Games of Risk & Games of Confidence
My conversation with Danny Crichton
“It’s 2024 and Hurricane Helen has devastated Hampton Roads, Virginia. Local naval facilities are paralyzed and crucial U.S. naval assets like the John F. Kennedy aircraft carrier (CVN-79) are severely damaged.
The hurricane and residual fallout from Covid-19 as well as labor unrest and economic concerns further complicate recovery efforts and future naval readiness.
Private industry titans, government and defense representatives, union leaders and local media are all at the scene.
Everyone wants America to succeed, but each person has their own motivations and personal objectives. Can they collectively find a path forward that helps their interests and the nation’s?”
Thus begins the first publicly available Riskgaming scenario designed by my guest today: Danny Crichton.
Danny Crichton is a man of many talents. He’s got a background in computer science, has worked in the worlds of foreign policy, was a managing editor at TechCrunch, and now serves as Head of Editorial at Lux Capital.
As Lux’s de-facto games master, Danny also devises their Riskgames: strategic simulations that immerse players in complex scenarios reflecting real-world challenges and dynamics. These games – whose players include senators, major generals, congressmen and, think-tank CEOs – include scenarios like ‘Hamptons at the Cross-Roads’ (that deals with climate change and maritime security) and ‘Powering Up’ (that deals with China’s global EV dominance).
Danny and I discuss the origins of Riskgaming and the lessons he’s learned in high-stakes games with tech founders and government officials. Plus, we riff on our shared Minnesotan roots, and discuss ways to combat the uncertain fog of war in our careers.
As always, if you like what you hear/read, please leave a comment or drop us a review on your provider of choice. In the meantime, here’s a sizzle reel to get things started:
Episode Cheat-Sheet
Games as a richer learning medium: When people want to get acquainted with a thorny problem (like geopolitics, international trade, capital flows) they often resort to textbooks and long-form 20,000+ word articles. If you’re a political staffer or aide to a senator you refer to a policy memo (if you’re the senator — a single page summary will have to suffice). However, these mechanisms of learning are poor, and do not situate you among the nuances of the problem. Games, however, are dynamic. “When you actually play an experience then you actually understand the dynamics, you inhabit it. I can teach you more in two hours than you could get from an eight-hour book.”
The world is a complex system where incentives rule: Danny’s position has allowed him to interface with people from many parts of the professional world: government leaders, generals, tech founders, non-profit volunteers. Each of these groups inhabit organizations that are sometimes at odds with each other, and trust is often only built through the correct alignment of incentives. Danny once created a game which ‘included 54 players working simultaneously trying to find different types of threats’. The real world can be even more menacing. Plain old tactics of ‘game-theory’ may not work in such a strongly interrelated world.
Embrace Radical Uncertainty: The world will continue to shift, and change in highly uncertain ways. The only way forward is to accept that. New technology is disrupting your business? Great. What tools do you have that will help you dig yourself out? What alliances can you forge? Leaders of all sorts often have very tight windows to make many decisions, sometimes with far reaching consequences. Danny’s approach to this is to become a rapid learner. And all it takes is a blank sheet of paper, and a reporter’s quest for truth: “And so I always start with the blank sheet of paper. I start with a problem and I start reporting. And I think that's been the course of my entire life [which] is always, you've got to get to real truth.”
I think risk gaming offers this ability, not just for VCs but also very busy CEOs, senators, et cetera, to step back from the daily fray and say, "How do I just get better at the basic decision making of my job?" What's been interesting as an observation, and I don't know if this will always apply, but what's been interesting is when I talk to staff members, they're always like, "This doesn't seem very useful. It's going to take five hours, the senator's very busy, he buckets things in five minute increments of time." And then I talked to the senator and they were like, "This is incredible. I would do this tomorrow." I think as you move up the stack so to speak, you start to realize how important strategic decision-making is in ambiguous circumstances and how hard it is to actually train yourself to get better at that.
— Danny Crichton
Episode Links
Transcript
Jim O’Shaughnessy:
Well, hello everyone, it's Jim O'Shaughnessy with yet another Infinite Loops. I'm super excited about today's guest. Danny Crichton is the editor in chief of Riskgaming Newsletter and head of editorial for our friends at Lux Capital. You, man, you are like, there's just so much on your resume. It could take us 10 minutes to read through that. You're a Fulbright Scholar, a former managing editor at TechCrunch, and you've worked as an early stage venture capitalist at General Catalyst and Charles Rivers Ventures. Danny, welcome.
Danny Crichton:
Thank you Jim, thanks for having me.
Jim O’Shaughnessy:
So let's start out with the background on what was the environment that led everyone at Lux to decide, you know what? We should launch a Riskgaming initiative. It was launched just April of 2024, right? So it's relatively new. So give me and our listeners and viewers a background on what led to that.
Danny Crichton:
Yeah, I mean, let's rewind the context, right? It was November 2021, I joined Lux and we were still in this kind of malaise post-COVID, right? So in our Menlo Park office, the health department still didn't allow more than one person per room. And so we still had to meet entrepreneurs either through glass or through Zoom that you sort of looked at each other, but you couldn't actually be in the same physical space. And so no one was going out to dinners. It was still hard to go out to drinks or networking events. And as we started to kind of reactivate this, people were saying like, look, it's great to be back at a meetup group or an event, a conference, a session, or whatever the case may be, but it seems like kind of the same old thing over and over and over again. And a lot of my background's in foreign affairs, foreign policy, and riskgaming, or what's normally dubbed wargaming, is a pretty common tool to connect people together to help them understand a complex subject.
And so it sort of hit me that we have this problem, which is we want to bring together people from technology, business, science, policy, all into a room together. And there's this tool we have, this gaming tool, that allows people to inhabit characters they don't normally get to do in real life and put it all together. And so in early 2023, I started writing the first scenario, one about climate change and national security. We trialed it with more than 200 early-stage technology founders. Thank God they're so sympathetic to some of our early forays here. And then it's expanded since then. So we've hosted generals, congressmen, CEOs, think tank presidents all across multiple games now over the last two years.
Jim O’Shaughnessy:
And what are some of the outcomes that you've had in all of the various riskgaming sessions that really surprised you and really conflicted with your thoughts about where it would come out versus where it actually came out?
Danny Crichton:
One of the things which has been interesting, I don't know if it was surprising, but when we play with a group of technologists, we often find that people start very, very cooperative, and then as the game goes on, they start to get more and more competitive as they sort of optimize for their own advantages, their own goals and objectives. But when we play in Washington DC with politicians, with policymakers, we actually see the opposite. We start with extreme competition. No one cooperates at all. Everyone's focused on their own goals and then they quickly realize that there's no way to succeed in that context and so they actually have to start to cooperate. So what's interesting is, at least among the technologists, they sort of end up in a very anti-social, very competitive place, whereas in DC and policy, we see kind of the opposite. They start very aggressive and eventually get used to each other and sort of work its way all the way through.
And so it's interesting because when we combine these groups together, it creates an oil and water feel where no one knows who to cooperate, people are sort of anti-competitive, and oftentimes it never comes together because the whole group is starting from different points of view. Those technologists who wanted to cooperate early on feel that there was treason at the table, so to speak. Meanwhile, the politicians who had been building alliances are now saying, well, look, catch up with me. I'm willing to work with you now. I know I sort of messed with you in the first few rounds, but trust me at this time. And you sort of realize how important that first round is, that trust that you either build or don't build, whether someone backstabs someone else. That openness that you have in the first couple of minutes of the game totally sets the tone for the rest of the evening.
Jim O’Shaughnessy:
Yeah, I had often advocated giving a trial to taking all of the think tanks in Washington, left, right, and center, putting them in a room together and seeing if they could come up with 10 things that they agreed on, and that they would all sign the policy, every one of them would sign it at the end, and didn't get many. I talked to friends at think tanks and they're like, that's never going to happen. Do you think that, and it did happen to a lesser extent, I think with Brookings and either, I think it was AEI, and they have different political views, and I think they did something on healthcare. What do you think about that process in terms of requiring a unified statement at the end of the gaming that everyone is willing to affix their signature to?
Danny Crichton:
I think what you're getting at is the importance of going from one-off games, so if you could look at the theory, you have one-off game theories, and then iterative game theory, where you actually play the same people over and over and over again, and the entire outcome changes dramatically. Because when we know that we're going to interact with the same people over and over again, we tend to want to build trust. We tend not to want to defect. We even if we have an advantage or an advantage we could have an opportunity for, we'll avoid doing so knowing that there may be consequences in the next 5, 10, 100, 500 interactions. And that's one of the reasons why entrepreneurial regions like Silicon Valley are so successful. Thousands of investors, tens of thousands of founders, we could all sort of backstab each other all the time, but we end up realizing we're going to run into the same people over and over and over again and so creates this level of trust.
