeply damaged Democratic nominee in Clinton, or someone who the party will not want to nominate in Sanders. That's where you have the kind of break glass for emergency Biden, Gore,

John Kerry uh type candidates. Right. Um, And so I think part of what Briden's trying to signal is to say, look, I'm here, I'm around if everything else is going wrong. You know, I'm at least thinking about this. I have a few people who are loyal to me. I can mount a campaign on the fly that's very different from actually run. And the way he sends signals it's very much like kind of his inside baseball you know, Marine Dowds Sunday column Right, those are kind of dog whistles too.

I think other Democrats, more than to like mainstream voters, like, hey, if something really goes wrong with Clinton campaign, then then I'm here. But I would say, you know, um, we are speculative or dismissive. A lot of speculation about Clinton is struggling. If Biden ran, that would at least be a more tangible sign that that some influential Democrats are

worried about Clinton. So right now. Statistically, if we had a guess, you're guessing Hillary Clinton is the Democratic Yeah, and you know, like I said, betting markets put the chances at eight percent, I put them marginally higher. I think, um you know, and I think of the of the percent, most of it's not Bernie Sanders. Most of it is that um Al Gore, Biden, Biden carry Gore right are the kind of the three people who could step in

and in an emergency. So it's fascinating that you say that when we were looking at the O eight election, And I'm curious that you of your perspective. My thesis was at the time, so the Iraq War by then had turned sour, the economy was in a free fall, we were in the midst of a financial crisis. The running on the incumbents track record, I said, no matter who the GOP put up, they're willing to get destroyed

by whoever the Democrats put up. But people have argued with me, well, if they would have nominated somebody instead of McCain, and someone less charismatic than Obama, perhaps it would have been a very different outcome. But I mean it was about the most difficult imaginable set of circumstances

for the Republicans to win. I mean not that I feel sympathetic for them, because you know, they had a president who made some mistakes, shall we say, but you know, you had the economy collapsing, a very unpopular war, the sitting president had approval rating. Right, it wasn't a hugely low number for a sitting president. You can you know, one could argue about my ship won by twelve points

instead of seven or something like that. Right, But that was about you know, there are some years that political scientists debate and say, you know, uh, he kind of was perceived to be kind of still in recession if you look back on revised cistics. Now it was actually doing. The recession was over before anybody realized. But oh way, it's one of those years where I don't know, Democrats would have had to make a huge error. Uh. And in some ways McCain wasn't a bad candidate. He was premoderate.

He broke from the GOP in a lot of ways of the the time when that party was really unpopular. Right. You know, um, I thought he was a better candidate in two thousand. He seemed to be less captured and therefore less well, that's part of people are are kind of concerned about which of these seventeen Republicans will get nominated to some except that all kind of get except

if it is a Trump or something. Right, Um, they're all going to kind of have the stamp of the party on them anyway, Right, And the ones who are more moderate Jeb Bush at least by the times you gets nominate, would be pulled more toward UM, toward the right. Scott Walker might be pulled more towards not the absolute left, but you know, become a little bit less conservative on some issues, and then they'll kind of go and do

what they need to do for the general election. But um, but to some extent um, you know, it is a party driven process, and the candiates own policy positions might be swamped by what the consensus is among Is it still true you you um run the primary to the right and then um run the main election to the center, or for the Democrats, run the primary to the left

and then tack to the center. For this, I mean that's the default, right, I mean I think, uh, you know, I think some partisans would say, well, it's all about turnout now, so you want to motivate your base, right. I'm always a little bit suspicious of that kind of the notion that where you can kind of have your

cake and eat it too. Right. Um, you know, it maybe works a little bit better for Democrats in the sense that there are just a few more Democrats and the electorate and Republicans, so if everyone turns out, Um, they might have an edge out of those numbers. Different what what's the GOP percentage? And how accurate is is that about independence? I always thought people just angry at

their party. A lot of people say their independence aren't really independence, right, but you know Democrats have right now about it five or six point edge and party identification, but it doesn't give them any well, els would be a whole another hour long segment right that time. Um, you know, but there are a lot of qualifications to that, one being that they Democrats get a lot of people who are only marginally likely to vote. Um. You know, if if we had mandatory voting in this country, a

lot of things would be different. That's weird hypothetical, right, Um. Without mandatory voting benefit who would benefit Democrats? The Republicans on the other hand, you know a lot of times um, people won't turn out necessarily. Also, one thing that's a little bit tricky is that, um, because the GOP has been unpopular, Um, a lot of people that formally would say, you know, I'm a Republican, now they'll say I'm an independent, but I lean Republicans look at leaners and it gets

a little bit closer still. Um. So I couldn't completely relate to that because we were talking before the show. I grew up. Jacob Javits was our senator, and that sort of center right republican you Kasik is probably the closest thing we mentioned. You know, Bush has actually surprised me to the extent that, um, he has run pretty explicitly as as a moderate right, which is an interesting approach.

I think maybe he figures that, look, I have a very long track record in public life, right, Um, I'm not going to fool anyone like Romney tried to do in two thousand and twelve, right, Um, and then looked like a flip flopper in the general election. You know, they're always cases like The one thing that amazed me is that in two thousand and eight, all three of the major Democratic candidates were against gay marriage. Right officially even though Obama had said, like in two thousand two

that he was pro gay marriage. Right. So I'm like, you know, are you really fooling anyone? I'm not. I'm not totally sure. And so I think Bush is saying, you know what I am? Who I am? I'm a moderate conservative, right, Um, you know, I'd be very electable.

Although this is the thing that's tricky for him is usually you'd have a candidate like that, um, like the McCain type candid who's a moderate conservative and they have good UM electability numbers, and Bush because the Bush family name is still not that popular, maybe he's not seen as that relatable. You know, He's head to head numbers are not any better than like the more conservative candidates against Clinton. Right, So let's talk about the head to

head numbers. So when we look at Clinton versus fill in the blank, Hillary versus Jeb Hillary versus Rubio, Hillary versus Kasik, who stands the best shot at at winning? Well, this is what we talked about. You know, I probably would not look at those head to head numbers very much right now. They're almost meaningless. Yeah, if you're gonna look at them, you know, candidates who have comfortable name recognition to Hillary, which is kind of almost no one.

