> The idea was simple, and also very much something I would have tried. There were a lot of different exchanges trading a lot of cryptos at lots of different prices. Sometimes the prices were different. When that happened, you could do various forms of arbitrage (and statistical arbitrage). Sam, being a trader and also being Sam, was a big fan of taking the free money.
This bot was completely insane to run in 2017. There was barely any way to hedge Bitcoin then, let alone *all shitcoins*.
I think you probably would have been talked out of it in 5 minutes.
Seems right that in the first 5 minutes you would raise concerns I'd have to handle, risk management would be important, but that's also the reason why you get to do the trade in the first place. Seems entirely solvable.
This was actually a mirror of my big crypto mistake:: I interpreted 'you can lose your coins easily and the usability and trading is currently so terrible' as a reason NOT to buy BTC at very low prices. Whereas it should have been a reason to buy.
But that's in hindsight. I could have bought BTC extremely early, for pennies each. It was obviously a scam though - and it totally still is! There's no inherent value. If the US government puts in significant regulations on BTC, the price might plummet with no bottom but zero. It could also plummet if people sell off enough to drop the value and create a run. There are individuals who could cause this to happen on their own.
That it became expensive and made a lot of people a *lot* of money is a fluke. There's a line of thinking that says it's okay to bet on things that cost you little but may become big. That's what VCs do, but that's also the theory behind dozens of garbage coins that came out trying to cash in on BTC. If you have an investment strategy based around cheap investments that may pay off, sure. That's how people have been using penny stocks for a long time. We shouldn't pretend that BTC was ever a good investment, rather than a lucky one.
I strongly disagree with this - in expectation BTC was a screaming buy at least until it got to ~$1000, and yes I did not realize it at the time but I am very confident I had enough information to conclude this. I'm not saying it was >50% to get to 10k or anything, but the alpha was very clearly off the charts.
I don't disagree with you that you could have looked at BTC at various stages and determined that you could reliably make money on it. I disagree with the idea that a poor business model and lack of sustainability was a good reason to buy it. I think you properly judged those as problems, but apparently large numbers of other people did not.
At base, BTC is now and always has been a bubble. People make loads of money off of bubbles, but that doesn't change them to be real value based on people making profits. All BTC profits necessarily come at the expense of someone else, whoever is holding the BTC after you sell it. It's use value is so much lower than the purchase price that having it only works in terms of speculation about future buys from other people.
If you are kicking yourself for not buying at $10 or whatever, sure. But I think your instincts were right and that given the opportunity to do that again, you should lean towards doing the same (prudent) thing - don't buy.
I think even a lot of the garbage coins were not terrible investments if you could stick to selling out a year later (the small handful that took off would more than pay for all the ones that didn't, VC style) - there is a shockingly large supply of Greater Fools.
Your story is "because that's who he is as a person", but I'm kind of still skeptical that there are people like that.
Some other people were speculating that it was because he was on such high doses of stimulants that his judgment was borked. That would make sense, and we do have suggestive evidence from the trial reporting that he's on an unusual amount of Adderall. (Entry-level ADHD dosing is not usually several times a day.)
Patrick McKenzie's story is, as I understand, "because he was a bagman for money launderers; the missing money wasn't lost in bad trades but funneled back to his secret crime clients." That too might make sense. Normally I'd be suspicious about a theory that posits a crime for which there's no direct evidence, but Patio's not a wild-speculation guy in general, and money laundering in crypto isn't an unusual occurrence.
But "he just likes to gamble, lots and lots, to an irrational degree?"
I...guess?
Is it plausible that anybody likes to gamble *this* much?
Zvi is being decorous in not using the p-word, in the same way that, say, Isaacson very carefully avoids using the b-word in the other famous recent biography.
I would be highly unsurprised, as I mentioned, if Patrick's story was correct. I would still believe the other part of the story, as well. No, I do not think it requires drugs. I think there are indeed people who are like this.
(If you are still confused, happy to chat more in private)
I agree that it's odd. It seems like there are some switches missing that even most other people who don't experience regular human emotions have. Maybe it was akin to a gambling addiction, with drugs on top (and maybe not just the prescription ones)?
He found a switch he could turn on that made obscene amounts of money with apparently no risk (to his uncalibrated perspective). I think there are very few people who would not be strongly tempted to hit that switch as often as they can, or find a new switch if the first one stopped working.