And so when you're talking about bringing different types of, let's say partisan groups together in DC, I see much of the same thing. Most people in the same partisan tribe interact with each other. And so there's trust on the right, there's trust on the left, but there's no connective tissue that brings the two together. And so whether the prime is saying, look, you have to work together to reach an end point, yes, I think that it starts to actually create that sort of relationship. But what's even more powerful is just to avoid the policy statement and just say, why don't you just do an obstacle course together? Realize that you're both humans, you can work together, you can improv, you can go hunting, you can go to a Spartan Race over the weekend or whatever the case may be. But as soon as you actually start to build that kind of teamwork coherence, realizing that actually you each have something to offer, you might find that the policy sort of automatically runs itself once you've built the trust and glue that holds those relationships together.
Jim O’Shaughnessy:
Yeah, I'm a big fan of Douglas Hofstadter, and in his book Gödel, Escher, Bach, he talks about the prisoner's dilemma. And they used to have, they might still have computer code competitions to play iteratively the game. And the one that I was intrigued by was the one that had the shortest lines of code, and it was cooperate first, and then if the other guy cooperates, you could continue to cooperate. If the other guy rats you out, retaliate, retaliate and rat him out, but then go back to cooperation, right? And so so much of what you're doing is due to our friends, Mr. Von Neumann, Mr. Morgenstern, and Professor Nash. And are you building in classic game theory situations into the process?
Danny Crichton:
Yeah, I take from the three you just mentioned, Schelling would be another, the whole family of scientists and researchers who've done this at RAND, at Harvard, and other institutions. So I'm always inspired by some of the canonical examples, prisoner's dilemma would be a great one, where each of us has an opportunity to defect so how can you build trust? Would institutions do that? But to me, what's interesting about Riskgaming and the opportunity that comes from being here at Lux is the world's just really complicated. The world is very complex. Incentives cross not just two or three parties, but hundreds of parties simultaneously. And so when I want to build these games, I try to really encompass as much of that complexity into the simulation as possible. Now, that doesn't mean it's not fun, it doesn't mean it's not simplified, it doesn't mean it's an overwhelming rule set of hundreds and hundreds of pages. It's amazing how much you can do around two or three decisions.
But for instance, we ran a game earlier this year around AI deepfake election security, and that included 54 players working simultaneously trying to find different types of threats with the goal of highlighting just how challenging it is when you're coming from business, intelligence, foreign affairs, think tanks to work together in an environment where no one actually sees each other on a day-to-day basis. And so you can't do that with game theory. I mean, you can do it with computation, but we can't do it in the sense of like a human can comprehend what's really going on there. And so I find that when you put people into a room, give them an individual role, particularly a role that they don't normally play on a day-to-day basis, they come at it with a fresh perspective and a fresh lens.
And so sometimes they do pull in from game theory. We actually have had players who are like, I read this in the textbook, and I'm like, it's so good because there's a secret here and that is wrong and you're going to mess it up. And so to me, it's people who are able to walk into radical uncertainty, not know any of their bearings, but start to figure out, okay, here's the resources I have. Here are the decisions I can make. Here are the friends, maybe allies that I can build, and who might be some adversaries down the line that I don't want to cause them to create themselves. I don't want to trigger them to become adversaries, but if they were to become adversaries, how would I protect myself going forward?
Jim O’Shaughnessy:
Yeah, and I was fascinated by one of the outcomes of one of your games where it was about rebuilding a shipyard in a area that was increasingly subject to flooding due to climate change, increased hurricanes, et cetera. And one of the things that kind of fascinated me was this idea that it seems, at least from what I read, that participants tended to move towards a more short-term outcome that was favorable to their character, the character that they were playing, and kind of ignored the longer-term outcomes of those kind of suboptimal decision making. And in addition to being dismayed by that, because it also seems to be a problem not only of American business and business everywhere, but specifically American business. This quarterly numbers, we've got to get this quarterly target hit. We've got the short term thinking, in my opinion at least, and this is just my perception, the people who focus on a much longer term time horizon tend to, and I underline tend to, come up with better, more optimal solutions. Talk a bit about that and your experience with that particular game.
Danny Crichton:
Yeah, I mean, that game was focused on Norfolk, Virginia. So Norfolk is the fastest sinking city in America, and it is also the home to three of the most important installations for the Navy, and that includes the home bases for half of our aircraft carriers, basically the Atlantic fleet. And so the question is, and the trigger for the game was why. Why would these installations be in the single worst city you could possibly imagine, which is one that's sinking as these shipyards on the coast are basically going below the water. And so the game really looks at why do we stay in status quo? Why can't we get out of this? Why can't we change? Why can't it be adapted? And so it models a mayor or a congressman, an admiral at the Navy, the CEO actually owns the shipyard, in real life the shipyard is privately owned, it's publicly traded on the NYC.
And so we started to look at those short term considerations that you're getting at. So the mayor and the congressman are running for reelection. Congress is every two years, mayors every four. Our CEO is looking at their stock price and quarterly earnings. They need to keep hitting these checkpoints, otherwise they're going to be fired by the board. And the admiral is looking to get promoted, so they need support in Congress in order to be able to go do this. And so all these characters can have an amazing pro-social, pro-trust outcome, but they each have to give up a little in order to get there.
They're going to have to give up some earnings in order for the company to rebuild the shipyard and make it more durable. The mayor and the congressman are going to have to make politically tough decisions, decisions that are likely to hit their poll numbers as workers say, look, you're going to raise my taxes. Yes, you're trying to protect the shipyard, but I have to pay those taxes today. And by the way, I don't even work at the shipyard, so why should I pay these other people who are working there or for the US government? Why shouldn't the feds pay for that? Isn't the Navy something that we all take advantage of and share?
And so we get into this challenge, and you could call it game theory, but I feel like it's more about the time pressure and when you can make those decisions. Because when I think of our CEOs and our portfolio, when I think about venture capitalists, the challenge isn't necessarily making the right decision. It's being able to make the decisions in the order and the timing that you want. So you know that AI is coming, but if your fund size is not at the right proportion to the size of the market, it's really hard to build a return because you don't have the right denominator in order to compete. If you're building a company and you're underfunded, you can't hire the right talent at the right price points. Or vice versa, you have too much money, in which case it's almost impossible to sort of generate the return that you need given how much you've already taken from investors.
And so you realize that you have these very tight windows in which you have multiple decisions to make, but you can't sort of ever get fired. There's a minimum, you always have to have enough credibility, enough share price, and very, very few people can do what's necessary over a long period of time and not have a consequence from it. And so I look at companies like Intel, which has been on a sort of downward slide for many, many years, one that I covered at TechCrunch and elsewhere, where, look, I believe that Intel has the capability, IP, and talent to do amazing things, but the leadership can't take five years off from the public markets and do what's necessary to get the company to where it needs to go because investors won't allow it.
Jim O’Shaughnessy:
Yeah, my old life was investing in public markets using quantitative models, and one of the things that continued to surprise me was the number of people, really, really smart people, who were just hyperbolic discounting. And everything was like, yeah, no, I can't make that decision for five years from now. I got to make the decision for right now. And what's interesting to me is that bleeds over into the way clients look at things. So we did a study that basically showed that looking at a money manager's three-year returns was potentially the worst and most misleading. And there's that other studies that duplicate it where they looked at the managers who were fired because their three years of underperformance versus the managers who were hired, presumably because they had great three-year performance. Well, guess what? Many years later, the managers that were fired were outperforming the managers that were hired. And it demonstrated the true economic negative impact of shorter-term thinking.
Danny Crichton:
Let me add here. 10 years ago, I was part of a utility company, an energy company's CEO forum. So we had about 18 energy CEOs from all across America, some of the largest that are out there right now. And the utility industry is obviously not a very innovative one, right? It's grids, it's transmission, it's distribution. You're getting power from point A to point B. There is innovation, but it's a very constrained fuel by equity analysts because what ends up happening for every utility CEO, and the vast majority of them have dividends that go into retirement accounts, that go to pensions, and so it's very tight on return on equity. Everyone really cares about every penny. And we asked the CEOs like, why didn't you move to the smart grid? Why aren't you upgrading your grid? We've seen, this was 10 years ago, but now we've seen fires in California. We've seen Hawaiian Electric having massive issues on Maui. We know that companies have been underinvesting in their infrastructure. And a CEO just said, look, if I'm the first person to move, this is the first mover disadvantage. If I'm the first mover to move, and I started investing in a smart grid, I started upgrading the infrastructure that runs and empowers the wealth of this company, my return on equity will go down, maybe even just a little bit, maybe 0.1%. But I am now at the bottom of all of my peer competitors in the industry. And the first thing that any equity analyst worth their salt is going to do is be like, well, what's going wrong with A company's CEO? In three months, this person has managed to take an amazing performing company and turn it into the dirt.
