Risk-takers, US election & the power of luck in politics – with Nate Silver

hello and welcome to the Americano show I am delighted to be joined here at the spectators offices by Nate silver who I probably don't need to introduce but I will say that he is the world's most famous election prognosticator uh and he has written a book a new book uh called on the edge the art of risking everything uh it's a fascinating read Nate uh and it's a pretty eclectic book I think to use an overused word um first of all I'd like to say during year when everybody wants to talk to you about an American election I know your your book uh the signal and the noise came out in 2012 which was another election year but that was probably more related to what you do or what you're most known for um whereas this feels like a bit of a gamble because you're talking about gambling and you're talking about risk yeah look if I never had to cover another election I would still have a happy life I'd be perfectly happy with that I don't really like um the people in American politics very much right I I think sometimes I'm too much in the spotlight for this stuff um trying to provide like a more arms length kind of rational voice to have a little bit of of fun with it to help explain explain probability and things like that to people but like um like you know I am not like a politics person I got into politics I was playing online poker and the US Congress passed a law to basically ban online poker or ban Payment Processing to online poker in 2006 and that that compelled me to follow politics more I wanted the people who passed that law to be voted out of office which they were by the way um you know with more time on my hands with poker no longer a viable source of income um it's 2008 I'm living in Chicago I go on University of Chicago Barack Obama a very different type of politician than the George Bush John krey era is running for president it's a big kind of Moneyball data driven era in the US and so it kind of yeah it would seem like it was all plotted out very carefully but it was more my my frustration with political coverage being very kind of momentum and Vibes driven and not very quantitative at all and and and to my surprise it kind of was the right time to get into American politics that like 2008 onward has been a fascinating sometimes exasperating era for you know for elections in the US and and it's been fun to have a a you know a seat on the train for it but the idea for this book was to Pivot out of politics um and instead of gotten gotten swept back in I suppose you mention the book that people have get very angry with you about your political prognostications uh less so with sports is this book in some ways an attempt to explain to people that it's the same thing for sure and when we say the same thing that's true in some ways and not in others right like obviously politics are are much higher Stakes although the fact that Sports is so popular around the world is something I think goes taken for granted sometimes um but they are like both strategic games in the sense of Game Theory where you have two sides trying to optimize their outcome against a smart and competitive opponent um you know they are both felds that lend themselves relatively well to data driven analysis in the sense that you have a lot of structure election data in the US and the UK for that matter is you know it's good data we um understand how elections work we understand the rules of the system and that makes for you know more um translatable problems as far as forecasting or or prediction go um but yeah look in politics especially in the US you hear over and over this is the most important election of your lifetime that it's very built on grievance I think people have trouble understanding the RO that like things like luck can play in politics where if Donald Trump turns his head in a different direction then Donald Trump may have had more than a bullet graz's ear right now um in 2000 the outcome came down to essentially random ballot design choices in the State of Florida and weather patterns and things like that between Al Gore and George Bush um people seem reluctant to embrace the idea that maybe their election wins aren't just by divine right um but are by twists of faith that are hard to foresee in advance well that's yeah I mean nobody could predict the future is is the obvious point but uh but I think what I'd like to start talking to you about properly is is this division you have between uh the river and the village in the book and this is in the bit about gambling um for those who don't know please explain to us what is the river who are rians what is the village who are The Villages so I'll start kind of in the in the US metaphor because the book is partly about America and what makes America weird um uh so the village is kind of the East Coast establishment which at the moment is very politically Progressive so the New York Times and Harvard University you know when a Democrats in charge the White House and the nonprofits and the think tanks around that Academia certainly um so the village is all about kind of the collective good and a kind of progressive Democratic framing of what that means um it's pretty risk averse people in the village are afraid of losing their social status of being of being canceled and losing their influence um the river you can think of is basically Silicon Valley Las Vegas Miami it's the world of calculated risk-taking it's people who are extremely individualistic who are sometimes contrarian to the point of being difficult um but who combin being really analytical with being really really competitive um so where the UK is I mean I guess the UK in some ways to extend the