But you know, for that reason, the Bush number, you know, I would discount almost all of it instead of literally all of it as I might for like Hillary versus Rubio or whatnot. Um, but usually there's correlation between how

moderate a guy is and how well he does among independents. Um. You know, because Bush isn't personally seeing that favorably, he might not have that edge this year, which is kind of why I say, if you're Republican, you might say, well, um, you know what, if we look at Rubio versus Bush, then we get a guy number one who's more conservative, right, number two who is actually a little bit more popular with independence, and number three avoids this whole voice versus

Clinton dynasty angle. Again, it's not my job to advise parties, but that's what Rubio is one candidate where that you could kind of go back and say, I totally get why they nominated him and what we're all discounting somebody. I guess I'm you know, I'm bullish on Rubio's You know, here's the thing that I find fascinating. Bright, Hispanic articulate, like a really like and you know, has almost a

Kennedy esk photo genetic photogenic sort of thing. But when you look at his policies and I know the Republicans are I don't remember was you or wrote about the gender gap on the GOP versus Democrats, he is surprisingly hard right. If he was a more centrist, moderate conservative, I would think that he would be the front runner for for getting elected in the general election. Well, and

we'll and we'll see, right. You know, one way this could play out, um, is that people sour on Bush and then you'll see Rubio maybe move a little bit to the the center. Despite what he said recently about abortion, no exceptions for rape, incest, life of the mother, Like you don't hear that from mainstream candidates the past twenty years. When he is pretty and there are fiscal systems to the try and quantify how conservative people are, and Rubio is very conservative. I mean, the GOP is a very

conservative party. So relative to the other seventeen people, he's about in the middle of the GOP field. But he would be, you know, twenty years ago considered very conservative. Um. But you know, um who's more to the right of Rubio, So Walker probably, Um, you know Walker is UM far enough to the right where he might kind of compete for votes with UM with Ted Cruz a right and Carson for that matter, right, UM, which you know, maybe

isn't a good place to be. There is gonna be some support for candidates like those, I think throughout the race. What's interesting about Trump is that, UM, he's really difficult to peg Ideologically, He's kind of all over the place. And his support too, was all over the place. Right, people with somoa's capturing Tea Party voters not true? Right. That coalition he has is drawn from all different kind

of ideological parts the GOP in. What's one reason I'm scipticle by him is that it's also drawn from people who don't traditionally turn out and vote in primaries. Right. It's a lot of people would say, you know, I like that kind of six it to the establishment and six it to the media. Right, they're fascinated by him. You know, does that translate into driving to the Iowa caucus in the snow when you've never voted before. You know,

we don't know, there's some reason to be skeptical of that. UM, but Trump is you know, if you kind of if you average out his policy positions, then they're kind of, you know, fairly typical right of center. But that means you have some things that are like radical Tea Party and some things that are like practically like socialists. Yeah, and it kind of averages out um and that is interesting.

I mean, you know, there were some article right today about how um so that was smart, uh, about how usually a candidate is so constrained right, and these disagreements remember in tuth as an eight the Democrats on on healthcare right, Democrats had huge fights about you know, did Obama have a employee mandate in his healthcare bill or

a mandate and Hillary didn't. I mean, these are you know, pretty minor differences, whereas Trump is saying, I'm not going to play by those constraints where you can be you know, an eight point two or an eight point five on something right, I'm just gonna be all over the place.

And that's how a lot of real people I think to write, um right there, their views aren't necessarily in tonally consistent, So so I would I would dispute that, like I think, you know, I don't think the party's views are that internally consistent either, Like I'm not sure why, you know, policy your views on taxation and gay marriage and um and uber and the war and Iraq should be correlate with one another that much. But you know, certainly, um,

you know, these views coalescent to parties and people often. Um, you know, our Democrats are Helbicans for like one or two big issues, and they kind of say, you know what, it's a lot easier to kind of agree with the party on everything. Right, So I kind of disagree with the idea that, oh, um, trumps use are more incoherent, because I don't think anyone's views are all that coherent necessarily right, but definitely it's it's maps differently than any

other candidates would. So at this point, given all the changes, well let's let's use gay marriage as an example. Yeah, the Supreme Court has ruled. I thought Case six. Answer was, and by the way, this is a kiss of death if I like you when a Republican primary, you're done last year. I thought, last last election, I'm like this huntsman guy seems pretty yeah, toast, that's the kisses I thought Case six answer was, Oh, that's who the responsible

adults on the stages. It's him, So you know, he's not. And a lot of people who are running cases campaign ran Huntsman's campaign. Um, same people. It's a lot of same people. And their strategy is, you know, you want to kind of appealed to to the media. So the first debate was interesting in that, um, you know, the media spam was that case that had done really well, right,

And I thought he did pretty well. I personally liked him, But I'm somewhere probably near you want to play the spectrum, and I know, right, um, socially progressive, a little, a little fiscally conservative and and to me, I I've assumed the abortion issue has been settled. I can't believe that's still ongoing. And and now it looks like the question of of nwage equality is settled. And the only one on the stage who seemed willing to say that was Casing.

But if you looked at Google searches, which they now releas stayed in real time, Ben Carson was doing really well. People were interested in him. Cross is much less crazy than I expected, not that that's saying any He was pretty he was pretty down, toned down and mellow, and I know I thought that were okay, right, Um, yeah, for sure, when he's gaining the polls a little bit. So let's talk about Carly Fianna, who did really well

at the children's tables. People have derisively called it, but I know her as a horrible CEO oversaw one of the worst mergers in technology history, you know, HP and Compact is described like a O L Time Warner. It was just a disaster, and she's laid off. I don't

remember the number, a hundred thousand, some huge number. Can she really be a credible candidate given that background, she she's managed to fail And again send your hate email to Nate Silvertie, but I think she's failed upwards and I'm wholly unimpressed with her as a candidate UM or as a as a CEO, as a corporate I thought she did well in UM in the debate and the

JV debate. It was very self possessed and kind of right understood the balance between you don't want to sound like a wing nut, but you want to make sure that you're memorable on that stage. People are kind of hap paying attention, you know. I know, I think it's a little bit premature. I think she's um she's not quite yet at the point where you're gonna see a lot of scrutiny. Um. But yeah, there's uh, you know, this is one of the things about comparing like Hillary

and Bernie Sanders, right. You know, if you compare a candid who has um received a lot of scrutiny from the media, from other members of their party, from the other party, versus one who hasn't, it's a really apples to or or just comparison. Right. Um. You know, Sofia a Mina has not yet been through that scrutiny phase. Right. Um. If she keeps doing well, then her reward is that she'll she'll then endure that, right Um. But you know it's not I don't know. I would think she would

have more chance me the nominee than Trump. Maybe you'd have to think about what about her is the vice presidential candidate. If Hillary is the nominee, put a woman on the ticket, I would imagine that would offset, um,

some of the gender gap that is inherent with Hillary running. Sure, although women don't necessarily um support women who aren't good on on quote unquote women's issues, right, I mean, certainly not every woman is pro choice, right, but you know what many are but but many are, right, Um, and so you know, Republican women might have a more difficult

time of it than than Democratic women. And what since, uh we're talking about Carly she kind of um caught some positive attention for her comments about the whole Megan Kelly Fox thing. Yea, how does that play into this? Uh? This is one thing that I, uh, one bad prediction we've made, right, I kind of thought that, boy when um, when Fox News started taking on Trump, that Fox were hammering and that Fox News wouldn't back down, and they did back down really really quickly to I mean, so

you sound as surprised as I was. Yeah, I was surprised because I kind of thought that. But look, Fox News, like any major media organizations, a complicated place, right. Um. You know, they want ratings and attention. On the one hand, the their hand, you know, um have some influence in Republican politics, so on the some Yeah, on the third hand, you know, you have lots of individual producers, some of whom are awesome journalists and some of them are not.