Addicted gamblers is an obvious example.
The problem with this whole scenario is that regular finance - trading, investing, whatever - has been through all of the bad ideas and is heavily regulated. The bigger and older players seem to be self-regulated as much as government regulated (with some famous counterexamples), but crypto was a new field that existing systems didn't account for. There was enough money floating around to be serious business, but not enough understanding in conventional banking circles to get a handle on it. A once-in-a-lifetime opportunity for a limited time.
I think SBF was young enough and foolish enough that he didn't see it for what it was, but instead thought he must be extra smart to have figured out what others could not.
My rough hypothesis is that he was/is driven by a combination of obsession with winning, idolisation of his own intellect, and a diminished capacity to feel normal emotions (meaning he compulsively chased the few experiences that gave him some 'reward').
A recurring pattern in his writings or interviews is reasoning on the lines "most people believe not-X, so they don't do Y. but I do believe X, so I will go all-in on Y!" Examples: ignoring the Kelley criterion based on spurious reasoning, funding politicians, the initial Japan/Korea arb, the political trade in Jane Street, etc, etc. Any time there's a potential "edge" caused by the false beliefs of others, he will try to exploit it, *even when it doesn't make sense*.
There are many cases where this reasoning is spurious (or plain idiotic), and SBF's life story contains tons of them. Maybe other factors make Y a bad idea. Maybe there are unforeseen risks with Y. Maybe the total costs of Y are hard to quantify. Etc.
He is still using this reasoning while awaiting trial: "my lawyers tell me not to talk to the press, but they think from a narrow legalistic perspective -- so I'll talk to all the media!!" It doesn't actually make sense, doesn't accomplish anything, doesn't fit with any kind of coherent strategy -- but it flips the conventional wisdom, so why not go all-in on it?
His eschewal of the Kelley criterion is an illustrative example. In your article (https://sarahconstantin.substack.com/p/why-infinite-coin-flipping-is-bad) you point out that SBF's reasoning was faulty. His rationale was something like "Kelley assumes diminishing marginal utility of money, but I have linear utility, so I can ignore Kelley!!!" You pointed out that Kelley still holds even if you have linear utility. But I think there was a psychological component at play, too.
Where most traders see things like Kelley as useful guides to action, SBF sees them as arbitrary restrictions that he wants to hack. On hearing "you can't be a perfect expected-utility maximiser, because of assumptions A1, A2, A3", most sane people would think "do those assumptions hold for me? let me check carefully". SBF thinks "screw that, I can prove those assumptions don't apply to me!! and therefore I win!!!"
This is how he ended up spouting nonsense such as saying that he'd repeatedly toss a coin with a 51% of creating a second earth and a 49% chance of destroying humanity. Most EAs would take this as a reductio ad absurdum, and check their assumptions, but SBF simply wants to prove some weird point (that he's the perfectly rational utility maximiser, or whatever). In practice he wasn't remotely rational, he was a high-IQ idiot who rotted his brain with stimulants, gambling and dubious theories.
I was raised with parents who had a strong value system. They taught me that the most joy to be found in life was in service to a bigger goal, especially in helping others. That turns out to be deeply true, so when I make decisions I think about more than just myself and my goals. And I'm deeply content in my normal suburban life with my wife and my beautiful daughter, and don't want to endanger them.
But if that value system wasn't there? Hell yes I would borrow money from others, lie, cheat, steal, whatever, just for the thrill of maybe number get twice as big. The thought of a 51% double-or-nothing bet is exhilarating to my particular limbic system in a way very very few things are. Some folks are wired that way. Plenty of folks lose their whole lives to gambling addiction. Intelligence might impact how ruinous this impulse is (because smart people make better bets), or how many resources they're able to accrue, but I mostly see intelligence as orthogonal to how much someone will gamble.
And a gambling addict with sociopathy will gamble everything they can get their hands on.
I’m no altruist, and I don’t live a life of service, but I never even used a credit card until my husband talked me into it; the idea of losing money (or taking out loans I might not be able to pay back) is so creepy to me that it boggles my mind that people actually do things like margin trading, let alone SBF stuff. I guess there’s a whole dark matter universe of people who like to gamble out there.