He's got to be fired immediately and the board is going to read that report and fire the person. And so they said, look, if we could collectively in this room, we represent more than half the distribution power in the United States, if we could collectively say, look, we're all going to lower our returns 0.5%, 1% and reinvest that into the smart grid, it would happen. But no one of us can be the first mover without serious consequences or on professional careers into the company itself.
Jim O’Shaughnessy:
That's fascinating to me because it just touches on so many things. So for example, in my mind points out a structural deficiency in some of our laws, for example. For example, you can't have all of those utility chairmen and CEOs in a room together agreeing to collude, right? Because of the legal structure that they have to operate under. But also a question for you. Do you think that that structural problem is what makes most of the innovation come from newer companies, startups, et cetera, and creative destruction, Mr. Schumpeter or... Do you think that's one of the things at play here?
Danny Crichton:
I think how you get evaluated for performance changes your behavior a lot, right? So for these big company CEOs, you have a collective action problem. Equity analysts don't care if you want to invest in the smart grid, they want a return on equity. But when you're a new company, you have venture capitalists on your board, you have angel investors who are looking for you to go bold and build big companies. Now that has downsides as well, right? We don't build as many sustainable companies 'cause we want to be very ambitious. We want to shoot for the stars. We want a $100 billion company rather than a $300 million sustainable business. But at the same time, you're not constrained by, look, there's 20 other competitors just like you. And any change to your business is going to change your stock price dramatically. And I find that very fascinating because when I was at TechCrunch, we were owned by Verizon, which was a large telecom, competing against AT&T and T-Mobile. And when you read the equity analyst reports for the telecom industry, I mean we just get every number is absolutely itemized across the three of us. Every change of the business is exactly compared. And they're different businesses, quite frankly. But they get summarized down into these reports. Whereas I think when you're in the startup world, we have market maps, but when there's only five people, it's very hard to say, well, this company is identical to that company, which is identical to that company—venture just gives you a lot more flexibility to make operational decisions that I think incumbents just can't make.
Jim O’Shaughnessy:
Yeah. And then of course, the solution to the problem with the wanting unicorns, wanting a hundred billion whatever on the startup side are companies like Tiny, for example, back in the day. They were like, no, we'll buy that 300 million in revenue. That isn't probably going to grow very much but boy, we love it. And one of the things that they were able to do was the venture capitalists were like, I don't want this in my portfolio anymore. And so there were a lot of incentives in place that would allow the companies to come in and create quite an amazing portfolio of companies with a hundred million in revenues, but that were solid and growing, but growing very slowly.
So not fitting the traditional venture bucket. I think that one of my prejudices would be, hey, let a thousand flowers bloom, and when many of them don't, some will just outright fail. Others, however, will reach that space in which the Tinys of the world are interested in them, etc. And the idea though gets back to incentives, which I think are also really fascinating here because good old Charlie Munger, if you tell me the incentives, I'll tell you the outcomes, how do incentives of your various players in the game, how do they affect your outcomes in these gaming scenarios?
Danny Crichton:
Well, the incentives are everything. And I would say what makes riskgaming (in the designs that I focus on) a little bit different from DC and sort of the typical war gaming model is there are no judges. And I also don't direct players to take any particular actions. What I do is I do just create the incentives. I try to get those to be as accurate to real life as possible, and then I just let people play. Because what I have found is rather than trying to control or say, look, you got to go down path A, I just put in the right incentives in place, and you're going to probably choose A. You could choose B, B might actually be better for society. It's sitting right there. No one is telling you you can't do it, except that your incentive, whatever it might be, your stock price, your poll numbers, your support in the small-medium business community, the number of subscribers in your Substack is going to be affected by the decisions that you make.
And I think to combine a lot of these different threads together, but I think of capitalism. I mean capitalism is the study of incentives. And I don't think enough people focus enough on asset allocators and their incentives. So we think about why is there so much money in AI versus other assets? Well, there are asset allocators who go, look, I see this massive expansion. I'm overindexed on maybe crypto or bio or a couple of different areas, and I want more exposure to this particular area. And I don't really care how I get it. And so suddenly you see a hundred billion dollars flood into a market that otherwise didn't have it a year or two ago. What we were just talking about, utility company CEOs. I am a 65-year-old pensioner. I need reliable returns. Electricity is not going anywhere. We're not going to slow down.
If anything, thanks to AI, it's actually going to grow. Utility companies see, over the last two years have been some of the fastest growing stocks on the market, but I want that dividend. I need cash. I am paying for my own medical bills, and it's really important that I have that. So those companies are constrained by those incentives. And so when you were talking about tiny versus the go big or go home venture model, to me the big challenge is can you underwrite different types of incentives? Can you create asset classes that are comparable, that offer those trade-offs that allow me to model them.
From the history, if you look, some books in the history of accounting or the history of insurance, all of our markets stem over the last 400 years from the ability of a Lloyd’s to underwrite a ship going overseas to, in modern day the collapse of a lot of home insurance markets where catastrophe bonds aren't enough to cover hurricanes and wildfires and other types of disasters. The only way we can build is to understand those incentives to be able to model them and to be able to understand them. And so I'm with Charlie Munger. I just wish that more people sort of grasped that and did the analytical work to appreciate it more.
Jim O’Shaughnessy:
Yeah, that was one of my go-to metrics to look at when trying to gather additional information outside of what our models watch the insurance industry, watch what's happening over there because that happens well before the downstream effects.
Danny Crichton:
Well, when you talk about truth, and I always joke, insurance companies are the most obsessed with truth in the market. They have to underwrite, and when they're wrong, it costs them a lot of money. And it's a competitive industry. And because it's competitive, you want to get as close to the truth as possible because otherwise you're leaving money on the table. And so it's a unique part. It's not always true in all parts of finance, but it is a unique part of finance where there is this sort of asymptote to the truth.
And that's probably most notable in climate change where regardless of your political views, regardless of what you believe, disasters, the cost of disasters, the cost of replacing a house has gone up. And so insurance companies have to be responsive to that. They just don't have the budgets or the reinsurance coverage to be able to cover those increased costs. And so regardless of any of our views, they just look at the data and they get closer to truth, and they are the sort of canary in the coal mines, so to speak, that indicates what all the rest of us are going to experience over the next couple of years.
Jim O’Shaughnessy:
Yeah. Speaking of AI, have you ever entered an AI as one of the participants?
Danny Crichton:
We have not. I have explored, I've definitely explored. There's a researcher at MIT who's done this around diplomatic negotiations. There's some really fun working papers. I find, I'll be curious how to do it over the next two years. I mean, to me, as we started at the beginning of the show, this was out of post-Covid. There's no devices allowed at the games. Everything's on paper. There are no iPads. I really wanted to make it a manual in-person cohesive experience. And I will say, people always say this, but then they start to use their phones, I've never seen anyone use their phone during one of my games. They just never think to even look at it. It's amazing how distractions go away when you're so intensely focused and you don't want to lose. But I am curious to build more virtual experiences and that would sort of open up the floor to more AI agents being able to negotiate, understand their own incentives, et cetera.
Jim O’Shaughnessy:
Yeah, I think you could probably, with the advances that are being made now, you could let the human participants not have a device, but then you could have one device that can listen and that can speak in a pretty good accent. I would be really interested in that because I build AI, every one of our workflows at O'Shaughnessy Ventures has an AI component built into it. And essentially one of the best use cases that we've found internally is just for steelmanning our own arguments. And, importantly, the arguments that argue against what we're doing. We want to look at steelman versions of both of them.
And those have worked out pretty well as has the, whenever I'm writing about something or thinking about, oh, maybe we should go and do this over here. I'll write up kind of a paper on it, and then I'll feed it into the AI and I'll say, so where are my weakest arguments? How could I improve them? And I have found that it's very efficacious in that particular and very limited use case. But the process has been incredibly helpful. Not only does it speed up the back and forth, et cetera, but it's really great because sometimes, and again, now we get down to almost an existential question here, but if I'm showing it to people who are on my team, there's all sorts of reasons why they might not want to disagree with me, right?
Danny Crichton:
Yes.
Jim O’Shaughnessy:
And when I have, it's like the idea that once you, the Peter principle, when you succeed to the height of your own incompetency, and it's like McNamara thought literally from what I've read, McNamara, who was the Secretary of Defense during the Vietnam War, he really thought we were winning the war. And he thought it because he had a horrible information flow. They were basically just saying, oh yes, of course, yes, Mr. McNamara, that works the way it does. And then when you bring it into the AI, he's like, you're a dumb motherfucker.