And you know it's a complicated place. But um, but you know, it was funny how explicit Trump was about the quid pro quel like, I got you terrific ratings right and made you all this money, and you're giving me this right and I'm like, yeah, you have a point, you know, And um, you know, it was amazing. I thought Fox did a great job on the on the debate. I watched about an hour of it, maybe a little more,

and at that point, you know it started. You know, it's when you're watching a game and it looks like a blowout and all right, I don't need to watch the stay for the fourth quarter. That's kind of but it seemed really interesting. The one name we didn't mention, um who I could give you a whole laundry list of reasons why he's terrible and I don't like him. But I thought Chris Christie did a really good job at the debate. So the Christie story is kind of fascinating.

We've been on um on the Chris Christie's toast man wagon, I recall, for a long time before it was cool, um, you know, in part because he's a guy who the party wants, Kenny, who's conservative but also reliable, right, And I think Christie is he is a guy who's not that reliable. He's no longer very popular with independent voters. But but who knows. I I almost think everyone at

the office dis agrees with this, right. I think it would almost help Christie if um if he dropped out of the top ten and then had the stage to himself at the next JV debate, right, and could be alpha male and totally dominant. I don't think Christy has is very likely at all to win the nomination, but he's a guy who could have a surge right in the media. The media is fascinated by Christie. He's a fascinating guy. He's great on his feet. I thought he I thought he hurt Ran Paul. I mean, he really

came at him hard. And then so there's guys who you know, there's Christie, Ben Carson, Ted Cruz, the three season. You know, they're all in the category of candidates I think could be the next candidate who searches but probably couldn't win the nomination, whereas Rubio Kik kind of slower and steadier. I think Rubio is a one candidate where you get the sense that he's not trying to win the nomination in August, which is probably a really smart thing to do. But my general impression of him is

that he's a little young, he's a little green. Some of the other candidates that compared him to some guy at this term in his first Senate term named Barack Obama, which I find am using. But you could see there's a political um uh, there's a wind at his back that that could help him. If not this, uh not,

then certainly. Yeah. Though I think one thing a lot of UM Republicans have learned is that UM and a lot of cands have learned is that you know, there's no time like today, right, you know, I'm sure are a lot of candidates this year, UM where you know Rick Perry, who was someone who had one of the better chances to beat Romney until he kind of imploded, right, you know, he was not in the top ten. Rick Santorum, who kind of sensibly finished in second place, was not

in the top ten. So there has to be a lot of these candidates now we're kicking themselves and saying, you know what, Um, Romney a little bit underrated as candidate maybe, but boy, you know it was a much easier nomination to win in twelve and probably kicking themselves for not having run four years. Why do you say undernominated as a candidate. I'm curious about that because in the end, it's hard to beat an incumbent president. Um, the economy improved to the point where it was okay.

I would take the other start of the arguments, say, when Romney was running against Obama, you had a very unpopular or at least it appeared to be unpopular. Um, Obamacare. It's since turned out to be quite successful. But here was a guy who put Obamacare only was Romney care in his state and the weakest economy, the weakest recovery

you've seen in half a century. Certain, it was a gimme, but like like the average elected in president wins the second term by by eight percentage points in the popular vote, No Ababa went by four three point eight or something. Right, So Obama kind of did underperform, But it was a huge electoral college blowout. It was so the electoral college can make It's kind of designed that way to make a relatively small edge and the popular vote be bigger.

UM in the electoral college, right, Um, And probably I think his campaign helped at the margin and some of the swing states to UM. But I don't know, I mean, UM, you know, I don't think Romney ran a terrific campaign. I think he was not a disaster though, either. And I think people, um, you know, again, you go back to the fact that historically incumbent presidents are re elected seventy percent of time UM and this was just one

of those. And a lot of subtle ways, like going into like UM a college football stadium or something, right, UM Michigan Stadium where UM US, at least it used to be University Michigan was tough to to beat there, and it kind of manifests self in all sorts of subtle way us to get better officiating, you know, the playing surface a little bit better, a little bit better

under pressure. It's tough to beat an incumbent president. And I'm not saying Romney was the most horrific nominee, but these guys get kind of tarnished with the loser brush or John Kerry is another one where he lost by by two points. You know, people are right in the middle of an active war, which at the time UM that war was somewhat popular, right where the plurality of Americans in the four right six months later, different six months later, I think um Bush might have lost that

election was really close. But um, but you know, at the time, the economy was had recovered from a mild recession, um and looked not bad, right, um, you know, and you had a war that was becoming less and less popular. But um it it's easy to have hindsight bias now right. And their models part of the two their models that look at factors like the economy measured in various ways,

you know, war measured in various ways. Incumbent see and most of them said, oh, you know, Carrie should lose by by several points, and you know he did about as well. I didn't realize it was that close. I thought carry had lost much more substantially Democrats, because you do have this kind of tied shifting in how Iraq was perceived and the Bush presidency was perceived, right, um, And I think you had Democrats kind of saying, boy, how can we not win this election? And that's kind

of parallel to two thousand. I didn't hear a lot of Democrats saying that to me. I thought it was sitting President nine eleven, still very fresh in memory. You're you're the old expression I kept hearing was you don't change horses mid stream. You're in the middle of an active war, and the tendency is to reelect the income in in the middle of an active war. Is that is that overstating the circumstances. Well, war, it depends on

it's a popular war and an unpopular war. Right in Iraq was going from a war that was initially very popular, uh, to one that wasn't And as easy as amazing as it sounds like, it's easy to forget the impact of September eleven, um, you know, and that still cast a real shadow on the way the race was contested to so things about Vietnam and seventy two and next we got reelected. That was a terribly unpopular war by seventy two, by the late sixties it was unpopular. But it just

seems to be a sense that's not well. S two is one of those cases where the kindo was actually not so bad and seventy to um. But that is a case where the Democrats blew any chance they had by nominating McGovern And in this convention where it went until three in the morning that the world gave his speech and they picked a VP who they had to then veto. Right, I mean that was you know. So, um, so people should remember that SENTI Too campaign because it

means like you know, this app doesn't always win, right. Um, there is this kind of you know five chance empirically that um that they don't um. But the odds are are you know, certainly in Hillary Clinton's favor uh, certainly against someone like like Donald Trump right where you know, and maybe at the party the GOP is weaker than it once was, but they're going to fight until they're dying embers to not nominate Donald Trump. He's kind of a trifecta of things that um that they wouldn't like.