I would build a model framing for thinking about this:
Different sets of people react differently to the same amounts of stimulus. This implies some people need a low amount of stimulus to be satisfied (people who like quiet cafes and calm music) and others need a higher amount of stimulus to be satisfied (parties, loud bars, sports games).
Different people have different risk tolerances. This is probably somewhat intertwined with the above, but doesn’t have to be for our model, so let’s separate it. Low risk tolerance people hate losing money, and hate paying interest. High risk people don’t mind losing money, as long as there is sufficient probable upside over time.
Sam comes across as a “high stimulus, high risk” sort of person, where he doesn’t mind gambling if the payoffs line up, and has an exceptionally high stimulus need (see the playing LoL constantly, always distracting himself with things). Other high risk high stimulus people would include people involved in most crazy adventure sports like free soloing, paragliding, base jumping, wingsuit skydiving, etc.
But it also turns out gambling and betting is an extremely rewarding stimulus, especially when you win. The high from winning is intoxicating enough that the signal can erase / blunt the feeling of losing, especially if you are predispositioned to not minding losing. So high risk / high stimulus people are more wired to chase the from the high of winning and don’t feel the pain of losing as much. And the stimulus from the actual betting process and the stakes at risk is hard to match with basically anything else (except for high adrenaline adventure sports as mentioned above).
One thing this comment points out that didn't occur to me in my original comment: SBF literally never lost, at least not in a way he'd recognize as losing. So while I think intellectually he thought of his risks as "positive EV value bets" or whatever, emotionally it probably felt the same as a night of craps where you win on every shoot.
And it's equally hard for me to imagine a person who can't even imagine a person who enjoys gambling. Vegas makes a lot of money.
I think Blackjack (appropriate username) has it right - different people have different dopamagenic needs, and different risk tolerances, and when high dopamagenic need meets high risk tolerance there's a danger of gambling addiction.
Psychologically I have a need for a lot of stimulation, but I have a low risk tolerance because I'm essentially content with a lot to lose. I'm stuck playing $5 buy-in internet poker for my fix. :P
Of course I don't steal or defraud people out of money to gamble, which would greatly increase the amount of risk I could take. But if I were a sociopath or narcissist, and didn't care about others, or was able to rationalize everything I did as "for the greater good" simply because I did it, defrauding people out of money to gamble would be a reasonable M.O.
Where does Patrick McKenzie suggest that? In the tweet linked in this post, he suggests that Alameda's alpha was from laundering / legal risk. This seems like a pretty reasonable hypothesis, but it's about Alameda, not FTX. If this were true about Alameda, what would it tell us about FTX? I don't see the connection to stealing money. Why would he steal money to give to other people? Did they blackmail him to create FTX? Why not just not? Why not come clean now?
It's such a red flag to start a startup and have the founder retain all of the stock. Either you only hire inexperienced people, or you unethically mislead them. Tricks like, the offer letter only describes the number of shares, and it turns out theres 1000x more shares outstanding than they verbally told you. Either way is setting your company up for disaster.
One lesson here that I wish more people would take is, don't join a startup where the founder is being secretive about how much stock you're getting. Every time a friend of mine has been in this situation, a year or two down the road they discover that 1. they were only given a tiny amount of stock, not worth anything near as much as the founders hinted, and 2. the general quality of people at the company is really low.
Obviously in a tech startup the stock might be worth nothing. But you should know what % of the company you own (with some flex for things like size of option pool, amount of future dilution, etc).
Yep. I actually did have one case where I was strung along for a while and then later treated right, but the cause of the stringing killed us in a different way anyway. So yeah.
For ModelBot, I wonder if the trade was some kind of market manipulation (e.g. pump and dump or similar) rather than arbitrage. That would explain the difficulty of backtesting it and also of running it with a smaller stake.
I wonder if the decision to name it "ModelBot" suggests that there was something more than simple arbitrage going on. "Model" is not the way I'd have described arbitrage.
> In elementary school he’d read the Harry Potter books over and over. By the eighth grade he had stopped reading books altogether. “You start to associate it with a negative feeling, and you stop liking it,” he said. (505)
I found this amusing due to doing a similar thing myself. Harry Potter grabbed me as an early understandable long-form book with lots of content to chew through. Then I reread it multiple times over the next years.