Danny Crichton:
Well, I could see a world where I don't have any criticism, but my AI bot, Bert, does. Bert, what do you have to say? Well, actually, John, this is a terrible idea and only saying this for a friend, but you really need to reconsider everything you're just doing right here. I do think it's interesting because we're getting at, if you think about the development of AI over the last couple of years, particularly AlphaGo, a lot of these were built around competitive evolutionary models. So you used AlphaGo against AlphaGo and competed with each other, and you could scale that up millions and millions and millions of times. Train these algorithms faster. And at some point it overcame the best professionals in the sport, their capabilities to compete. I think the same thing is true in organizations, but you don't have that loop, right? In every organization, we have social etiquette.
If I'm the one who's argumentative in every meeting, no, you're wrong, here's five things... It's very frustrating. No one wants to either be, well, some people want to be that person. I don't want to be that person. Even if it's in the pursuit of truth and the pursuit of profit, whatever the case may be, whatever the organization's goal is, it's a very awkward place to be, to be the annoying gnat constantly pointing out flaws, but AI can be that, right? It doesn't understand context. It doesn't understand etiquette as of today. And so it is much more direct. And so maybe that's something that we'll see as a norm change over the next 10 years as AI... I mean, as we're recording this, you have an AI bot that's taking notes. You can imagine in future meetings, look, you made an argument. How do we go and do this?
We already see that in the context of sales calls where there's better training with AI listening and saying, look, here's how you would improve. Here's how you can increase your performance. But it's been mostly individual, right? I'm doing a repetitive motion. There's a lot of training data. There are better and worse ways to do sales, and so I can be trained to optimally conduct that business. I've never seen in the organization group context. And to me, that is going to be a huge unlock, particularly in bureaucratic societies like America where the paperwork, the processing, the middle management, you add all these layers together and it takes a large company, five, 10 years to buy a pop machine, whatever the case may be. And so I have a lot of potential there per AI to sort of cut through all of our niceties and try to get ahead.
Jim O’Shaughnessy:
Yeah. And we've definitely found that it can be very, very helpful just internally on our various verticals. Another thing you did and described, and I'd like you to talk to me and for our audience and benefit, is this idea of pair programming. Where you take a very senior person and put them with a junior person, not necessarily, you know what I mean, yeah, and what did you find there?
Danny Crichton:
Look, apprenticeships were the foundation for modern society. I mean, if you go back to the Renaissance, the idea of a journeyman going into, becoming a master was intrinsic to the guilds and the trades that exist in the 1400s, 1500s, Florence, going up to the Netherlands in the 15 and 1600s, we sort of lost that culture, right? Because we went into the industrial era. There started to become this massive explosion in the number of people. I mean, it's actually really, really important. It's one of those grounding analyses you can do. But look at the number of people who existed in the world, either by region or just globally over centuries. And you see this exponential growth in the 1800s and 1900s due to an increase in food and economic wealth. And so we went from a model of individualized training and education where someone will train me on how to go do this to the industrial education model of public schools, institutions, infrastructure, going all the way up to secondary, tertiary at the university level.
And so suddenly you weren't a journeyman to a master, you were one of 50 students in a classroom with a blackboard, with someone yelling things at the front. Now, if you read the research in education studies over the last five to 10 years, one of the single most powerful interventions that economists have found over and over across random controlled experiments, all kinds of different econometrics, is the individualized attention of a tutor. If you have tutoring at exactly the right moment, the level of progress that a student can make is incredible. And that's true across math, reading, it doesn't matter what the skill is, when you have an intervention that's individualized, even for a limited period of time. It doesn't have to be four hours, it could be for 20 minutes. There is an absolute quantitative, verifiable improvement in that student skill. And so to me, I think we've seen this in coding as you’re pointing out, this idea of pair programming. You put it as senior and junior, but ironically, it's actually, even pairs in parallel with each other can actually be very effective.
Just having a second pair of eyes in the same way that I, as a journalist writer, work with editors and vice versa as an editor to other writers, having someone read your work and being like, look, I know what you're trying to do here, but you're too involved in your own coding. You're too involved in your own writing. I got to take you out of this, I'm going to do it this way. And you're like, oh my God, that makes total sense. Why didn't I think of that? It's because you got too into it. You got too obsessed over some details. And I think you could expand out though from those sorts of fields to basically everything from the medical professions and nursing, into business. I mean, those of us who make business decisions usually don't get to do it in a collaborative element.
We don't get to work with our competitors and say, well, what are you doing in the exact same context? This just happened in our industry, how are you responding? Because it's antitrust and it's competitive, so I can't actually work with my peers who would make that same decision. And so I think one of the biggest challenges, even when I think about, what we try to help with our founding CEOs, is how do you connect peers who are not competitive with each other, but are making very, very similar decisions? Because if you can learn and pair and take that knowledge from each other, just like having an individualized tutor, and that's very, very powerful. But based on the evidence as well as my own anecdata.
Jim O’Shaughnessy:
Yeah, the individualized tutor is another use case from AI that I find very compelling. We're investors in synthesis schools, which is essentially doing AI tutoring. Results are pretty extraordinary. We are also investors in Stability AI, which just went through a recapitalization. But much of the earlier work there was also, we did an experiment with the African Nation of Malawi where they asked us for AI tablets with tutors, and they found that the kids who were getting the AI tutoring on the tablet were way ahead of the kids in the traditional classroom environment. And my observation was, well, I think it might not necessarily be just the AI, it might be the fact that they're getting individual attention.
Danny Crichton:
Exactly right.
Jim O’Shaughnessy:
Right. And so I'm a huge believer in that. And as far as your comment about the numbers of people, absolutely true. I don't remember what decade it was, but Robert Anton Wilson wrote once that, I think he wrote this in the eighties, that there were more scientists and researchers alive at that point he was writing it than at any other time in human history. In other words...
Danny Crichton:
More people, yes, yes.
Jim O’Shaughnessy:
Yeah, take all of the scientists of yore, and there were more people doing those, living today than all of the accumulation of human history. So that numbers and scale problem was an interesting idea to me. How have you changed the structure of the game? And we, for example, one of our O'Shaughnessy fellows, William Zeng, attended one of yours at Maxwell Social. And he thought it was great, and his feedback to us was that he'd like more, right? He said, I'd like more, I'd like more free... Well, he got the fellowship from us to advance open-source quantum computing, so...
Danny Crichton:
There you go. There you go. Doing okay.
Jim O’Shaughnessy:
He's doing okay in the old gray matter…
Danny Crichton:
The old noggin, yes.
Jim O’Shaughnessy:
But what kind of changes have you made or are contemplating to the underlying structure of the game?
Danny Crichton:
Yeah, look, we do have our scenarios, right? So risk gaming is sort of a family of scenarios. We want to keep building them. We're up to five at various forms of draft. We have two coming out in September. I would say in each case, I actually start to know those. So I start with no assumptions. I start with no models. I don't even start with the number of people who are playing. I actually just start with a problem. I start with a thesis, a question I'm trying to answer with the game. So with the first one on climate change, international security, the question was why is the shipyard [which is] most important to the United States's power projection overseas, in the single worst city you could host it anywhere in the United States? Because that to me was a contradiction. And ironically, it was one that no one had covered since I did the game and I don't think it's triggered from the work that I did, but there have been cover stories in the New York Times, Wall Street Journal, Bloomberg.
I mean, it's getting more attention than it did two years ago. And I think that's how you sort of reduce surprise long term. We looked at the Chinese EV market. We had a third one on the future of AI and national security focused around defense procurement. A fourth one, as I mentioned, on AI and deepfakes, particularly around this year of elections. The majority of humans on earth are voting this year for something. And then the fifth one is actually focused on chip fabs in Phoenix. And so the water usage balance between farms, the chip fabs, and residential growth. And so in each of these, I just start from the basics. Who are the players? Where are the trade-offs coming from? Who is sort of competitive with who even if they may not realize it?
So we're starting to build the Phoenix game right now. You have farms who have had water rights going back 150 years off the Colorado River. Now these chip fabs, this new industry that's funded with the US Chips Act. Billions of dollars. President Joe Biden has stopped by, has launched this big initiative here with TSMC, Intel, and other chip fabs, but they need the same water that everyone else needs. And Arizona is a desert. And so do you knock out these farmers, many of which are families wedded to the state, some of which were there at the founding of Arizona. And then all of this is in competition with residential growth. As humans, we need water as well, our houses need water even if you don't have a front yard. And so in Phoenix, you need proof of a hundred years supply of water in order to build a new house.
