Going to be a super high-volume week, we are up to four and a half hours already!
Fun fact, the voice I've given Nate for these was originally created as "Contract Devil" for Planecrash.
If you're finding value in this AI-multi-voiced podcast version of the post and are already supporting Zvi, please consider chipping in through a Substack subscription. The ElevenLabs MV2 model sounds great (at least to my ears), but it can get pricey at high volumes. (Support Zvi first and foremost though.)
I think private equity has distorted VC — as they go into increasingly risky and earlier stage businesses the only investments left for VC are the ones where they would never consider taking PE money for cultural reasons or the ones where the opportunity is so far out on the probability curve that it breaks PE folks’ brains.
And the long shot math breaks VC brains too. If your expected 2x return was calculated by multiplying 1e9 by 5e-8, I am going to question the robustness of some of the modeling assumptions to put it lightly.
Plus these are the good VCs! Most VC is pure herding and pattern matching where original thought is actively discouraged. If the raison d’etre of your fund is “Sequoia wouldn’t take our LPs’ money” then those same LPs will reward you for looking as much like Sequoia on the outside as possible and don’t understand the adverse selection you face. Complete cargo cult investing.
My other theory is that SV succeeds because it’s the only place in the world where someone who asks for an 8 figure check to fund a company with no existing business isn’t 100% definitely scamming you
Erik Torenberg writes that what is especially important to SV VCs is the social capital returns of being on the ground floor of a major investment (ex. FB, Google, etc.) with the hopes of generating the prestige of knowing the founders before they hit it big (couldn't find the article). However, as Rabois points out you can't be too contrarian otherwise other VCs will be unwilling to invest in future rounds.
SV is contrarian contra the rest of the world, herd-like internally. The believe (with good evidence) it's the crazy ones (Musk, Neumann, SBF) that generate the 100x returns.
I’m not surprised that Bitcoin partisans would hate Ethereum: once you endorse later cryptocurrencies rather than just Bitcoin, the first one, then Bitcoin should be worth zero as the potential supply of cryptocurrency is infinite.
Yesterday I wrote a comment (first one here for six months) on the gambling instalment only to find, when copying it into Substack, that comments were restricted to paid subscribers, which was unannounced and seems unprecedented (perhaps even a mistake?). Because I‘m so slow at writing (managed only one Substack post in 2024) that even a quite short comment takes me time, I‘m going to be impolite enough to add it here. So the gambling post was another one that I found highly interesting, e.g. the insights about Phil Hellmuth‘s modus operandi, and then not least the parts about sports betting against the house, as all I knew in the area of sports betting was Betfair Exchange (and not even much about that), where players bet against each other.
With poker I‘m more familiar, but before reading the post I don‘t think I really knew about the thorough investigations into Lew‘s other live-streamed play that found nothing suspicious. And still I would have gone higher already than 95 percent on no-cheating. Only the no-cheating story makes any sense, it may have been not a misreading of her hand (though I guess that‘s possible) but simply a spite call, a total blowup, caused over time by Adelstein‘s constant aggressive play; and then she confabulated stories to rescue some reputation of competence (which women at poker tables seem to care a lot about —- even though appearing incompetent is usually +EV). People who maintain that she was cheating misunderstand this because they think, many even correctly, that they themselves could never „lose their mind“ like that.
Lastly, this may be just a nitpick, or maybe not, but game-theoretically optimal poker strategies, while they may take lines that look unnatural, never sacrifice EV.
I think a big differenc between VC and poker is that the economy isn't zero sum. So it's a different mindset towards risk. I wouldn't say the places them outside the river. It's like I read the rules and this casino has set min and max bets wrong, so I am using a strategy that wins me 3% an hour. It looks like I have no tolerance for risk, because it's not a risk just a tactic or program I'm meathodically executing.
A big root cause of apparent irrationality in Silicon Valley is that startup founder utility incentives are not the same as their financial incentives.
A simple model of founder outcomes is, you have a small X% chance you make $Y, which is more money than you need for anything in your lifetime. Financially the EV is proportional to XY. But the expected utility is basically proportional to X. So on the margin do you really want increase Y by 2% and decrease X by 1%? Nah.