metaphor is like east of the US East Coast right you come over here and I I went to school in London for a year um but you see subtle things where there's more paternalism or more risk aversion in the in the UK right you go to the tube and there are the guards on the subway so you don't trip and fall into the into the tube which I suppose makes sense right um you know you go to the store and and you're trying to buy beer and after 11: p.m. the beer is boarded up and things like that you go to the Uber and there's a blinking light if you if your seat Bel isn't on so just like a little bit more handholding here and but less but less entrepreneurship I mean the number of um the number of unicorn companies in one region of the US silon Valley is like 5x more than in all of the UK um and you still have capital and Venture Capital fleeing to to the Americas and obviously that's to do with the the structure of American society capitalism and so on but also do you think it's genetic because Americans you know a lot of you uh came here because you were exploring a new world you were you they were the Ping people willing to take a risk that probably a lot of Europe went yeah I mean look um Las Vegas was founded kind as an offshoot of like the gold brush basically right where you had people go to California first and you it's spin off into Nevada to M silver um so yeah look people have always selected in the US to um to move there because they wanted to find kind of Fortune potentially and that Pioneer spirit I think is is still a big part of the story and Silicon Valley attracts like Talent from all over the world um at the same time I also think that there are more warning answers like that in a constitutional system where the US Constitution is hard to change um there are more veto points um so we tend to have by default kind of fewer regulations we also have a conservative Supreme Court that tends not to like tends not to like liberal Progressive laws for example and so um so yeah I mean you know things like product safety laws things like free speech I'm a big Free Speech fan so I'm good with that part right um but yeah you see a Divergence where um you know I was a college student here in 2000 I guess and back then the US and the UK had roughly an equal GDP per capita within a couple thousand pounds um now the US is about $76,000 per year per capita and the UK is 46,000 so you've had a huge Divergence on the flip side um life expectancies in the UK are three or four years longer than in the US despite the country seeming like sister countries in a lot of different ways so at some point um Europe maybe even more so than the UK but also the UK I think kind of took an offramp away from maybe growth toward sustainability or quality of life and the US is still on the accelerator toward toward growth growth growth well and as you mentioned about gambling the the US has a an odd relationship with gambling because online gambling here has been uh pretty legal for a long time now uh it's just it's coming back into legality now I I'm not quite sure where we are at the latest but uh in in the US gambling is very much done at the state level and so um so you've had uh sports betting is now present in I think about half the states in the US um you know online poker is only five states or so yeah the UK is it's a weird role reversal right I mean um you know the UK is I think more rational in some ways about gambling yeah you go to like there's like a little casino on on the main road or something right it's like not really that big a deal oras in the US it's like we can be very puritanical on the one hand but then we have the biggest you know gambling site in the world Maybe tied with maau in Las Vegas it's like some kind of type of shrine to American capitalism or consumerism and Vegas is bringing in record revenues and so we we're a little schizophrenic about gambling in the US for sure and I found your book to be a good guide to how to be a risk taker um because you yourself a rivian you identify as a rivian yeah I would say yeah well I'm probably a little bit of both a little bit of both but you have uh gambled you've been a poker player you have bet on Sports and even a person who's very well off uh very large sums of money um did do would you say you're low in neuroticism which is kind of a key thing to be a successful rivian I think I'm pretty low n i mean so the people it's people who have high openness to experience um but low neuroticism right I I tend not to dwell on negative emotionality and these things are kind of hard to find because typically if you think of um in the United States Democrats or liberals or progressives whatever you want to call them they're the ones are more Multicultural and more open to different sexualities and genders and things like that and more open to immigration more open in general but they tend to be pretty neurotic they tend to you know have a lot of Hang-Ups about a lot of different things um have lower mental health at least self-reported um whereas conservatives are just the opposite so if you can kind of combine like the openness to experience with the lack of neuroticism like that's a skill set that I think is you know helps people to thrive in in gambling type tasks and does it help help them to understand value because appreciating value or expected value is a key part of being able to take risk success I mean I think in both senses of the term value I mean yeah obviously you know what you're trying to calculate in some level is will I on average come out ahead if I make this this bet or take this risk expected value um I also think though people in the river are often quite philosophical um you know someone like Peter teal is not for