And so what you actually have is this tension between, look, I want high growth, high industry, more jobs, but those jobs bring people, they now need more housing, and there's only so much water to go around. And given climate change, it's getting tighter and tighter over time. And so I love being different types of people together, figuring out, well, what are those exact trade-offs? Maybe the farmers and the chip fabs actually work really well together. We can actually do work, we can report, we can talk to people who are on the ground going to do this. Maybe there's not as much of a conflict that it appears from a distance.
Or it might be the opposite. There's another group that we didn't even consider that's part of this nexus here and that we can add into the game. And so I always start with the blank sheet of paper. I start with a problem and I start reporting. And I think that's been the course of my entire life is always, you've got to get to real truth. And the more that you sort of assume through secondary, tertiary sources, the more you're going to get wrong.
When we're covering some of these issues, you can talk to the people who've actually done this. In the case of our naval game, we talked to the head of procurement for the Navy and we interviewed them, and we got information from them that was very crucial to understanding some of the trade-offs that the Navy had to make in the context of the game. And so I think that ultimately this is in some ways a replacement of the policy memo, that long formed journalism article that would've been 20,000 words that you don't really remember the conclusions from because it's a lot of pictures and words and it's hard to remember. When you actually play an experience then you actually understand the dynamics, you inhabit it. I can teach you more in two hours than you could get from an eight-hour book.
Jim O’Shaughnessy:
Has the outcome of the various games that you've conducted impacted investment at Lux?
Danny Crichton:
I don't know if it's impacted investments so much as I think it does help improve people's broad decision making under uncertainty. What I mean by that is we invest in new spaces and Lux invests across the heart sciences. You're looking at areas of science where in some cases there's no investment that's ever been done before. We're not looking at 50 companies in a software as a service SaaS space and saying, well, "Which one is fastest growing and you can get all the metrics and download into an Excel spreadsheet, run some number crunching and have your answer." We're in cases where it's like, is this technology two years away from commercialization, 10 years away, or a hundred years away? And we have to do work, we have to figure out how do we make that decision.
And most importantly, because oftentimes these categories change, we have to actually know what would we be wanting to see that would change our mind that we're going from the academic lab into the commercial world. Is it a specific discovery? Is it a slow progression of the technology, it gets better and better and better over time? What is it that would trigger us to change our minds? Because I think we've had Danny Kahneman who unfortunately passed away last year, but he visited us on my podcast about two years ago and he's like, "People never change their minds. People are locked in." That's one of the theories [in] Thinking Fast & Slow, is it's very, very hard to change our minds.
I actually think you can, but you need good decision-making apparatuses, good mechanisms to say, "Look, I'm wrong. Here's how I would know I'm wrong and here's what I would do if I saw this sort of information." The idea of a pre-mortem. I think that that has been taught through these games and the Lux team and more broadly where we come in with an assumption, we leave and we're like, "Well, we got a lot of it wrong early on, what went wrong?" And we're iterating and learning all the time through that. And that's harder to do in the context of a partnership meeting, it's harder to do in the ebb and flow of a investment team where we have a couple presentations every Monday, there's just always chaos, everything's always going on, you're always onto the next thing.
I think risk gaming offers this ability, not just for VCs but also very busy CEOs, senators, et cetera, to step back from the daily fray and say, "How do I just get better at the basic decision making of my job?" What's been interesting as an observation, and I don't know if this will always apply, but what's been interesting is when I talk to staff members, they're always like, "This doesn't seem very useful. It's going to take five hours, the senator's very busy, he buckets things in five minute increments of time." And then I talked to the senator and they were like, "This is incredible. I would do this tomorrow." I think as you move up the stack so to speak, you start to realize how important strategic decision-making is in ambiguous circumstances and how hard it is to actually train yourself to get better at that. And so I've just generally found that very successful people really enjoy improving their own cognitive machinery and are always looking for excuses to do so.
Jim O’Shaughnessy:
I agree. I knew Danny as well, and we actually got into a chat once, because he was maintaining people don't ever change their minds. And I was like, "Well, if you'd indulge me, of think I try to do that all the time," and I went through how I do that. One of the things that I'm always thinking is, "I'm probably wrong." Basically because we get this confusion between an opinion and truth. An opinion is really a perception, and your perception is different than my perception. Everybody has different perceptions. But if you're developing a thesis or a mental model, how do you test the efficacy of that? Well, you put it into practice.
How well does it work? Pragmatically, is the use of that way of thinking leading to success or leading to failure. And I'm only speaking directionally here. If it is mostly leading to success, you can still learn from the times where it isn't leading to success and try to refine and iterate. If it's not leading to success, then you've got to be willing to say, "That theory was wrong." I think that one of the things that worked very well for me was I don't attach those theories to me, to the definition of Jim. I don't say, "That's a hill I will die on." I subscribe to the General George Patton idea that I'm going to let the other poor dumb bastards die on their hills.
Danny Crichton:
I hadn't heard that one before. Look, I think what you're getting at is this idea of intellectual humility. Unfortunately, it sounds great. It's one of these values that everyone should have intellectual humility, we all want this, but in reality, all of us have jobs in which we have to make decisions, oftentimes under very trying circumstances, very quick with limited information. The reality is, and it's not just unique to American culture or business culture, I think it's a global problem, is the most confident people tend to rise to the top. There's a parallel to the Peter principle that we were talking about earlier, but it's like the confidence game matters a lot.
People like confident people. You don't want to go to your doctor, get four tests and go, "I don't really know the answer, but if I had to guess it'd be 30% this, 40%." It's scary. It sounds like you don't know what you're talking about. You prefer to go to a doctor and be like, "Look, I know I went to Harvard Medical School and the answer is this." As a patient—and look, I have a stats undergrad training, and so I come from a probabilistic background, a computer science background, I love the first doctor. To me, the first sign of a first rate intellect is someone who not just has a humility, but has a sense of, look, there's a couple different paths. This is where I'm leaning to.
But look, I could be wrong and this is what I'd be looking for on the other paths I'd go. But the vast majority of people, whether it's in medicine, in the professions, your lawyers, whether it's in venture capital, CEOs, we want someone confident next to us who's saying, "Look, I know the right answer. I have confidence. I don't want to know that there are six other alternatives and maybe you're wrong and we're all going to lose money, I'm going to lose my health, I'm going to lose my house in a mortgage dispute, I'm going to lose my bankruptcy proceedings or my divorce." And so I actually think it's one of these values that's constantly in conflict with the reality of life, that we really put a value on humility but to be successful, you can't really have that.
And to me, venture capital is the asymptote of that. You will get nothing done if you do not have audacity in a partnership meeting. I've sat across three partnerships over my career, very different groups of people. But the reality is if you don't come in pounding the table, believing in a founder and entrepreneur against opposition from your partners, you're never going to make it. And you can't overcome opposition with humility. You can't be humble and say, "Well, I kind of think that this person is really good, but you all could be right," because no investment will ever get made. It's interesting because what you really need is this extreme level of confidence with the recalibration occasionally of your thinking and thoughts over time, that keep you aligned with truth that's happening in the marketplace.
Jim O’Shaughnessy:
Well, I chose the Captain Kirk solution when we launched our very small venture fund. We don't have any LPs.
Danny Crichton:
That solves a lot of problems.
Jim O’Shaughnessy:
Yes. I solved a lot of problems with that. But as you were talking about the doctors, I actually used to use that as an example of why using empirical contested investment models made sense, and I used doctors as my example. I would say, "Which doctor are you going to believe? Are you going to believe the doctor who you go in and say, doc, these are my symptoms, I'm feeling horrible. What can you do?" Are you going to believe the guy who says, "I just got these little yellow pills by the pharmaceutical rep, and I've got a really good feeling that these might help you." You're going to run out the fucking door. However, my argument was I would trust the doctor no matter what... Even if he gave me for Provisos and a variety of the type of probability analysis, I would trust the doctor who said, "I think that this is your condition. Here are the 10 years of double-blind tests on this specific medication that I'm recommending for you. What do you think?"
Danny Crichton:
Right. Yes.
Jim O’Shaughnessy:
But you are absolutely correct. You are 100% correct about we always go with the person who seems to be the most confident.
Danny Crichton:
It's a confidence game. Fake it until you make it is a confidence game.
Jim O’Shaughnessy:
I love that you bring that up because that's where con artists come from, confidence. The funny thing about it is it works everywhere. You're not onto anything new there. And so maybe I've just been very privileged to have reached a point in my career where I'm not really trying to convince anyone of anything, other than the things I want to get done with the company that I've started. I guess I have to convince the people who might want to work for me.