Number one, he's not very popular with independent voters. Um. People talk about how you know kind of Trump is extremely he's a new populism take of the country. Well, you know he is really awful. Favorability ratings with the electorate as a whole about negative positive right um. Number two he's now that conservative. And number three, you don't

know we're gonna get with him, right. Um. So you know there's kind of the three criteria that the party would say, boy, you know it could this could be a real disaster. And and yeah, so gun to the head today you're are you looking at Hillary Bush or Clinton Bush or is it Clinton somebody else? I mean, you know, I kind of think, um, on the GP side, it's about Rubio Walker and other Right, Really, that's fascinating. I wouldn't have thought maybe I'd diminished Walker's chances a bit.

He struggled in the polls after the debate somewhat unless he really comes back in the next debate. He seemed kind of soften, not ready for prime And that was some reputation he had, is that this guy is does find one on one but maybe his trouble standing out a little bit. At the same time, you know, um, he's leading the Iowa polls or at least as a leading non Trump candidate, depending on what poll you look at, right, and he kind of is that's a new category. Leading

non Trump can today. But you do almost have to pull Trump. It's a little bit how uh, in a weird way with um Ron Paul in two thousand twelves where Ron Paul was going to get of the vote and if the vote were divided evenly enough among other candidates, then um, then that my have been enough to win. He came pretty close, right, you know. You know, but

being the leading non Trump Candida. I mean the term front runners used um sometimes in horse racing for a horse that uh jumps out to really quickly, but it's not going to have the stamina stamina to go the distance. Right. And so you know, if you call Trump for front runner, I kind of think in that more ibronic sense, it might be more worthwhile. But you could also hold on to and then not really grow from there. There are a lot of guys that, um, you know again, Pat Robertson,

Pat Buchanan, Paul Huckabee, Santaurum. Right, Um, you know you can win Iowa in a field with seventeen candidates, with twenty doesn't mean a lot, you know. One poster recently said, Um, okay, let's just take this down to three candidates. Um, you know, Trump, Walker, and Bush. I think it was and Trump barely gained anything, whereas Walker and Rubio pick up support from the sense

Christie's and and the fi Arenas and the whatnot right. Um, And so as that field consolidates, right, then Trump will have trouble or maybe it won't consolidate, but then you're gonna have like know him with a plurality of delegates. Um, this is a scenario that journalists dream of, um, brokered convention, madness, brookered convention or at least, you know, maybe resolved before the convention. But and there's a lot of backroom dealing. Trump is not going to benefit from that because the

Party of Salesman faciates his guts. Right. So, um, you know. But the one thing that would make me more bullish about Trump's chances. You hear he's actually investing in staff in Iowa, right, is he or is he not? You never know. It's there's a lot of smoke mirrors with any candy at the stage the race. Um. And also you can invest in something and not do a good job of it. Right. But you know, um, you know again the one candidate who defied the establishment and one

was McGovern seventy two a long time ago. Very different cannate than Trump, very different. Um. But you know it wasn't just uh he got lucky. He had lots of grass roots support. He understood how to in caucuses, He understood the delegate rules. He understood you know, party conventions and how the delegates are allocated that the caucus can

change later on. Right. Um, you know, so if he if Trump has that side of the operation, really good logistics, and understands the rules and has had a lot has a lot of lawyers willing to litigate when the GOP tries to change the rules. Then then you know, maybe i'd make his odds six percent instead of two percent. But still, yeah, I think so, Because I was the last political question I wanted to ask you was what happens in the head to head between Trump and Clinton? UM,

I think Clinton wins thirty eight states or something. It's that so so Trump's core audience is his core audience. He's got those folks, but as other candidates drop out of the race, he's not gonna necessarily attract those those voters. Now he's not that popular, right, I mean, you know, again his kind of unfavorable reading ORG about Hillary because she's now like five points it's like, you know, forty to forty two or something. Now right with Trump, it's

like sixty five to thirty. He is not very popular apart from UM, apart from you know, a certain number of Republicans. Um, of course you could also run as an independent. Is that likely to happen or is that a lot of noise? I mean, ordinarily you'd say no, but I think Donald Trump doesn't like to hear no for an answer, right, And you can certainly decide. Um, you know, I'm having too much fun, right, right, And it's gonna be grateful whatever my next show is, when

it's gonna be great. Right. And he seems like he's a guy he wouldn't mind spying people. You know, It's again, it's not clearly he's really a Republican. Rights Like, Ordinarily a Republican would say, you know, at the end of the day, I owe an oath of loyalty to my party. Right, Trump might say, you know, I don't care if it's Jeff Bush or Clinton, right, or Rubio or Clinton. You know, it's all the same difference to me. Right, I'm so

all over the place, So so I don't know. You know, Um, there was a story out some time ago that last year Trump spoke to Clinton and um, yeah, he was encouraged to run or or he wasn't discharged from running or something like that, and the conspiracy nuts all uncrazy about it. Yeah, I think it's quite a conspiracy. But you you clearly you have a guy who's running as a Republican who does not care about the long term misinterest of the Republican Party at all. Right, Um, you

know he's a potential spoiler. That's very different. And and even though I think he's very unluckly to win the nomination, the chance that can make it difficult for the GEOP to nominate to candidate, difficult for them to coordinate a message. And the chance I still think is low but not negligible that he would run as independent. Right, So I think he's um mostly bad news for the GP. I was trying to make. I always want to make my own contrarian case, gendering a lot of enthusiasm where to

look reasonable as compared to him. But you know, he certainly brought a lot of excitement so race that probably wouldn't be here August. But what I say is Trump can be a problem for the GOP, even if he's very unluckly to win. Right. Quite fascinating. So let me shift gears on you and talk a little bit about you and some other aspects. You know. We we quickly went over your background on the on the radio portion. Um, prior to Pakoda, there was a quote of yours that

I really liked. You. You were working at KPMG in Chicago, and someone had once asked you what's your biggest regret in life? And you said, spending four years of my life in a job I didn't like? Yeah? Is that accurate? Is that a true? Uh? I've been true, true quote. I've been lucky enough. And so whether you work, it's been half your time working or you know, man with onum work and I we bet listeners of the show are somewhere higher on the spectrum of how much energy

they to vote to work. Um, you know it's time. You can't get back really, and even though the day to day can be a grind any job, just to have a job where you fundamentally our challenged intellectually, were you? Uh, where you enjoy yourself, where you have ownership of the work, where you like the people I work, you work with? I did like my colleagues ATKPMG. That was not the issue at all, right, But um, but a really great team I work with now, I mean that's you know,

that's really important to say the least. And then and then you quit KPMG and you start playing poker. Yeah, how did how did that work out? Um? So there was kind of a poker boom, which I would call more of a poker bubble really in the mid two thousands. So, uh so my buddy at KPMG is like, Hey, we're gonna get a game going. I'm like, I'm really competitive.