Picking out books to read became harder and harder. I did end up associating going to the school library with a negative feeling, of having to use cover art and limited statements to discern what a book might be about and whether it would be any good. I was picky, and in reality none of my tastes were being properly appealed to (Harry Potter was just familiar and dense enough along some dimensions). I read other fiction, but at times I just gave up and picked something I knew I liked, and once you reread the first Harry Potter book it is far easier to continue onto the next. That would then make the next week or two of reading during class free-time a breeze, though a predictable one.
Also what's wrong with his commentary on Shakespeare? He is ignoring the possibilities of 'being in a time with lots of pressures just makes you a better writer' and 'his writing is perhaps just in fact that good, update on that'.
Also, people do definitely ask the question of 'who the best writer is'. Though I think the quote is more trying to gesture at the (classic?) point that there's at least a hundred writers as good as Shakespeare or better, just that he gains a lot of weight by by being early and taught in all the classes.
I think the primary issue with his Shakespeare quote is that it's completely disengaged with the subject matter/reality. As a method of art-criticism it's essentially equivalent to saying "discard all art before the industrial revolution on a bayesian basis". His comment judges the art without any reference to the enjoyment from viewing it, the influence on society, the addition of words/phrases to English and simply says "well he played against a weak field so it's probably not very good".
Putting it another way: Do we discard the art of Michelangelo (dead about a month before Shakespeare was born) b/c the Bayesian priors on sculptors, painters and inventors are so weak?
Being even less kind to SBF: If I ask someone to judge a book, they ask when it was published and tell me it's crap based on the year of publication, what I've learned is that person is terrible at modeling art judgements (and possibly terrible at knowing emotions exists).
I think of it more as a 'you should be skeptical that it is really as good as all that, and start thinking about alternate hypotheses like self-feeding fame and an aura of historical mystery'. I would say that much art does actually fall into this category. Various household name artists are legitimately good! There's pieces that are centuries old which I think capture a certain feel better than anything else I've seen. I just also think there's simply better artists (so if for quality, focus on them), artists with more information on how they painted (so if you're trying for inspiration), etc.
(The classic response is that art culture is about signaling, which is in part true)
There's certainly reasons to read it purely for effect on society. However, for a lot of English classes (for example) it would potentially be better to focus on some other short stories - unless the English class is specifically focusing on important historical aspects (but many of those simply don't). I agree that completely throwing it out is the wrong move.
> Being even less kind to SBF: If I ask someone to judge a book, they ask when it was published and tell me it's crap based on the year of publication, what I've learned is that person is terrible at modeling art judgements (and possibly terrible at knowing emotions exists).
I'd come in with a high prior that the book will be worse along various axes that I enjoy. There are certainly books from hundreds of years ago that I've enjoyed, like I enjoyed Faust as the first that comes to mind, but they often had focuses and writing styles that weren't as evocative as the best writers I've seen are. Also they tended to not include story elements which I'm not as interested in, which isn't a mark against them since a lot of pieces I like simply were nonexistent or minor.
These are also from the ones selected for lasting the ages. The real issue I'd say with the quote is ignoring the selection by people finding Shakespeare/Michelangelo/etc's works good, which gives it significantly more bits than random writer/artist from that time period.
As well, the quote isn't saying that the book is necessarily awful, it is saying that you should roll to disbelieve that he's among the greatest authors.
The murder part sounded to me like it was gesturing at the idea that calling something murder isn't significant. There isn't some label hidden on the act that says 'murder' and thus you should get more worked up about it. That the problems with murder come from the effects, not whether you term it murder or not, ranging from the person dying to a person being willing to kill others.
I'm kinda mostly only commenting this because I think (though I'm operating without the surrounding context of the rest of the book here) the way the post takes these quotations more aggressively because of what he became, rather than them being something that would be too objectionable if they were in some random post on LW.
I am fully aware of what I did with the framing here, and yes I do see the real and important point that is part of it, but one should not ignore the Yikes, either. And it echoes so many common themes throughout.
What he said is also bog-standard Academic Moral Philosophy about why murder is bad. If you ask a real philosophy professor what's bad about murder, you'll get pretty much the same speech SBF made.