This causes things like taking less money to avoid down rounds, or preferring a big-name VC to negotiating harder.
To me, "River mentality" is actually not a great mindset for startup founders. You want to get your head completely in the space of the product you're building. You want to talk to users and make something people want. You don't really want to spend your mental energy estimating probabilities. VCs are fundamentally more "River". But overall Silicon Valley is not entirely a fit for the River sort of thinking.
Isn't it just that SV is a different river, where downside risk is effectively capped and social consensus makes certain rational plays out of bounds? A founder *should* take super-Kelly risks that would wipe out the bankroll in a River situation, because there is more value in having tried and failed in SV than in having backed slowly away from a table offering -EV bets.
It's more like, the actual "investing" part of Silicon Valley isn't that important to the mindset. For investors, sure, but very few people in Silicon Valley are VCs or angel investors. When you're a founder, you spend a very small part of your time doing the negotiating part of raising money. You can be a great founder and be not that great at raising money. And non founder employees (most of the people in Silicon Valley!) spend near zero on it.
Instead, the important part to Silicon Valley mindset is actually building a novel product. You need to think different here. Thinking in the same way as everyone else is a disadvantage. You need to make something people want, you need to talk to users. You need to insulate yourself emotionally from the ups and downs of launching things that people like or don't like. You need to deal with the chaos of starting an organization that is starting from nothing but growing fast and therefore inevitably you have people in new roles who don't know how to do things yet.
The River is this "let's get 1% smarter by constant tuning" mentality. Which is great for playing poker or running a hedge fund when you are constantly dealing with financial exchanges that you need to value. But founders should just not get into that mental space. They should deeply understand their product, they should build something people want.
Peter Thiel seems like a very interesting person, and I'm jealous you got to meet him. I read his book "Zero to One", but I feel like it barely scratches the surface of his thinking and worldview.
great stuff, thanks for writing down your thoughts like this.
one common theme seems to be "animal spirits". people who just want to be out there doing things, they've got joie de vivre, libido, whatever you want to call it -- they seem to be very successful and to not fail as much as you'd expect.
i suppose you can partly appeal to the fact that optimism in the face of uncertainty is a very sound decision making strategy. if you believe the active inference people, it's the only decision making strategy any of us ever use. but this seems like only a partial explanation.
Most River-nature activities discussed have social value that is either clearly negative, like gambling or plausibly so, like quants and hedge funds, which are facially neutral, possibly have some diffuse benefits, but have large negative tail risks.
Silicon Valley venture capital has large positive social value, but it the least River-like of the activities discussed.
Vegas gambling only took off because of a kind of industrial policy imposed by the local Village people.
I’m not sure how I feel about the implications of this confluence of facts! I’m not a 5 on a River version of the Kinsey scale, but I’m not a 1 either. As some point one is reading the DSM and muttering “not *all* psychopaths…”
Just a random comment about the "coming from wealth" thing.
The null hypothesis here is: "tech founders are similarly likely to come from any social class", which seems untrue - they're surely far more likely to come from a high income background. I haven't done the maths, but they're probably also more likely to come from the top 1% or top 0.1% than the null hypothesis suggests: e.g. Reed Hastings (Netflix), SBF (FTX) and Steve Jurvetson (Draper Fisher Jurvetson) are all from fairly elite-level wealth.
All I know is, the Irish ones are shillelagh points...you get the most coordination utility by picking one and sticking something with it, rather than swinging at random.
Shilling for Schelling sounds like a company I'd invest in, too. Or the opposite, for sellers...
> I do think that even if you could have otherwise played it safe and been personally set, you need to be willing to risk ruin if the stakes are high enough, in the sense of being willing to make enemies, or perhaps face political or legal consequences or even potentially violence. But risking financial ruin in particular, through losing money, past some point? You’re simply never getting odds to do that.
I didn't really follow this bit, can you expand on what you mean? Which combinations of monetary vs non-monetary risks vs rewards are you saying make sense vs don't?
Podcast episode for this post:
https://open.substack.com/pub/dwatvpodcast/p/book-review-on-the-edge-the-business
Going to be a super high-volume week, we are up to four and a half hours already!