So I started practicing online at like Yahoo where you play for free, and I'm like, you can't really play for poker without without real money, right, So some online site was like deposit Bucks and you can withdraw it and it's like basically free money. Right. Of course, I got kind of hooked and started staying up all night

and playing poker. And at the time, you know, I played a little bit, and Kyle just a couple of years out of colle at that point, and you know, the quality of play was really poor and just kind of using a very basic ABC strategy. You can make a little bit of money if you started to, you know, um bluff a little bit and be a little more aggressive.

Could do even better than that. So but yeah, I kind of made my living mostly playing poker for a couple of years, so you did pretty well and otherwise I made a couple hundred thousand and then lost some of it, but enough that, you know, it was a very cool experience. But ultimately it was like, you know, it's kind of like the proverbial hundred dollar bills sitting on the ground. It kind of dried up once people realize that that's money to be made. It got arbitraged away.

That's the famous quote, if if there ever was a magic formula, it would eventually be whittled away as everybody started using the formula, at least in the stock market. That's uh, that's that's how it works. Let's talk a little bit about hunches, right, And I don't know if if Malcolm Gladwell is the best person to use as an example for that, and I think he's sort of

backtracked on on some of his early earlier statements. But when you look at the work he's done with outliers that after you've done something for ten thousand hours or some ungodly decade amount of time, you know enough of your you subconsciously recognize the statistical spread at the the probable outcomes that you could select the best just out of out of habit. How do you, um, how do you look at that sort of approach of intuition or haunches or what have you. I mean, I'm suspicious of

the ten hour hypothesis. Right. Um, you know, but one experience is pretty new for me now is that now we have a whole bunch of employees, a whole bunch, but um, you know, two dozen at five thirty eight, and I often work with our our younger writers and analysts son on the problem. And um, you do see the benefit of experience there, right, where you're developing some little model or formulate to address like a question in baseball. Right, And even though the younger writer I might work with

is like super smart, as smart as me. Um, I've just been doing this now for for ten or fifteen years, right, so I can say, you know what, you're unnecessarily complicating the problem. They're right, And that's just gonna make it hard to explain and make the model. Overfit is technical term or you know what oh here in other words, overfit meaning it's geared towards what happened previously, and you're

making it. Yeah, it's like too rigid mask right or commercially like, you know what, you're making an approximation here that's just way too clumsy, right, and you're missing the whole gist of the problem. We're trying to solve here, right, Like that intuition for kind of which method works. Um, you know that's built from experience, I think, but you know what I mean, it's you know again, intuition kind of makes it seem like you're not spending much time

thinking about the problem. It's kinda like you've invested that time before and developing some some expertise. Right. Um, So

I softened on that a little bit. Um. But the problem is that you know, um, they're very any systems like politics, for example, where kind of it's are all the Daniel Kneman system one versus system two type of thinking, where the instinctual reaction is to overreact to things, right, and to say, oh, here's a new poll that came out that shows um, Bernie Sanders ahead of Hiller Clinton

in New Hampshire. This is really dramatic, and all my friends are talking about this poll, right, and you kind of ignore the fact that there have been file other polls of New Hampshire in the past two weeks that show Clinton ahead, and also that New Hampshire is one state and pulls every other states so show Clinton ahead, right, And that this happens kind of in every election. Cycle. Um, so it's very useful to kind of, I think, slow

down and and not overreacting in the market. We call that the recency effect, where you have this long series of of of long data series and a trend supporting it, and then you'll get something within a statistical range of possibilities. But that's off trends, and that's so much of you know, we aren't trying to predict what the polls will say tomorrow, right. You know, there's a lot of people who want to

know kind of what's the mark are gonna do today? Right, We're kind of like, well, you know, here's your kind of not ten year time horizon, right, but here's your six month time horizon. Right. That we still think Hillary Clinton is very likely to be the Democratic nominee. That we're not sure who the GOP nominee will be, but we'd be short Trump stock as it were, right. Um. It also helps to quantify these things to some extent too. So in some sense it's kind of like, um, if

you ask me what's your chance of Trump winning? We don't have a model yet, we probably will at some point, right, but you know the nomination I say it's two percent. I mean, it's kind of a spit balled estimate, but it's useful to have to put at least the order of magnitude on the table, you know, a small, non zero chance, because there are journalists who will say, you know, right, an article and say, oh, of course Trump is unluckily to win, but here are all the reasons he has

momentum blah blah blah. Right, you know, and if you read that article, the sense you might get is that, you know, kind a pretty decent shot, right, you know. But if you actually kind of actually had them say we think the chances about two percent or say zero to five percent, at least the order of magnitude, right, and that would be useful, right, you know, because in some sense, if we say a lot of our calls that are like, oh, you know, no one gives Bernie

a chance, but but he could win. It's like I give a chance, you know, five percent or whatever. But the journalists or Pundy who writes a column saying, you know, Hillary is not inevitable. I mean, in a literal sense, I agree with that. It's not a percent. It's not quite close to say it's a rounding error, right, But you know, it's pretty low and to actually say is

there a real beef here or not? You know, my view is at five eight, if we think a candidate has a chance of winning, then most of our coverage should reflect the reasons why she probably will win, and occasionally we would have a piece saying that here's something

of her to be really worried about. Right. The ratio in the mainstream media is like almost the reverse, right, where it's like they're so many reasons why the percent might come through and there's very little, just a perfunctory reminder that, oh, by the way, she's leading in the national polls by thirty points and no kind has ever

done that before, right, you know, without being nominated. Right, you know, so the ratio is a little bit askew, even though I'm not sure there would really be a beef.

If I if I, you know, sat down with the calumnists that I read a sarchastic tweet about, right, you know, he or she might agree that, yeah, Bernie Sander's chances are you know, five percent or ten percent just kind of in you know, I want the thrust and the kind of tone of the articles, you're right to kind of reflect that reality, not just as a caveat in the fifth paragraph that the that's the problem with narrative and the problem with human speech, which developed around a

time when there was no written language, that narratives are more memorable, the more exciting, and the greater the horse race. What you give up in accuracy, you gain in page views. And that seems to be my gross It seems like there's some reduction that can occur. You know, we're lucky in the sense that kind of because everyone else is saying one thing that we can get a lot of attention by kind of saying no, this is probably pretty wrong,

you know. But you know, I think there's also something mathematically accurate, right, um, But I think there is also there is also a group think that sets in apart from just a page viewing everybody on the same bus here in the same nonsense and having that's and that's

somewhat literally true. So I went to literally Yeah, So I went to New Hampshire for a week or so in two thousand and twelve, right, and for the first time in my life, I went to um a presidential debate, one of the primary debates, right, And I kind of thought, you go to the debate, and you have credential and you're in the debating hall, you kind of see how things play in the room. No, it says you're heard it in this giant gymnasium with two thousand other journalists