everybody but he's certainly a very literate guy who knows kind of the you know his history of European Enlighten Enlightenment and Beyond philosophy and postmodern philosophy and things like that um the people who are pretty widely read for the most part um and who are trying to kind of understand what like the governing rules of the system are and they're less kind of they're Less ad hoc I mean one critique you can make of the village is that everything is so political it's all about you know what's good for partisan purposes then you kind of rationalize later on why it fits into your Paradigm um and the rians at their best which they often are not are I think more long-term and strategic focused and do you think with the tech Revolution these types of ravarian like Peter teal or Elon Musk or Sam ultman uh Sam Bank Freedman you talked about Sam bankman freed I should say um they are the sort of people who are pulling uh not not by themselves but their success is what's causing a lot of problems in America because because of the tech Revolution these people have had such AST astronomical success so quickly that the river is sort of getting further away from the village I think so you know the top VC firms in Silicon Valley probably make returns of about 20% per year and because they keep reinvesting those returns then that grows and grows and grows um you know if you invest in the S&P 500 then you make seven or eight% per year on average So 20% compounding means that you have an exponential increase in in wealth and power and even if you're kind of like a somewhat unrepentant University of Chicago educated capitalist like like I am I suppose then then you have to worry when kind of that much power is concentrated in so few hands where they become more powerful than than governments in a lot of ways I mean you kind of have evilon musk kind of functioning as a as a quasi governmental entity in certain ways now or like Brit Britain or excuse me Brazil is Banning Twitter or attempting to which I you know I don't approve of but like but it's almost like he's like a state power it's like embargoing a country in in in a war or something like that um so I think we have to think carefully about that especially with ai ai is not the traditional startup where you have a garage somewhere in Los Altos California and Steve Jobs makes a new computer or something right this is a case where um where the already powerful companies that have enough computing power and capital to train models for six months at a time at a very high density of of compute um are are coming out ahead it's the microsofts and the Googles that are are gaining as a result of this and is that why uh the village is necess so necessary because if AI is as as you mentioned in the book The nuclear bomb of the the technological Revolution uh we do need institutions of government that tend to be more Village than River to to moderate it to control it and to actually regulate it yeah look Sam Alman who's a smart guy he's not Sam bman freed or something like that right but he'll say that like actually oh this technology could um potentially destroy civilization if it's misaligned in the wrong way if we develop things that are smarter than people and very capable of than people and they don't like people very much then um then that could be not very good for the human race right and he'll say that I guess to his credit for being honest but like but I'm not sure that like any private Enterprise should have the power to like um to you know press a button and run some nonzero risk of like destroying civilization with the next model training run and again I'm someone who most of the time 95% of the time is like yeah I think these regulations probably do you know are not doing not good necessarily but this is a case where I think um where we should have a broad consensus that like that you know having smarter regulation is helpful but smart is also the key um you know it's a little bit of a kind of prisoners dilemma where um you know if we overly regulate AI then China might develop it or some other country might develop it um you know or you might have people who develop it in the gray Market um you know you have Elon Musk for example with his grock engine on Twitter is is you know making pictures of naughty stuff that like um chat GPT or or whatever won't do but like so you you know there are to use again the nuclear metaphor there are some potential arms race type Dynamics and the uh relationship that seems to be developing between a lot of not a lot but a few very important Tech entrepreneurs and Donald Trump um suggests that there is a a rejection of the democratic Village uh now and you are seeing almost a hatred of it in a lot of these Tech circles does that tie into what you're writing about in the book yeah look I think if you took a survey of everybody in the river you would still get more Democrats and Republicans or at least more Harris voters and Trump voters um because like they tend to be pretty educated people and in the US we had this big educational Vibe where the educated classes tend to vote Democratic in the US um but absolutely you've seen at the tip of the spear people like Elon Musk Peter teal um um you know Bill Amman as a hedge fund manager against the presence of MIT and Stanford or not Stanford MIT in Harvard and Penn um you see more open conflict and I think it's partly Personality Clash they see uh they see the village as being very collectivist and RIS givers and they're very risk on and individualist um it's partly kind of narrow economic self-interest where um you know if you read the reporting on Elon Musk before some of the culture War stuff he was just mad that the Biden admin Administration