Danny Crichton:
Well, I would actually center this existentially a little bit on, if you were to connect a few dots from earlier in our conversation around utility, CEOs, building company, LPs, what you're getting at is how many people do I have to be confident to? As a venture capitalist at a traditional fund, you have to sell to LPs, you have to sell to your partners, you have to sell to other firms, you have to sell to founders. I have to be confident with all these groups, which means even if I don't really believe, but I have to have an answer, I have to have an immediate response. I have to be aggressive and audacious because otherwise someone else is going to compete and beat me. As a utility company, CEO, I have the same thing.
I'm the CEO. If I don't sound confident, the board will lose trust in me, shareholders will lose trust. In the regulated sector like power, the Public Utility Commission may lose trust in me, and I'm out and someone else is going to be taking over for me. And so one of the reasons I've always stuck to writing particularly in the latter half of my career, is there's less people I have to prove myself to. Readers can walk away. Many readers actually enjoy articles that have more intellectual humility precisely because they're not looking for a glib answer, but they are looking for more depth. I only have to prove it to an editor. I have to prove it to one publisher. I don't have this network of approvals that's required to get this work done.
I think that that is what's challenging with so many other professions, particularly client services ones, think consulting, investment banking, lawyers, doctors. I constantly have to play a confidence game to the people paying me. I have to pretend I know what I'm doing, even when I don't. Imagine being McKinsey asked for $5 million on a client engagement, you did six months of work, and the correct answer is, we don't know. That is truly all the information. Can you imagine? Imagine that MD, that partner, walking into a room with the engagement manager and said, "Look, you put your trust in us for six months, we've held up two quarters of results. You put five, $10 million in this field study, and we have no idea." It's impossible. You can't do it. You have to show a bias to action, you have to show a confidence. I'm reminded only because it's funny that it was circulating, a McKinsey slide from the Department of Sanitation here in New York City where they were like, "Well, it's garbage bins. That's how we're going to solve the rat problems." And it's like, on some hands, you're completely right, on the other hand, it probably shouldn't take 5 million bucks, but you could probably be pretty confident that at least the garbage bins make sense. And then you have our mayor in front of Gracie Mansion with the Empire State Of Mind song going on in the background, putting a trash bag in his bin and closing it up and making it sound like we're the most innovative city on the planet. Confidence. It's a confidence game.
Jim O’Shaughnessy:
I know. I had a exactly that same reaction when I saw that video. I'm like, "Oh, dear God."
Danny Crichton:
Shameless confidence goes so far. It's amazing. And New York is like the Carnival Barker city of world. We're filled with people like this.
Jim O’Shaughnessy:
I'm reminded of earlier in my career, and I always looked at it in that I changed from being a proselytizer, like, "This is the only way you can do it," to, "Hmm, I was wrong about that. There are lots of ways you can do this." I had never brought in the idea of the confidence angle. Thank you. I think I'm going to reassess that little part of my history. But I do remember, my first company was O'Shaughnessy Capital Management, and we were hired by Merrill Lynch to design a stock selection strategy for one of their unit investment trusts, which were very popular at the time.
Old joke, Merrill is not this way anymore, listeners, but this is back in the dark ages of the 90s. I heard a lot of jokes when I was at Merrill, which was remember Chernobyl? And so that was still much more in the public mind back in the 90s. The jokes that I heard at Merrill Lynch that did make me laugh, even though it's horrible joke, was, "Hey, how would you clean up all of the nuclear waste at Chernobyl?" And the answer was, "Put four points in it and give it to the guys in retail."
Danny Crichton:
Incentives.
Jim O’Shaughnessy:
Incentives, that's right. But to get hired, I had to literally convince, just on my own, 10 separate committees within Merrill Lynch, and I felt like Sisyphus man. They had a big campus down in Princeton and literally I had a drive from New York or Greenwich down there and like, "Okay, which mill am I going through today?" You're right. You are absolutely right. Even though I did answer several questions where they got into things really in the weeds, my answer was, "I do not know, but I will find out and report back to you." That seemed to work.
Danny Crichton:
That's like, you have confidence in results while humility about having it immediately, which is appreciated. See, my version of that was getting interviewed by Bain. It was at the final interview with the managing partner, whoever. We had this case study, as they do with all the business consulting interviews. And I came up with some elaborate [process on] how to save the business. It's heading towards bankruptcy. I was like, "Look, there's all these things you can do." I came up with 2025 ideas, and the answer was, "Let the company go bankrupt." That was the answer. The managing partner at the end was just shocked at how many ideas I came up to try to save this business. And I said, "Well, that's just stupid. I could have said that in a tweet and walked out the room. That's the answer?" I got dinged for lack of client facing personality. And so I was like, "Well, clearly that's not the right pattern for me. No one wants any solutions going on here."
Jim O’Shaughnessy:
I love that though, because I have a very soft spot in my heart, and I have many people who like this who are my teammates here at O'Shaughnessy Ventures. They score very low on agreeableness on the big five, and I find it very refreshing when somebody looks at me and says, "What? Are you fucking dumb?" I just think it's great.
Danny Crichton:
I come from Minnesota. I grew up in Minnesota, and so-
Jim O’Shaughnessy:
Me too. Me too.
Danny Crichton:
I didn't know that. Where in Minnesota?
Jim O’Shaughnessy:
Grew up in St. Paul, Minnesota.
Danny Crichton:
There you go. I was in Eden Prairie.
Jim O’Shaughnessy:
Oh, well there you go.
Danny Crichton:
Different cities.
Jim O’Shaughnessy:
No wonder we're getting along so well here.
Danny Crichton:
There you go. I love the Minnesota's in the news all of a sudden, but that Minnesota nice mentality of you can be disagreeable without being in your face. I do think that there is that difference on the coast to the Midwest where we're a little bit more passive aggressive, at least in the traditional sense. Our way of expressing disappointment or disagreement can sometimes be like, "Well, I don't think that that's the correct answer," and it's not direct like you're wrong, and you have to piece together what the real answer is. But to our earlier point about AI bots, being able to interact with other folks disagreeing is something that I think we are losing a little bit as a society. Because of social media, because we're intermediated with our devices. We're losing the ability to have a long conversation where you can have debates and say, "Look, I agree with points two, five and nine, but I don't agree with the others and here's why."
To your point about think tanks much earlier, I do think you can get a lot more overlap. There's not a lot of incentive to do so, but the reality is you never put anyone in the same room. No one ever collaborates and comes together, as we pointed out earlier. And so to me, the ability to disagree to create ecosystems of competition, that old idea of the free market of ideas, the marketplace of ideas, is something that is lost, I think in 2024 and one that I always seek. I always love a debate. I was the person in the lunchroom. I always got kicked out of the dining hall as an undergrad because I would come in right as it opened, there'd be a group of people who would show up at the table. We would all argue for four or five hours. And then eventually they were like, "Please leave because we need to clean up the whole place." We did that every single day, seven days a week. And then after college, you're like, "No one wants to go out for five hours and debate what is the meaning of justice?"
Jim O’Shaughnessy:
I would've gone with you.
Danny Crichton:
Yeah, exactly. These days, people don't even want to drink. You're supposed to even get started. But building up those Mechanisms of debate helps people, I think, sharpen their thinking skills, sharpen their theses, and it's just hard to get these days. People are scared cross that line.
Jim O’Shaughnessy:
Your a remark about being from Minnesota reminded me of, my sisters used to tease me that I actually was not born and raised in Minnesota. My parents actually flew to New York City and adopted me and brought me back to the Midwest. And of course that's a joke. I was born in St. Paul. But really, it didn't take with me. Minnesota nice didn't take too much with me. When we were going back after moving east in 1991, I looked at my wife as we're landing at Twin Cities International Airport, and I said, "Welcome back to the world of false sincerity and passive aggressive behavior."
Danny Crichton:
It is true. Non-confrontational. I say this in startups, false sincerity, non-confrontation, passive aggressive behavior, is the single worst ingredient in innovation.
Jim O’Shaughnessy:
Totally agree.
Danny Crichton:
Scientifically, I agree. If you want to have an intellectual conversation, you have to have disagreements, and you have to have professional disagreements. You have to create mechanisms for going to do that. If you haven't grown up with it, and obviously most of our listeners have not, you have no idea what we're talking about, but if you grew up in a suburb of Minneapolis, I can assure you, the false sincerity is something that you experienced every day and only notice when it's missing.
Jim O’Shaughnessy:
And if you want a crash course on it, watch The Coen Brothers Fargo, because I think that that captures it quite nicely. Moving on to the work that you did, I was very interested in what you were trying to achieve at TechCrunch when you were trying to launch TechCrunch Plus. And so didn't work out. What'd you learn? What from that experience informs what you're doing now, et cetera?