Yeah, in an earlier time, I read The Lord of the Rings six or seven times because (1) my parents' preferred punishment was to send me to my room, and LotR was on my shelf,* and (2) I was bored a lot of the time, we didn't get a computer until I was 15 or 16, and re-reading LotR was likely to be more enjoyable than a random book off my parents' shelves. (Although I did find some winners - thanks, mom and dad!)
* In hindsight, maybe they "punished" me more than I recall? To be fair, they would send me to my room for an hour and I would start reading, then stay in there for four.
Being a complete layman financially, my guess for the reasoning to not activate ModelBot is that it would have potentially leaked the details of how the bot worked to competitors, based on transaction histories. If it barely works, someone has the opportunity to refine it before you (you have one 100x programmer, they have 101 programmers) or simply copy you. No chance to test it in public without a loss, even at small volumes. Otherwise I agree, it makes no sense not to set expense limits on the bot and calll it a day.
As I understand it the arb was across exchanges, which means you wouldn't reveal much that was useful other than you were taking good prices, and everyone knew such arbs were possible. I doubt this was a concern.
And if leaking information is for some reason a concern, just build a simulator for the exchange APIs and try it in a sandbox. There are lots of ways to build confidence short of running something at 100% in production without supervision.
Minor nitpick: do you *actually* believe that electron suffering is something EAs care about or was that a tongue-in-cheek remark to prove your point? Because there is a forum post on it, but it is an April Fools joke.
Almost all EAs are not so foolish. But choosing a "more real" example that to me was almost as absurd - of which there are several that I am very confident are real - seemed like a counterproductive distraction here.
Thank you for writing this. More than anything else recent (recent enough you can send to a normie and tell them to read it and they don't got "but it's from 2 years ago!") it's a fantastic argument in favor of "maybe people should actually do some thinking and thinking about thinking".
I would add, to the unanswered questions section: How is it impossible for someone so good at games/puzzles to be in Bronze in League of Legends? This is like hearing someone is a competitive-at-world-champs level chess player, but really struggles with tic-tac-toe and checkers. There's probably some good SBF mentality questions to mine here.
My guess is SBF was in the habit of being massively distracted and thus losing matches? In Brawl you often get 30-60 second breaks and can somewhat move them around, in LoL not so much? Or the micro--optimizations the game is all about simply bored him, so he never bothered doing them right, and that kept him at bronze.
Or perhaps he was intentionally losing to maintain lower ranking.
I do find 'why did he keep playing that game instead of switching to another' to be a good question. But I suspect that being punished for distraction was a bug rather than a feature - the book describes the constant pressure to act as something he loved.
That's a fair point. In league, if you stop paying attention for 60 seconds coming back and winning becomes extremely hard (when you combine demoralization effects to your team plus actual game state changes).
EDIT: on further reflection him being in Bronze makes a lot of sense given the information available. He joins a game expecting to play, his EV/attention calculus shifts and he afk's for 30-60 seconds and his team has a conniption while falling behind. When a more normal person rejoins they are put off by the external negative effects and chat consisting people telling them to jump off a bridge (hypothetically, in a video game). SBF discounts this to 0 (self assessing that he had to act has his EV fluctuated wildly by the moment). But, he also gets to enjoy a lot of fun gameplay as bronze people are terrible so as long as his mechanics are remotely decent he can probably personally outplay them and still have fun (even if he loses b/c his team is playing terribly after tilting off the face of the earth).
High intelligence is helpful but neither necessary nor sufficient to be good at League.
This is not the kind of game that you can completely solve intellectually from the first principles; there are too many moving parts—it's practically intractable for a human mind, even if you do remember all the mechanics and interactions (that's a huge if, btw). So you have to make educated guesses. And while intelligent people are usually good at that, it also requires a lot of domain knowledge.
And then you also have to make these decisions fast, because the game is not waiting for you. Which you can't without good reflexes and/or good foresight. You need a lot of experience for both, and intelligence only helps you so much here (you can learn more efficiently if you are smart, but there are limits, and it's also not automatic; you have to actually make decisions to train specific skills to take advantage of that). I know quite a few not-all-that-intelligent folks that are quite good at making split-second decisions in league simply because they've seen most standard situations in the game enough times for their brains to successfully pattern match the situation with optimal response. You simply can't do that with intelligence alone; you will be too slow.
Now I don't know if he had the necessary prerequisites for developing the kind of skills required to excel at League or how much he played, but frankly, it does not matter.