did not invite Tesla to various EV initiatives because they were not unionized um and he was mad about Lena Khan regulation at the FEC and he and they're mad about you know high taxation in California and other places and the kind of demonization in the liberal media of of billionaires you know I don't know why they're so worri they're billionaires you can it's okay to be not be sympathetic if you're a billionaire um and then I think they kind of like find the culture War stuff later on it's like I think I think if you are someone who um you you know you're you're kind of like oh I have to be a good Progressive Democrat right and then you kind of start you know you start getting pilled a little bit and you feel kind of naughty right now you're allowed to like say all those things I think the kind of covid era or if you want to call it the covid kind of wokeness era in the US I think a lot of you know particularly middle-aged and older men I'm probably susceptible to this myself right um felt like you weren't allowed to say certain things for a few years and kind of once things change there's a regime change then you may overcompensate the direction for a period of time I suppose someone like the comedian and TV presenter Bill Mah would be a good example of somebody who has sort of become increasingly closer to the Trump worldview I think in some ways but I think there are also nuances and and subtleties here right I mean the thing is like if I think about like bluffing or strategic communication um sometimes people say well I'm a little bit more moderate on this issue you know trans rights or Gaza or you know think of any controversial issue that you can right and a way to signal that like um that like oh actually I'm conservative I'm just kind of like saying this as a as a dog whistle or a hand but like you know I'm somebody who un like you know a lot of these K you know if you ask me what's your sense on on you know the war in Gaza right I'll say like I don't know I feel really sympathetic for both the Israeli and the Palestinian people and I I'm I'm kind of torn right it's like not like a dog whistle anything else just like actually how I happened to feel and I think you know there are people who are true moderates or or uncertain or just say I don't know it's kind of above my pay grade on some issues um and they're the ones that you often like don't hear from very much and do they make predicting elections are they the hardest thing to predict because they tend to be perhaps a bit more contrarian yeah a little I mean you wor less tribal you worry about the voters that you don't hear from um you know on the one hand obviously any party would rather have more enthusiastic voters than less on the other hand sometimes that enthusiasm translates into being more willing to take a poll um so you get sampling error that um that you reach the enthusiastic people you don't reach the unenthusiastic people but the unenthusiastic people still still wind up voting potentially does a lot of the book is about Bitcoin um are you a believer in Bitcoin um I think the blockchain is an interesting technology it's one of the more interesting Technologies of of the decade I suppose I I don't think right now the applications of it are are tremendously World altering necessarily I mean you know right now it serves various functions I mean like you know the technology of nfts is a pretty cool technology to provide digital ownership of of digital assets and proof of um the chain of custody and things like that it's like you know and but like most Technologies it takes like 50 years to figure out kind of what the long-term applications are look I think Bitcoin is partly partly an ideological movement um the idea that um after the financial crisis that we are worried about governments printing money and we're worried about fiat currency and to fully realize it' be quite a radical idea right to have a financial system that's decentralized and based on um Trust on the blockchain instead of by government Fiat I mean that's you know that's a very radical idea if that were able to come to fruition right now it's mostly used as a form of of gambling I'd say though it is gambling isn't it it's it's uh I mean the idea of its value is not properly understood even perhaps by the people who most believe in it to some extent it's trying to predict social behavior among kind of people in your in your social class right it's trying to figure out what's the timing of like surges and busts um you know the game theory Dynamics are actually are actually quite interesting where even I mean even you set aside like Bitcoin for now something like the bubbles that we see in like GameStop or AMC stock for example where it's like coordinate coordinated efforts to like undermine all these hedge funds and and you know I'm not sure if you even say manipulate prices but like to squeeze short sellers it's like quite fascinating dynamics that are a product of like the the internet age and meme culture and the way that you know these are mostly younger people and younger men in particular the way they tend to like scheme and bind together so that part that part of it I think is like quite interesting and it thrives on distrust all those things thrive on distrust with with the mem stocks it was def distrust of the financial system with Bitcoin it's this idea that central banks have too much power is that the prisoners dilemma at work well the prisoner dilemma occurs when there's when there's a lack of trust um so ironically you know if you have people if you have different Traders there are relatively few big wigs for example in a in a stock like AMC or or GameStop um you know if they can coordinate and say we're not going to