all checking one other Twitter feeds in real time. It's like the literal definition of group think being being manufactured. UM, And so I want to emphasize that line, the literal definition of group think being manual. And so we're talking about kind of the conventional wisdom. You know, sometimes in markets you embody the convention wisdom with what the market prices, right,

but there's some separation. You know, if you look at UM, I mention there are betting markets where you can go and UM by Stock so to speaking Bernie Sanders Hiller Clinton, and they have not seen Clinton's odds changed very much, right, whereas the tenor of the news coverage has changed quite a bit. That that's fascinating. I'm gonna shift gears on you again. UM. So, who are your early mentors that you kind of come out of Chicago and you create

or at least amplify a form of statistical analysis. Apply it to baseball, Apply it to UM campaigns, and elections. Who who motivated this? I mean, you know, Bill James, kind of the guyfathers stuff and in terms of not just having the satistical child to being able to communicate it to white all agins and kind of ask essential questions about about baseball. Um, you know, I looked up to some of the guys that work with a baseball perspectives,

but I'm pretty self motivated on the whole. You found your way into this basically, Hey, here's something that's not being done right, and here's a bit of way to do Yeah. Like, boy, I think, you know, I think the coverage I watched on TV during two thousand and seven is very frustrating and so um so boy, I'm gonna have to do it myself, right, you know I kind of if no one else is gonna do it right,

you might as well step up. That's challenge me now, right where I have a bunch of great people I'm working with, right, and it's kind of like, you know, how can I figure out not to do everything well? It's a challenge learning how to delegate that stuff. Let's let's you mentioned, um, Bill James, Let's talk about books. So obviously Kamon's book very influential to thinking about thinking. What other books have really uh, either inspired you or

you've really enjoyed fiction non fiction that doesn't matter. I mean, there's a very wonky book called UH by Phil Tetlock was press new book coming out and new becoming out soon, right. It was called Expert Political Judgment explic Forecast. So that's a good kind of te temic book about about forecasting. But um, and how the experts are indistinguishable from a random person in public when it comes to making these predictions and forecasts a year out. Um, they're like barely

even better than undergraduates. I think was the most amazing Um, you know, I mean, I don't know. In terms of UH storytelling, I think Michael Lewis is a terrific storyteller. I guess my answers are quite conventional and boring. Here. Did you ever read James Glick's The Information? I've read parts of it? Yeah, dude, that is so up your alley. I will check. I you know, I love his stuff. Do you want to talk wonky? So my undergraduate was applied math and physics before I switched to Polly say

and so his book. I found him because of the book Chaos Making of a Science, and it's you know, way into the weeds on physics. But few people can take hard science and make it a narrative. That's interesting. I read that on a beach one summer and found it absolutely information just absolute fascinating. Diet Tribe on the history of of information science. Yeah, I mean, I think all these things are you know, kind of history of

nala tripe. But I know, you know, to be honest, I traveled quite a bit and I used to catch up reading on planes, and now that I have people to manage and there's Internet connections on the planes and stuff like that, reading for pleasure has has declined. Unfortunately. Um so the net. A related question, what other mathematicians or statisticians either influenced you or influenced your approach to

looking at at what you do? You mentioned Tetlock. Anybody who's really a political scientist, anybody else really leap out. I mean, there a couple of guys like Colombia whose name is and be familiar, but you know, Andrew Gelman is a guy there, Bob Arricks, and who are political scientists, um who have still at elections for a long time, and and we're helpful. But you know, um, you know, I, like I said, I'm pretty self taught for for the most part. I picture you as a lonely mathematician, light

burning in the window, and uh, working by yourself. Is that kind of how this all came about? Sort of? It's different. I mean, I live with a partner and I manage people at work, right, so I don't have as much, um much alone time as as the old days. Right. Um, you know, I'm one of those people who's kind of very much in the middle of the extroversion introversion curd.

I can go, I can go crazy on either extreme. Um, but um, but you know, yeah, some of this pursuit is pretty pretty solitary inherently, I think, right, uh, sure, you're thinking about what the status quo is doing wrong, right, and trying to come up with ways of thinking about it that makes more sense and as more probitive value,

more statistics. It seems like if you kind of start, you know, so I think about this Clinton versus Sanders thing, right, Um, you know, before the campaign began, if you kind of had told yourself, well, you know what, no matter what happens, there's gonna be a strong incentive for for the media to make it seem like the race is really competitive and it will be exceptionally clever in UM. In ways, almost all this is unconscious, by the way, Right, I'm

not saying they're deliberately putting their finger on scale. Right. Well, you're boards are doing the same thing every day. You're boring. There's so many polls, so many indicators to look at, so many ways you can Um, it's a very large constellation of information, right, so many ways you can draw

threads together that inevertly. You'll see some stories written about how quote unquote unexpectedly you know, Bernie Sanders is doing is doing well, right, and you can kind of brace yourself for that in in the abstract, UM, but it's kind of hard to do that when you're confronted with it. Another example is that almost every year, there are some exceptions. UM,

a candidate gets a boost in the polls after the convention. Right, so the bump Rubio, Bush Walker, whomever, Trump, Um, we'll have his convention, Uh, get a five point bump in the polls and it'll fade, right, And this is you know, pretty predictable. Um, it's gonna be really hard when a Democrats experiencing that not to think it's something new and different this time, and the press will event all types of reasons why you know this time is different and

it it probably won't be. You know, debates can produce little bumps too, but it's kind of like you're about to go on on a roller coaster. You can kind of, um, you can say, well, I know we're gonna go up this hill and then come down it really fast, right, and there's and I'd relate to it, but you know, it doesn't. You can still kind of suspend your disbelief, and of course, if you're an roller coaster, you want to suspend your disbelief and have a lot of fun

with it. Right. But you know, my job as someone who covers campaigns, someone who's more empirical but also critical, is to kind of stay at a distance and say, you know what, I know, you guys think that this Trump thing is new and different, and by the way, it probably won't evaporate tomorrow. Right, it might take some time, but um, the odds of a guy who basically isn't a Republican being dominated by the Frolican party is pretty darned low. So on a related note, I have to

share a funny story with you. So we just this past weekend. We we had family members in from Chicago and I remember No. Eight them saying, I know you think Hillary is going to be coronated, but this Barack guy, who we've seen in town for years, don't underestimate him. He's he's got something and he surprises Hillary. He gets the nomination and then Lader in the year. As as the election move forward, I want to say August September, October of eight. Um, my sister in law, we we

just had this conversation this weekend. I mentioned I'm gonna be seeing Nate Silver on Monday. She goes, oh, he made my oh eight so much better? Why she goes, well, I really thought there were big Obama supporters. I really