There are many things necessary to be successful in League, and the right mindset is a big part of it.
You cannot have the kind of views and values that SBF apparently had and be good at League. It is a game where your bad bets bite you in the ass really reliably and really fast. If you keep making high-risk calls, that is guaranteed to cause you to lose, often. And you can't lose 100 games, but then win 1 and come out positive like he could in crypto.
"If you keep making high-risk calls, that is guaranteed to cause you to lose, often. And you can't lose 100 games, but then win 1 and come out positive like he could in crypto."
He was playing while in meetings and stuff, right? Where his attention was (intentionally?) split? I don't get the impression that he was playing to win, so much as playing to keep himself from getting bored.
I was also, to some degree, friends with Sam when I lived in NYC. At least on the level of occasionally talking. We also talked about him leaving Jane street. I wish I wrote down what exactly he said. I don't remember him telling me he was going to run or help run anything. He just told me he was going to take a job. This seemed kind of dumb but 'take a job at an official EA org regardless of other opportunities' seemed like the kind of thing EAs do all the time. I just figured Sam's EA job quickly didn't work out for some reason. I'm still not sure if he was blatantly lying or the EA job just fell through. But I don't have any notes. Definitely feels like a bit mysterious to me.
It's more gonzo than gonzo. I knew the outlines of the story already, but the various cognitive train wrecks kept me glued to the entire review. And not just SBF. The entire cast is extremely smart in some ways yet makes dumb decisions that could be trivially avoided, throughout. And these people are all rationalist-adjacent at least. "The first principle is not to fool yourself – and you are the easiest person to fool."
I saw a trial quote that the thing where they bribed chinese officials wasn't in the charges. Why wasn't it? Seems like a blatant crime for which there's a lot of evidence, no?
Yes, it is a direct violation of the Corrupt Foreign Practices Act, and everyone involved is super guilty. As I said, of all the crime. What, they should charge all the crimes no matter how small?
Regardless of whether you want to call it a 'hack' the MOBL exploit is at minimum commodities/securities fraud. Its roughly the same exploit that landed Avraham (another rationalist) in jail and will almost certainly cause him to serve prison time. This sort of exploit is absolutely not ok even if some people's intuitions differ.
the ethics dont seem very complicated to me! I'm all for ignoring unjust laws, but your argument for doing so needs to be a lot better than 'enriching yourself'. Im happy to even accept personal autonomy as a reason for doing things like using illegal drugs. But wanting millions of dollars, which you will be moving from other people to yourself, is an extreme weak justification.
Appreciate this. What a fantastic review! I also read it and agree that it is an awesome book.
But I never believed Lewis was earnest. He was SBF's guest and Lewis surely knows that the crimes he described alone are enough to convict SBF. Emphasizing this in is unnecessary and imo turns the book into something less interesting -- a polemic on SBF. By writing the book in this way he:
1. Ensures future subjects will be happy to let him tag along and speak candidly with him
2. Keeps the focus on the big picture that you end with (although he never specifically states it kind of like the Alchemist)
3. Produces what's probably the first widely-read description of how powerful EA is becoming.
I have to wonder if Lewis is touting SBF's innocence in the hopes of selling more copies of the book to people wondering what the proof of his innocence is.
> The idea was simple, and also very much something I would have tried. There were a lot of different exchanges trading a lot of cryptos at lots of different prices. Sometimes the prices were different. When that happened, you could do various forms of arbitrage (and statistical arbitrage). Sam, being a trader and also being Sam, was a big fan of taking the free money.
This bot was completely insane to run in 2017. There was barely any way to hedge Bitcoin then, let alone *all shitcoins*.
I think you probably would have been talked out of it in 5 minutes.
Seems right that in the first 5 minutes you would raise concerns I'd have to handle, risk management would be important, but that's also the reason why you get to do the trade in the first place. Seems entirely solvable.
This was actually a mirror of my big crypto mistake:: I interpreted 'you can lose your coins easily and the usability and trading is currently so terrible' as a reason NOT to buy BTC at very low prices. Whereas it should have been a reason to buy.
But that's in hindsight. I could have bought BTC extremely early, for pennies each. It was obviously a scam though - and it totally still is! There's no inherent value. If the US government puts in significant regulations on BTC, the price might plummet with no bottom but zero. It could also plummet if people sell off enough to drop the value and create a run. There are individuals who could cause this to happen on their own.