sell until next week um then you can actually squeeze get a short squeeze you can actually have that be profitable um so that requires like local trusts and Global distrust I think um and maybe we kind of trust our local communities now that we self- select a little bit more into in online spaces you can kind of pick and choose who your friends are and you may have a tremendous amount of like loyalty to them but um but distrust of kind of The Sovereign right distrust of like you know all the institutions in the US I suppose in the UK I don't know the data as well but you know a government and Academia and media and the church everything but the um but the military we've seen decline in trust in the US well I think we we tend to be a bit up you know they say politics is Downstream from culture or whatever the phrase is and I think in Britain we just tend to be a bit Downstream on both from America perhaps I mean um you know look it's still a culture that is compared to the US relatively less partisan the fact that you see these big swings you know to and from labor and lib Dems will Surge and recline and Nel for I mean you know I think Americans are or maybe I wouldn't speak for other Americans you know you know I'm somewhat envious of having a system that seems to have healthy multi-party competition as someone who likes risk or who's interested in risk and doesn't mind taking quite big risks uh presumably in your latest uh writings about this suggest this you might be reluctant to take risks on predicting this election because it's so unusual in so many ways well look we have a a model that we publish every day um and I you know I kind of made a calculated risk that like um it's worth doing this as a way to build an audience for the newsletter that I'm writing silver bulletin um you know and so far the reception's been really good I mean um you know the subscription based model if you have content that's differentiated and specialized is is a really good business it's much better than the ad supporting B the economics of of the news industry in the US where Remnant ads don't pay very much so like I'm happy I did it I understand that if we by election day have Trump as an 820 favorite and Harris wins or vice versa um it won't be great for the business but like that's that's a risk I'm taking going in I can't I can't control that all I can do is make the best best bet so to speak that I can and set the best odds that I can and there's been a minor controversy uh over 538 the company you you set up and you sold to Disney um and it seems as though they were using a model that they needed to update can you explain what happened so 538 um was the brand I founded in 2008 um it was bought by Disney um in 2013 first with ESPN the sports company and then ABC News 2018 and um about a year and change ago a year and a half ago they laid most of the staff off and then my cont I wasn't technically laid off but my contract was going to expire anyway and we just kind of like let it expire um they hired a new person but I sold them the name 538 they I kept the models that I designed they were just licensed to Disney I know it's like more def than people want but basically they hired a new guy to build a new model and he designed a model that was buggy and broken they can try to be euphemistic about it and say oh there were just assumptions that were wrong no this model was broken and should not have been published um and they were kg about replacing it I mean apparently the staff at 538 I still dearly love like the people that not the new guy but the people who were there originally um they're really hardworking and and good journalists um but yeah and they and they kind of changed the model when Biden replaced terrorists but they took to month before they published it and then tried to um tried to kind of up fisc the fact that it had been radically transformed and radically changed and so this is kind of against the values of of data journalism as I used to call it or Journal ISM in general these models are hard to build but I imagine they're very hard to build but I also Imagine almost impossible to change from Biden to Harris how do you obviously I realize it's very complicated but how do you change that because it's such a substantial thing and yet it's not a complete rupture it's actually pretty easy right I you basically change one cell in a spreadsheet and say you know candidate number 42 which is KL Harris instead of candidate number 26 which is which is Joe Biden and then it kind of feeds in all the all the Biden or the Harris Trump polling data instead of Biden Trump now we did wait for about a week until she became the de facto not the official yet but the deao Democratic nominee so Biden dropped out and I forget which day it was right a day later she secured de facto the nomination and so a week after that we turned back on the model um maybe a bit prematurely she was still rising in the polls at that point um but look I mean the challenge is more that we're almost having like we're almost having like a snap election you know you guys are used to that we're not used to like a snap election in the US and so you can argue that the assumptions that are embedded in the model about like the two-year-long election process that we have don't apply as much and there's probably more like intrinsic uncertainty um and you have other crazy stuff you have this assassination attempt against the former president right you have RFK Jr enterry and exiting the race um you have this catastrophic debate for Joe Biden so it's been an incredibly strange set of news Cycles the model is doing the best it can I mean ironically it kind of defaults towards saying basically 50/50 um you know after the