thought he was gonna lose. But every time I'm sorry, in every time they went to they said, you were reassuring that here's what we are statistically, and while it's not impossible for him to lose, you know, the electoral College makes it really hard for Mitt Romney to garner two seventy votes and and here's the likely and she said it was a despite the media drumbeat despite how how close this was. So so she credits you for making her two thousand twelve much better, and I get

that coming a lot. Yeah, that exact kind of sentiment. That because you can kind of lose yourself in the media coverage and because yeah it's weird, like I, um, in our election coverage, we kind of always wind up um. Incentive wise, We're always kind of rooting for the favorite, which is I'm not that kind of guy, right, I'm

like an underdog, unpredictable kind of guy. Um. But you know, the media will take a race where one can it kind of objectively quote unquote has a nine chance of winning and really really sell the case that it's neck and neck. Right. Um. Papers, Yeah, it's how you sell papers. And so I kind of always wind up um, you know, being the person who says, you know what, this is

not that schucking and prediction. But Donald Trump probably not going to be Republican nominee, Bernie Sanders probably will not be the Democratic nominee. And we will, believe me, one year, we'll get one of those things wrong. I mean we've had cases where um, where another context sports and whatnot, where you know, unluckily things have happened. But um, but you know, it can really throw people for for a loop.

Um but you know, the fact is also in fourteen we had from the very first time a couple shift forecast the GOPS favorites to pick up the Senate from from Democrats, and a lot of people were not remotely close to seeing that. No. No, it really depends on the presidential coattails, and it depends on this Yeah, and people you know, and you saw it quite a bit of it in in reverse. I think like Harry Reid,

you know, he's like, this guy's an idiot. He always under said, well democrats would do see, you do get it from from both from both sides a little bit, right. I'm not saying this is you know, we're not going to burn another hour here. I'm not saying the parties are exactly identical. But um but you know you would

get comments like how did Nate Silver turn into a Republican? Right, It's like my views haven't really changed that much, right, just like I think, you know, here's where the polls and the fundamentals point is that you have, you know, last year's mid terms, howquckly we forget right that, you know, with this big blowout for the GOP just last year. Right, Um, but you know, all these states are being fought for in red states, and Obama's unpopular and democratic turnout looks

like it might be bad. Right, It's not that complicated necessarily, So I would be remiss if I didn't mention unskewed polls dot com. Yeah, I found that horribly, horribly amusing. Basically, the media is biased. Nate Silver doesn't know what he's talking about. We're gonna skew the polls or we're gonna unskew the polls, and Romney's gonna you know, he's gonna run the table. How often does something that blatant obvious concerted that was really like a How often does that

sort of stuff pop up? Oh? No, I mean it'll be there'll be versions of that, probably on both sides by the time we get to next year, Right, and the more dangerous versions of versions where like, you know that guy, I've never met him personally, Maybe he's a nice guy to have a beer with, right, he seemed a little bit kind of screwy in the head, right, just just enumeroate, But it will not be that guy.

It'll be someone who, um has sort of a sophomore acknowledge of statistics, right, who can actually kind of make superficially persuasive arguments for why, um, why my hypothesis is true, and you absolutely will see that, and I know there's a market for you know, one of people I'm realized about five thirty eight is there are now six or seven websites that do a forecast similar to rors, right, and we're always kind of somewhere in the middle of

the consensus, which I think is usually a good place to be consensus of consensus analysis. Yeah right, Um, but there will be someone let's say we have Rubia with a seventy chance to win the GP nomination, and that's where the average is, right roughly, I can guarantee you le'll be someone who says, no, it's s Hillary, Um, no, it's Rubio right, right, Um, so they look like a genius even though statistically their their work is so that

be someone who looks like yeah, right. And if there's someone who um, you know, it's the same problem with with Wall Street analysts, I guess right, you know, on average, the person someone who makes an outlandish forecast will be in first place, even if their meetian outcome is worse than the person who makes like a more kind of lower case could see conservative going back to Tetlock and others, when you look at punditry and people go on TV.

Someone who goes on television and says, you know, the future is inherently unknowable. I don't. I can't tell you whether dow will be in one year and someone else says it will be at nineteen thousand seven fifty. Not only is that person less likely to be accurate, but they're more likely to be believed and and so that's the enhancement I did. I did a program for the BBC on the UK election and we had some guys we hire really smart guys and like everyone, they way underestimated,

uh how well Tories would do. Right, But this BBC program was there for a week and my personal take was like, you know what, this is a really weird election. There are four parties or five parties that are assailiant and and um, there's not a lot of agreement in the polls, and like so I was fighting so hard to avoid how we go to commit to anything, And of course their whole stick was brilliant genius stistition comes from the United States, and was your you know, like

tells us exactly what's gonna happen the elections. There was there was tension there, I think, right, it was like a week of trying not to answer this question. Um, well, the data is uncertain. Using a methodology, that data is certain. But in the end we published a model on five thirty eight and like everyone else, it blew it. So but you know, um, but yeah, it's a tricky thing to do. And that's a much shorter election, and that's a a very different process than in the United States.

It's a much shorter election. It's only it's only six weeks, but when you add multiple parties to the race, it gets much more complex, like really fast. Um. But you know, we haven't talked about this. I know, we don't have too much more time. But the fact that, you know, the polls aren't as reliable as they used to be, at least in some non US context. So more cell phones, less reliable polls. Yeah, you know, Israel and in Greece and whatnot. We have a Canadian election. We'll see how

the polls do there. They've had trouble in some Canadian parliamentary elections and whatnot, or provincial elections. Um. But you know, UM, polling is not fool proof either, and that's why it's like, you know, that's why I say we got lucky in part in two thousand and eight. The polls in two thousand twelve were okay but not great. Actually, Obama beat his polls by an average of three points? Is that unusual for the victors who have won by that? It's

about It's about average. You know, a three point miss is about average. The thing is, if you'd had that three point miss any other direction, then Romney would have won the election, or at least made it really really close. All Right, I know we only have you for a few more minutes, so let me get to my last two questions. Um. This is the millennial question that that

I we talked about earlier. So you've carved out your own kind of unique career by taking a couple of subjects you really liked and applying it in a in a unique and useful way. What would you what sort of advice would you give to millennials who are just coming out of school now or whatever we want to call this generation? What sort of career advice would you give the recent graduating class. I mean, you know, learn how to code if you want to become a journalist. Right,

that's pretty more more important. Um. But you know, I mean, uh, it's a combination I think of of working really hard, but um, but not tolerating yourself being bored at work. Right, I think that's pretty that's pretty important. Um, it makes a lot of sense. Um, I mean, you know, uh, learning to critique the conventional wisdom, including the wisdom that your friends might have. Right. Um, did you get a lot of pushback on on what you were doing from

from friends and colleagues. No, I mean they always know me as someone who did different things, so I don't think there was ever that much pushed back from it, exactly. But I do worry that we kind of enter a universe now where, um, where it's kind of so easy for an opinion to form on social media or whatnot. There's so much kind of communication and sharing that I think that can be a little bit dangerous, right. You know,