Fun fact, the voice I've given Nate for these was originally created as "Contract Devil" for Planecrash.
If you're finding value in this AI-multi-voiced podcast version of the post and are already supporting Zvi, please consider chipping in through a Substack subscription. The ElevenLabs MV2 model sounds great (at least to my ears), but it can get pricey at high volumes. (Support Zvi first and foremost though.)
I think private equity has distorted VC — as they go into increasingly risky and earlier stage businesses the only investments left for VC are the ones where they would never consider taking PE money for cultural reasons or the ones where the opportunity is so far out on the probability curve that it breaks PE folks’ brains.
And the long shot math breaks VC brains too. If your expected 2x return was calculated by multiplying 1e9 by 5e-8, I am going to question the robustness of some of the modeling assumptions to put it lightly.
Plus these are the good VCs! Most VC is pure herding and pattern matching where original thought is actively discouraged. If the raison d’etre of your fund is “Sequoia wouldn’t take our LPs’ money” then those same LPs will reward you for looking as much like Sequoia on the outside as possible and don’t understand the adverse selection you face. Complete cargo cult investing.
My other theory is that SV succeeds because it’s the only place in the world where someone who asks for an 8 figure check to fund a company with no existing business isn’t 100% definitely scamming you
Erik Torenberg writes that what is especially important to SV VCs is the social capital returns of being on the ground floor of a major investment (ex. FB, Google, etc.) with the hopes of generating the prestige of knowing the founders before they hit it big (couldn't find the article). However, as Rabois points out you can't be too contrarian otherwise other VCs will be unwilling to invest in future rounds.
SV is contrarian contra the rest of the world, herd-like internally. The believe (with good evidence) it's the crazy ones (Musk, Neumann, SBF) that generate the 100x returns.
I’m not surprised that Bitcoin partisans would hate Ethereum: once you endorse later cryptocurrencies rather than just Bitcoin, the first one, then Bitcoin should be worth zero as the potential supply of cryptocurrency is infinite.
With a statement like „traders didn‘t even know Black Scholes“ you can make Nassim Taleb angry, see https://www.sciencedirect.com/science/article/abs/pii/S0167268110001927.
Yesterday I wrote a comment (first one here for six months) on the gambling instalment only to find, when copying it into Substack, that comments were restricted to paid subscribers, which was unannounced and seems unprecedented (perhaps even a mistake?). Because I‘m so slow at writing (managed only one Substack post in 2024) that even a quite short comment takes me time, I‘m going to be impolite enough to add it here. So the gambling post was another one that I found highly interesting, e.g. the insights about Phil Hellmuth‘s modus operandi, and then not least the parts about sports betting against the house, as all I knew in the area of sports betting was Betfair Exchange (and not even much about that), where players bet against each other.
With poker I‘m more familiar, but before reading the post I don‘t think I really knew about the thorough investigations into Lew‘s other live-streamed play that found nothing suspicious. And still I would have gone higher already than 95 percent on no-cheating. Only the no-cheating story makes any sense, it may have been not a misreading of her hand (though I guess that‘s possible) but simply a spite call, a total blowup, caused over time by Adelstein‘s constant aggressive play; and then she confabulated stories to rescue some reputation of competence (which women at poker tables seem to care a lot about —- even though appearing incompetent is usually +EV). People who maintain that she was cheating misunderstand this because they think, many even correctly, that they themselves could never „lose their mind“ like that.
Lastly, this may be just a nitpick, or maybe not, but game-theoretically optimal poker strategies, while they may take lines that look unnatural, never sacrifice EV.
I think a big differenc between VC and poker is that the economy isn't zero sum. So it's a different mindset towards risk. I wouldn't say the places them outside the river. It's like I read the rules and this casino has set min and max bets wrong, so I am using a strategy that wins me 3% an hour. It looks like I have no tolerance for risk, because it's not a risk just a tactic or program I'm meathodically executing.
A big root cause of apparent irrationality in Silicon Valley is that startup founder utility incentives are not the same as their financial incentives.
A simple model of founder outcomes is, you have a small X% chance you make $Y, which is more money than you need for anything in your lifetime. Financially the EV is proportional to XY. But the expected utility is basically proportional to X. So on the margin do you really want increase Y by 2% and decrease X by 1%? Nah.