debate which is in a week um maybe the the first maybe the only debate that's where you might actually have more confidence in saying oh we actually um can lean toward One Direction or another which direction I wouldn't want to guess right now and I mean usually the pompous thing to say about uh presidential debates is that there may be a crucial moment but if there isn't it's not going to make much difference but I think because of the circumstances of this election is it fair to say this election this debate will be much more decisive yeah look you're compressing two years of campaign into five months basically so if you want to look at it heris maybe everything is like four times more important um but also I think Harris um you know has had kind of a a Glide path in August where she I think did a very good job winning the nomination I think gave a very good speech at the convention um but had for the most part like pretty friendly media coverage now you see more scrutiny about her relative lack of doing like candid interviews in the press for example in the US um but none of that matters except as much as the debates right um where I mean that first debate people anyone who says the debates don't matter like okay you just had one of the most pivotal moments in American political history happen was it just you know eight weeks ago or six weeks ago or whatever um and the fact that like you know um probably both Harris and Trump think they're going to win that debate potentially um I think they both can't kind of live down the expectations entirely um um and well no I mean it's it's going to be it might be the most important remaining day of the election yeah another thing that people often say don't really matter that much is vice presidential nominees um again in this election would you say they matter more I think they matter more because people are very reminded of the fact that um Joe Biden just literally was replaced by his vice president in the middle of the campaign and um if Donald Trump looks the wrong way or the guy you know is able to get another shot you know know what happens in and in this year um so there's heightened scrutiny as I think there should be I mean if you're if you become vice president or even named on a ticket your odds of becoming president at some point even though the VP job itself is kind of ceremonial um go way way way way up right um so people should pay attention um look I think um JD Vance was probably not a very smart pick I mean JD Vance was a pick that Trump made at a time after he had um just had the assassination attempt right and it was still Joe Biden and he was really struggling with like Nancy Pelosi and other Democrats and you know by their reporting they kind of thought they had the race in a bag so they're like hey why not let's pick Jade Vance he'll extend our kind of Maga timeline he'll kind of troll the Democrats a little bit he's a smart guy you know the kind of Silicon Valley ravarian will will like him as a fellow traveler so to speak um and let themselves get very flat footed and he was branded as weird by democrats I think he um I don't know he's only been a senator for two years I think Ohio's not really a swing State anymore I think not the most natural political Talent um so I think that was probably a fairly bad pick I mean on the Democratic side um Tim Walls was popular at first I think maybe you're having a little bit of a sell by date you see Harris lagging not behind but you see Pennsylvania stubbornly remaining you know a one-point race um and to have had the governor Pennsylvania who's very popular there Josh Shapiro on the ticket if you had an extra two points in Pennsylvania right now if you're Comm Harris then you could breathe a lot easier in the Electoral College would it be simplistic to say Waltz is a village pick Vance is a Rian pick I think like wal was the consensus pick of the democratic party for sure right Nancy Pelosi who has political instincts that are pretty good most of the time you know wanted Tim Walls John fedman wanted Tim Walls Bernie Sanders wanted him um you know I think Shapiro is seen as a guy who has like a little bit more ego it's it's always been hard for presidential candidates to pick p s that that overshadow them right um you know Obama although Obama wasn't at too much risk of being overshadowed I suppose but they tend to be pretty bland I think Sarah Palin would be an OB Sarah Palin okay would be the exception right possibly yeah but that didn't go very well for John McCain exactly so yeah she's probably doesn't help the rivian case for highrisk VP picks uh and given that we know that luck is an important factor in in these things and people talk about October surprises uh in deciding elections how often would you say October surprises have decided elections and that presumably is all down to luck yeah if you have some you know some oil tanker boycott in the Gulf War or things like that right even things like what the weather forecast is an election day in different parts of Pennsylvania can potentially affect things I mean in an ordinary campaign then I'd be saying oh things actually get locked in a little bit earlier you know so for example part of it's because of early voting voting is actually starting today in in North Carolina for example so there are votes being cast I don't know quite what time it is in the US I'm on jet lag still but like there probably there have at this moment been votes cast now in the election um but again with this kind of Quasi snap election compressed time frame then obviously October is potentially magnified Nate Silva thank you very much for coming on to the americ podcast I do hope we'll get you on again of course thank you so much