I don't think I'm thirty seven. I don't think any thirty seven year old should say I have the whole world figure out, you know, certainly not any twenty seven year old or seventeen year old should say, you know, um, you know, my beliefs about the world are just the way things are, you know. I think I think there's there's that tendency a little bit um, not just with millennials, but with everyone now kind of the age of of social media. You can kind of feel for rocket bear

term kind of smug about your view in the world. Um, and that can be a dangerous thing. Enough people, there's enough selective perception and enough things around that reinforces your views, especially with the Balkanization of the Internet that yeah, you could you could find applause for just about any opinion on the spectrum. Yeah, it's really easy to you know, it's part of part of what Trump's doing is you know,

I sold Julier. He's not actually that popular, but the twenty five or the country that really likes him, he can kind of wallow in their world, right, I Mean, Trump is I'm kind of going on attention here. He is incredibly entertaining for sure to write like so one thing that I think does separate him out from um, from say, the harmon kines the Michelle Bachman's of four years ago. Was the guy has real talent for or something? Is it politics exactly? Your showmanship or demagoguery or somewhere

in between those three things. I'm not. I'm not quite sure, right, but you know there's a charisma there. He's hosted or at least he's been significant in a television show very very fast. He has a sense for for theater. He can poke the eye. I mean, you know, um, I'm not that's a he's got a natural gift for that. I'm not saying I like Trump. You know, I wouldn't vote for him, but I love the fact that he had his helicopter around the Iowa State Fair because it's

all these rituals that are silly in some ways. Right, He's gonna be like, I want to be the evil villain and just kind of circle my helicopter around the state fair grounds and like driver one's attention, right, And like, um, was he giving kids, healthy kids helicopter riots and telling them he's he's Batman? Literally? Right? Um? That's great. So I like people are willing to defy connection a little little bit, right, and he certainly he certainly has done that.

But you know the reason why we kind of get end up as being perceived a very anti trumpet just because you know, we think the media takes this fascinating story and trumps up so to speak, the chance will actually be nominated by the GOP right and also great narrative, but also the fact that it is August, right, the polls you see in August aren't measuring anything real in a sense, people are hypothetically thinking about a vote they're going to make in six months when half the candidates

will have dropped out, right, Um, when they'll have four times as much information. The idea of all you have a national primary, you don't actually vote kind of one state at a time, and so you know, um, some people kind of point, um, the Trump's polls as kind of self evident evidence that he's succeeding. Then I'm a little bit suspicious of that. But he certainly is ah entertaining in a way that I think raises some questions about the way campaigns are are covered. Right well, there

certainly raises questions about the way they're run. But the coverage is part and parcel of that. The coverage is not not necessarily doing democracy any uh any favors. I think I mean, you know, kind of I guess the two grand thesis about Trump are that, you know, kind of Trump reveals what's wrong with the GOP, or that he reveals what's wrong with the media, right and kind of being Yeah, there some combination thereof I mean kind of you know, kind of an amateur media critic, I

kind of lean towards that explanation more. But there's you know, there's something he's tapping into. I think I would resist the implication that he's all that popular or that people are responding, um, all that literally to the substance of his policy proposals. I think there responding to the affect to the you know, kind of he to the nerve. Yeah,

I think that's I think. And certainly if you look beyond the poll's terms of is he popular or not, people like him or not, there's no doubt that there is enormous interest in Donald Trump, right. Um. You know, I think we did read some metrics from Google, like six of the news coverage of the GP campaign has been about Donald Trump. That's amazing, But the Google searches for the campaign have been about him. So the public is even more obsessed with So you know, you know

that won't translate into support. Yeah, I want you to be president, but there certainly is a fixation with him that is not purely a media creation. So last question, because I know you have other places to be and I've kept you here for a long time. Um, what do you know about politics, campaign, sports statistics today that you wish you knew fifteen years ago when you were starting and you could pick any of those fields to

work with? Oh gosh, um, I mean there are technical answers to this, right, kind of wish I wish I had when they took me a while to understand it's like really fundamental, but I think it takes other people a while too. Is that um, is that a Cisco model is built on past data and you're making a big assumption to therefore say the model is predictive? Right? Sometimes that's true. In baseball, the conditions are so stable that if you build a modle that explains the past, well,

you're kind of, by definition usually predicting the future. Well, right, you can extrapolate better with that stable set up than you can perhaps with politics. But in politics, you know, not always true, and you get into areas like finance, right, and there used to be a correlation between which conference on the Super Bowl and how the stock market did. Right, Yeah,

but that was always so I understand. My job has always been to point out that's a correlation without a causation, and whether it's crazy things like the Hindenburg almen or the death Cross, my job is to go back and look at these things historically and say, well, here's the net result of it, and this doesn't work this, this has a strong correlation, but it's a random accident. And here's what we get out of this. Um. But the leap of you are making a leap of inference. I

was gonna say leap of faith. That's not really about faith, right, but a leap of inference where you say, okay, I fit a model the past data, therefore I'm predicting things like no, you fit a model of the past data. And then there are a lot of factors you have to consider in terms of how well that model will predict going forward. Right. Um, So you know, whenever I see you know, I think a lot of academics make

this mistake, just like anyone else does. Right, But they'll say, you know, I've developed a model too to predict campaigns right and perfectly predicts every presidential election. It's like, no, you didn't predict anything right. You know, if you publish a mouth for the first time in and you say, perfectly predicted all fifty two presidential elections, however many there are in the past, Like, no, your record is zero

for zero forfeited, what not? Not fifty two? That that was might be with Anthony Robbins model that was heavy on gold, have on bonds after a ten year gold bull market and a thirty year bond will market, both of which were unlikely to continue going forward. It was that overfitting predicting the past as opposed to looking forward. Um. You know, in the whole question of fascinated by is kind of you have this um tension between the fact that on the one hand, um, you know, the consensus

is usually better than an individual forecast. It's very hard to beat the market when there's a robust market, you know. On the other hand, you know, understanding that there's group think and that everyone can kind of be delusional together, and that pension I'm kind of I'm kind of fascinated by as as we all are. Nate, thank you so much for being so generous with your time. Definitely this This has been really just an absolutely fascinating conversation. We've

been speaking with Nate Silver of five thirty eight. If you enjoyed this conversation, look up an Inch or Down an Inch on iTunes and you can see all of the rest of our conversations. If people want to find your work, you're at Nate Silver five thirty eight and at spelled out five eight dot com. I want to thank Mike Batnick, who did a yeoman's job as our head of research, and Charlie Bohmer as as my producer. Um, you've been listening to Masters in Business on Bloomberg Radio