Danny Crichton:
It's most essential, right? The media industry was changing throughout the twenty-tens. Facebook rose up, social became dominant for a lot of people in terms of our audience reading, the advertising markets became more and more concentrated with Google and Facebook and digital advertising. And so like most media companies, we were looking to do a subscription product. How do you get people to pay out of pocket? You can get rid of the ads, you can get rid of the networks, you can get rid of ad salespeople, all this infrastructure we need to get revenue from ads, and build a deeper, denser relationship with the reader.
So people want to commit, they're buying into it, they want to read us every day, and that's really positive, I think, for any media company. In our case, this is back in 2017, 2018 when we were building this out. This is prior to Substack. Actually, ironically, Chris Best, Hamish, and the third co-founder over there, I met them when they were starting YC and they were starting to think about subscription. We were all in the same milieu together. They wanted do it as a platform play, I ended up at TechCrunch trying to do it at a larger established media company. And so a huge focus of it was on deeper analytical profiles of a lot of unicorns.
One of the centrepieces was really to say, hey, we were going from a world of 20, 30 unicorns, remember back in 2017 when there was literally, you could count them, you could actually memorize them. And then we went from 30 to a hundred to 200 to, I think the highest number I ever saw was 965 unicorns. And a lot of them were these companies that people just knew nothing about. Flexport was one that I was very focused on. No one understood logistics, but it was a valuable company. Why is it valuable? Where did it come from, and could we do deeper reporting on that that TechCrunch has not been able to do because we have the subscription, et cetera.
And we complemented that with a focus on founders, so how to build companies. And so you would read these profiles and say, well, how did these people build successful companies? These are sort of the lighthouses coming out ahead of me. And then we complement that with sort of tactics coverage. How do I do marketing? How do I build an engineering team? How do I scale up a product as my users grow those sorts of playbooks? Why didn't it go right? One part of it was definitely corporate. We never had the resources to get it done right. And I think this is a good example of
“where the best laid plans of mice and men / Go oft awry”.
The simple reality is to build a brand like that, we needed five to 10 writers and we had two. And you can do a lot with two people, but you can't do as much as we needed to do to make it an effective product.
And so it's one of these classic catch-22s where it was a really good idea. I actually think it was right. But if it's not resourced, you will not succeed. You cannot win a war without the right army behind you. So that was A. B, we sort of got annihilated by the fact that there was a lot of decentralized Substacks, newsletters, podcasts that came out of all kinds of different places that just ate different parts of this pie. So Acquired, that started back in 2015, obviously has grown to, I mean just recently they were at Chase Center in San Francisco with a practically sellout crowd with Mark Zuckerberg. So I mean, over a course of almost 10 years, growing to these very long podcast profiles of notable companies the same way we were doing. The Generalist by Mario, Packy and a bunch of others with Not Boring. There became this whole world of the deep profile of tech companies that were competitive and oftentimes free.
And then, on the tactics side, some of the venture capital firms started launching their own publications, but most notably YC did a pretty heavy job on that. They really covered the gamut. They had a lot of writers and, obviously, direct access to a lot of the founders to this. And so we were never able to compete. We never had the resources, we didn't have the infrastructure, we did have the audience. And that's sort of the tragedy of the situation is know TechCrunch is read by tens of millions of people. They were already on the site. That was the sort of magic of the place. We didn't have to find these people. We didn't have to convince them to come. They were already here, but we had to deliver them a product that they deserved, and we never did.
Jim O’Shaughnessy:
And what did you learn from that experience that you brought with you in your new capacity as running the newsletter, the games themselves, the podcasts, et cetera, of what you're doing now?
Danny Crichton:
It's really important not to get locked into long-term plans. As I mentioned earlier, TechCrunch is owned by Verizon through Verizon Media. Verizon in the frenetic up-and-down world of media being owned and marooned within a large telco where prepaid or post-pay subscriber counts on 5G -up-sold plans as sort of the metric du jour. It is really, really hard because you would get these budgets, they'd be over for two years. What happened was we would have the budget, we were spending it appropriately, we'd hire folks and then midway through they're like, we actually need the budget to be half of that. And I was like, well, we were on target for the year. So when you say half, you just mean the final two quarters of the year. You're just saying we're going to fire everyone at the end of July, which we did.
I mean, it was a very sad context, but every person I hired, essentially at TechCrunch, was either fired or laid off at some point or departed because they knew it was coming over the course of that period, because we kept having this investment, and then the pullback investment and then the pullback.
And you see this model not just in corporate America, which I think is very, very common across the Fortune 500. The year of the belt tightening. Which senior vice president has the power and influence for the next three months, and then someone else becomes the golden child and does super well? But you also see in areas like international development where the World Bank will get into a project, they invest a lot of money, then they pull back, then they invest in other project, they pull back. And the funny thing is, if you're on the receiving end of this, it's almost always the case you'd prefer less money, but more consistent cash than the rollercoaster ride of budgeting that goes into these places, right? Because you can actually plan, you can actually effectively choose a strategy. And so when I think about risk gaming and a lot of the stuff that we're building here, I'm marooned in a very small venture firm, or at least small in terms of the number of people. I can see everyone here.
So it's a very different place from Verizon, which was a hundred thousand people globally and has a major Fortune 100 company in New York City. But I think the key is you have to be adaptable. So, earlier this year we hosted a game in partnership with Mike Bloomberg. That came out of an introduction of an introduction, which became an interesting opportunity, which became a full risk game, which included eventually five major generals, two congressmen, a couple of think tanks, CEOs, et cetera. We never would've been able to do that if that was locked into a budget. And I have two similar opportunities like that later this year. I have no idea if they're going to happen. There's not even a budget here, which no one actually knows about. But assuming they come together, the place is flexible enough that I said, if we can get a couple of senators into a room, how about a budget?
And so having that sort of flexibility and the ability to pivot strategy regularly makes it feel much more like a startup versus that large utility company CEO, that have to plan 10 years in advance. And I just think that with change in the world and how fast things accelerate, what topics we want to cover change, what ways we want to cover those topics change. So how do we play these games? How do we design these games? The people we want to connect with are going to change? And so I don't want to plan two years in advance. I have no idea where I'm going to be. I don't even know if I'm going to be in America. I don't even know if we will be here in two years. We might all be in caravans driving around in rusted out old Fords. So I have a very existentialist present bias of saying, look, you can only plan so far ahead, and then you go into the fog of war. And so, the more adaptability you have and the faster your iteration speed, the better your offer you're going to be.
Jim O’Shaughnessy:
Just from our earlier conversation, I would say, Danny, not very confident…
Danny Crichton:
It’s very interesting. The government puts together a four or five year strategy, the-
Jim O’Shaughnessy:
(in a glorious accent) The glorious five-year…
Danny Crichton:
Yeah, it's a five-year plans, and ten-year plans. And it's interesting because some governments actually follow through with their plans, but many don't. And I always think it's very interesting because what you really want to have confidence on is that, no matter what happens, you're going to learn quickly. You're going to have rapid learning, you're going to adapt to ambiguous circumstances, and you can mobilize and be effective. And so I covered disaster tech for a long time at TechCrunch, climate change, national security, a lot of these issues where there's crises that show up and what you learn is you can never know what's going to happen. I mean, you think of some of the huge flash points that have taken place over the last year. We couldn't predict when, where, how, why. No one's saying that some of these aren't obvious, but we didn't know exactly what would happen or how intense or when they would take place.
And so, the only thing you have is the ability to mobilize. You have the ability to immediately assess, here's what just took place and here's what needs to happen right now given what we're seeing live on the battlefield or in real life. The same thing is true with disaster response, which I covered for many years. All of a sudden, there was a hurricane that hit Maui, knocked out a historic city. We were completely unprepared for that. It was a kind of one in 1000 year storm. It happened that the land had been very dry. And so the wildfires that struck through this happened to be able to spread in a way that would not be usual in Hawaii, which is very wet as a tropical place. And so you had this unique confluence of factors, and we were really unprepared for how to respond to that.
And so, I just don't think what's going to happen next. And so your ability to assess, adapt, I'm going to use the old SOAR acronym, but your ability to kind of do the OODA loop, so to speak, orient, observe or observe, orient something, something. The devise act, it just becomes really crucial. And the speed at which you can conduct that to me is your competitive advantage against anyone else.
Jim O’Shaughnessy:
Yeah, agility is something that we prize very highly when looking at startups to invest in. And I'm also reminded of an article of a friend of mine, Dan Jeffries, who's a technologist. He was writing about the, I don't want to call him spiritual guy, but the enlightenment writer, Jed McKenna. And in his essay, he did this thing that the moment I read it, I was like, oh, you are absolutely right. And he was talking about the people that he most admires are generally people who are the closest to what he called truth. And what I was interested in was who he identified. He identified special forces because life and death situation. You could be looking at the map, everything could have changed from the time you did your halo jump to get to the ground. And I've had the privilege of meeting many of these guys, and they're so impressive. And you can just watch the agility of their thought when they're thinking, which I think is amazing.