This causes things like taking less money to avoid down rounds, or preferring a big-name VC to negotiating harder.
To me, "River mentality" is actually not a great mindset for startup founders. You want to get your head completely in the space of the product you're building. You want to talk to users and make something people want. You don't really want to spend your mental energy estimating probabilities. VCs are fundamentally more "River". But overall Silicon Valley is not entirely a fit for the River sort of thinking.
Isn't it just that SV is a different river, where downside risk is effectively capped and social consensus makes certain rational plays out of bounds? A founder *should* take super-Kelly risks that would wipe out the bankroll in a River situation, because there is more value in having tried and failed in SV than in having backed slowly away from a table offering -EV bets.
It's more like, the actual "investing" part of Silicon Valley isn't that important to the mindset. For investors, sure, but very few people in Silicon Valley are VCs or angel investors. When you're a founder, you spend a very small part of your time doing the negotiating part of raising money. You can be a great founder and be not that great at raising money. And non founder employees (most of the people in Silicon Valley!) spend near zero on it.
Instead, the important part to Silicon Valley mindset is actually building a novel product. You need to think different here. Thinking in the same way as everyone else is a disadvantage. You need to make something people want, you need to talk to users. You need to insulate yourself emotionally from the ups and downs of launching things that people like or don't like. You need to deal with the chaos of starting an organization that is starting from nothing but growing fast and therefore inevitably you have people in new roles who don't know how to do things yet.
The River is this "let's get 1% smarter by constant tuning" mentality. Which is great for playing poker or running a hedge fund when you are constantly dealing with financial exchanges that you need to value. But founders should just not get into that mental space. They should deeply understand their product, they should build something people want.
Peter Thiel seems like a very interesting person, and I'm jealous you got to meet him. I read his book "Zero to One", but I feel like it barely scratches the surface of his thinking and worldview.
great stuff, thanks for writing down your thoughts like this.
one common theme seems to be "animal spirits". people who just want to be out there doing things, they've got joie de vivre, libido, whatever you want to call it -- they seem to be very successful and to not fail as much as you'd expect.
i suppose you can partly appeal to the fact that optimism in the face of uncertainty is a very sound decision making strategy. if you believe the active inference people, it's the only decision making strategy any of us ever use. but this seems like only a partial explanation.
Most River-nature activities discussed have social value that is either clearly negative, like gambling or plausibly so, like quants and hedge funds, which are facially neutral, possibly have some diffuse benefits, but have large negative tail risks.
Silicon Valley venture capital has large positive social value, but it the least River-like of the activities discussed.
Vegas gambling only took off because of a kind of industrial policy imposed by the local Village people.
I’m not sure how I feel about the implications of this confluence of facts! I’m not a 5 on a River version of the Kinsey scale, but I’m not a 1 either. As some point one is reading the DSM and muttering “not *all* psychopaths…”
can't wait for the next one.
What a great series.
Just a random comment about the "coming from wealth" thing.
The null hypothesis here is: "tech founders are similarly likely to come from any social class", which seems untrue - they're surely far more likely to come from a high income background. I haven't done the maths, but they're probably also more likely to come from the top 1% or top 0.1% than the null hypothesis suggests: e.g. Reed Hastings (Netflix), SBF (FTX) and Steve Jurvetson (Draper Fisher Jurvetson) are all from fairly elite-level wealth.
Is a Shilling point a vendable Schelling point? "Getcha equilibrium here, no communication down, all genuine offers beaten!"
All I know is, the Irish ones are shillelagh points...you get the most coordination utility by picking one and sticking something with it, rather than swinging at random.
Shilling for Schelling sounds like a company I'd invest in, too. Or the opposite, for sellers...
> I do think that even if you could have otherwise played it safe and been personally set, you need to be willing to risk ruin if the stakes are high enough, in the sense of being willing to make enemies, or perhaps face political or legal consequences or even potentially violence. But risking financial ruin in particular, through losing money, past some point? You’re simply never getting odds to do that.
I didn't really follow this bit, can you expand on what you mean? Which combinations of monetary vs non-monetary risks vs rewards are you saying make sense vs don't?