So he identified special forces, emergency room doctors. Again, you have no idea. You got to be ready for anything that might come in that door. And so it really inculcates that kind of thinking. And then, interestingly enough, traders at big banks or big institutions, because it's not a real death, it's a metaphorical death.
If you lose a hundred million dollars for the house, it's your death.
Danny Crichton:
Yes.
Jim O’Shaughnessy:
Bye-bye. But I really enjoy that article because it really crystallized for me a lot of things that were kind of nebulous in my own thinking pattern. And that's where I got this idea of no, no, no, it's how agile are they, right? And I'm a huge fan of Daoism and Lao Tzu had this wonderful thing, the metaphors for life and death. He's like that, which is supple lives. He gives the idea of the new plant, when the heavy snow lands on it bends and the snow slides off, and then it springs back up. If it's an old tree branch, it is brittle and the snow piles up, it breaks off. And so Lao would say, thus, to be stiff of opinion, you'll always be in the wrong, and you are a cult of death. Be supple, be agile, be willing to change, and you are a creature of life. And so I think that's a pretty cool idea for just trying to keep all of our thinking, at least somewhat agile and somewhat open to ideas.
Danny Crichton:
Well, and you think in the post-COVID world, I mean, life changed. We spent a year locked in our homes. I remember doing phone calls with people on the sidewalk where I could see them, and you could see me, but we were on the phone together. And so it felt like we were in the same place, but you're over there. And so we got used to remote work. We got used to distributed teams, decentralization. And then you've seen this fight as you see some companies, some executives with cognitive flexibility. Hey, there are different ways we can work together and how companies are organized. You see how cities are bouncing back. In New York and San Francisco, we've really struggled to bring these cities back. The commercial real estate in Midtown remains anemic compared to what it was just 3, 4, 5 years ago because we have no idea what to use with this space anymore.
We're still hoping that there's some tenants, some company that will buy out all these spaces. Meanwhile, the prices, in some cases, the land is going down towards zero as landlords are going bankrupt, selling this stuff off. Even the Flatiron building here, classic flagship, mired in legal wrangling because the repairs to it aren't matching the sort of rents that it will expect. And so it's completely covered, and it has been for years at this point. And so I think the ability to adapt, look, it starts with the individual. And so you see a huge range. I mean, you're in the venture industry, much like me, you know the founders who are unbelievable, and they're adaptable and they're just a whole higher order level of thinking if they don't lack confidence, but they do know to be confident when they need to change directions. Something has changed in the industry and they go another way.
And you see this in everything from an Uber that's started in taxis, but added in food delivery into the AI world, where people are starting an enterprise, going to consumer or vice versa, started an infrastructure and now we're trying to do applications. But then it starts at the organization. It's not enough to just be flexible yourself. You're also part of groups, you're part of society, you're part of companies, you're part of workplaces. And groups can have flexibility. I mean, it's interesting. You would think that very flexible people would work very well in groups. And I actually don't know if that's the case at all. In some cases, I actually think that the most effective groups tend to have two or three leaders who are very supple, as you were just describing from Daoism. And a lot of folks who are sort of willing to just go whatever way the wind blows, so to speak.
And so, they're just adaptable to what the priorities are of the organization at any given time. And then you scale up to nations, and you scale up to cities. I was recently doing an interview with Paul Collier, author and development economist at Oxford, who just published a book on left behind places. So, think cities. I was born Youngstown, Ohio, so think Steel Towns, rustbelt. And he asked, why are some left behind places still left behind, but others have actually caught up? What makes these places different? And he basically emphasizes its rapid learning under radical uncertainty. Some places realized that the steel was never going to come back. And so mayors, community leaders, business leaders, came together and said, what do we need to do? What's the next thing? Where can we go? Almost like a band on the horizon, say, we need to go this direction. It's a whole new way.
And the ones that were able to try to chart that path were really, really successful. And the others who doubled down were not because they didn't recognize that the world had changed, and it had changed at a level. In this case, was NAFTA and trade and everything else. It traded in a way that was not solvable at the city level. And so when I think COVID, I think about some of the changes that AI is going to do to schools, to corporations, to our workplaces. There are a lot of people completely unprepared and who will double down on the past rather than recommit towards the future. And so I love the metaphor. Of course, those metaphors are always so much more poetic when they come from historical works that have lasted the test of time. They have our analyses of adaptability to the AI revolution. I mean, does anyone going to read that in 4,000 years? I don't think so.
Jim O’Shaughnessy:
I love that idea though. That's going to stick with me. I've always had an aversion to “because that's the way we've always done it” and well, because you can't argue with 200 years of success. Yes, you can. And the way that you've always done it, if it stops working, you might want to stop building it that way or doing it that way because, even if it had an illustrious, glorious history under completely different circumstances and zeitgeist and societal types of behaviors, not going to work now! So always keep your mind open to where things are going as opposed to where they've been. I do think it's very important to know where they've been. I do think having context around history is vital because human operating system doesn't change very often. And we keep fucking up the same things time and time and time again. But if you know that, you can arbitrage it, right?
Danny Crichton:
Exactly.
Jim O’Shaughnessy:
Well, listen, Danny, this has been amazingly fun. Thank you for joining me. At the end of our podcast, we play a little game, and we are going to dub you the emperor of the world. And as the emperor of the world, you cannot kill anyone. You cannot put anyone in a reeducation camp. But what we are going to do is we're going to hand you a magical microphone, and you can speak two ideas into it. And these two ideas will incept the entire population of the world. And whenever their next morning is, they're going to wake up, think to themselves, I have just had two of the greatest ideas. And unlike all the other times in my life, I'm actually going to act on these two ideas starting right now. What are you going to accept in the world's population?
Danny Crichton:
Okay, that's a very bold question, one that I should have been prepared for, but two come to mind. First, the power of decentralization, the power of breaking up the large and making them small. So I think of everything from the monopolists in our own tech industry to large governments, large polities, massive cities. Humans are designed for smaller communities, evolutionary socially. Dunbar's numbers roughly 150. And with some memory aids, you can push that number higher. But the reality is that we live in a world of 8 billion people going towards 10 or 12 billion depending on who's counting. And so it can feel very anonymizing where we just feel like we're just a number in this mass sea of humanity. And so, I think always in my entire career, going to smaller organizations has been more successful, has been more fun, has been more interesting, has been frankly more productive. And so I wish to give that to more and more people that you can leave those big structures, break them up, shred them, and empower in the same way that Schumpeter is talking about creative destruction. That kind of flexible organizations rise, fall.
Well, that doesn't happen if they're monopolists. And so breaking up all these big things can really change things. That's one. Two, it really comes from a lot of French existentialism, and I'm a huge [fan of] Albert Camus, Antoine de Saint-Exupery and a bunch of others.
Jim O’Shaughnessy:
Love them all.
Danny Crichton:
Is sort this power and maybe Smith and Sisyphus into the plague of our actions every day. That agency, its funny, because you said, assuming they had agency and they're willing to do the actions. Well, that was sort of the lesson I was going to give is you have amazing agency every single day to change the world. Whether it's in your own community, whether it's in your own family, yourself. Viktor Frankl. There's just a huge long list of books. Man's Search for Meaning. You can be in the most trying of circumstances, but you are always the arbiter of your own fate. You are always the arbiter of your own decisions, and you're always the arbiter of your own morality. And so the decisions that you make, how you evaluate them, how that contextualizes within the ties that bind you to society are always within your hands.
And I think we can sometimes go on autopilot and say, well, that's just what the company wants, or that's what the country is saying I should do, or my boss, or whatever the case may be. But you always have agency and the people who use agency tend to build it as a muscle. They get used to it, and they realize that the consequences when they go against the grain, when they're contrarian, tend to be a lot less than they expect. The fear does so much more than reality when we make right decisions in our own heads. So the power of agency and the power of smaller groups.
Jim O’Shaughnessy:
I love them both. I especially love the power of agency. I'm a huge fan of Viktor Frankl. Love that book. It's one of my foundational books that I recommend and give away to a lot of people. How can our audience find you online? In New York City?
Danny Crichton:
In New York City, online @DannyCrichton on Twitter, DannyCrichton.com, on Lux Capital's website at the Manhattan Institute. All the different institutions. Would love to meet people or reach out anytime.
Jim O’Shaughnessy:
Terrific. This has been great. Thanks for joining us, Dan.
Danny Crichton:
Thanks, Jim.
 | A guest post by
|