That it became expensive and made a lot of people a *lot* of money is a fluke. There's a line of thinking that says it's okay to bet on things that cost you little but may become big. That's what VCs do, but that's also the theory behind dozens of garbage coins that came out trying to cash in on BTC. If you have an investment strategy based around cheap investments that may pay off, sure. That's how people have been using penny stocks for a long time. We shouldn't pretend that BTC was ever a good investment, rather than a lucky one.
I strongly disagree with this - in expectation BTC was a screaming buy at least until it got to ~$1000, and yes I did not realize it at the time but I am very confident I had enough information to conclude this. I'm not saying it was >50% to get to 10k or anything, but the alpha was very clearly off the charts.
I don't disagree with you that you could have looked at BTC at various stages and determined that you could reliably make money on it. I disagree with the idea that a poor business model and lack of sustainability was a good reason to buy it. I think you properly judged those as problems, but apparently large numbers of other people did not.
At base, BTC is now and always has been a bubble. People make loads of money off of bubbles, but that doesn't change them to be real value based on people making profits. All BTC profits necessarily come at the expense of someone else, whoever is holding the BTC after you sell it. It's use value is so much lower than the purchase price that having it only works in terms of speculation about future buys from other people.
If you are kicking yourself for not buying at $10 or whatever, sure. But I think your instincts were right and that given the opportunity to do that again, you should lean towards doing the same (prudent) thing - don't buy.
I think even a lot of the garbage coins were not terrible investments if you could stick to selling out a year later (the small handful that took off would more than pay for all the ones that didn't, VC style) - there is a shockingly large supply of Greater Fools.
I am still confused.
Why was SBF betting so aggressively?
Your story is "because that's who he is as a person", but I'm kind of still skeptical that there are people like that.
Some other people were speculating that it was because he was on such high doses of stimulants that his judgment was borked. That would make sense, and we do have suggestive evidence from the trial reporting that he's on an unusual amount of Adderall. (Entry-level ADHD dosing is not usually several times a day.)
Patrick McKenzie's story is, as I understand, "because he was a bagman for money launderers; the missing money wasn't lost in bad trades but funneled back to his secret crime clients." That too might make sense. Normally I'd be suspicious about a theory that posits a crime for which there's no direct evidence, but Patio's not a wild-speculation guy in general, and money laundering in crypto isn't an unusual occurrence.
But "he just likes to gamble, lots and lots, to an irrational degree?"
I...guess?
Is it plausible that anybody likes to gamble *this* much?
Zvi is being decorous in not using the p-word, in the same way that, say, Isaacson very carefully avoids using the b-word in the other famous recent biography.
Can some kind soul please tell me what either of these words is (ideally both)
Psychopath?
See section 3: https://www.astralcodexten.com/p/highlights-from-the-comments-on-elon
Rhymes with "shmipolar".
I would be highly unsurprised, as I mentioned, if Patrick's story was correct. I would still believe the other part of the story, as well. No, I do not think it requires drugs. I think there are indeed people who are like this.
(If you are still confused, happy to chat more in private)
I agree that it's odd. It seems like there are some switches missing that even most other people who don't experience regular human emotions have. Maybe it was akin to a gambling addiction, with drugs on top (and maybe not just the prescription ones)?
He found a switch he could turn on that made obscene amounts of money with apparently no risk (to his uncalibrated perspective). I think there are very few people who would not be strongly tempted to hit that switch as often as they can, or find a new switch if the first one stopped working.
Addicted gamblers is an obvious example.
The problem with this whole scenario is that regular finance - trading, investing, whatever - has been through all of the bad ideas and is heavily regulated. The bigger and older players seem to be self-regulated as much as government regulated (with some famous counterexamples), but crypto was a new field that existing systems didn't account for. There was enough money floating around to be serious business, but not enough understanding in conventional banking circles to get a handle on it. A once-in-a-lifetime opportunity for a limited time.
I think SBF was young enough and foolish enough that he didn't see it for what it was, but instead thought he must be extra smart to have figured out what others could not.
My rough hypothesis is that he was/is driven by a combination of obsession with winning, idolisation of his own intellect, and a diminished capacity to feel normal emotions (meaning he compulsively chased the few experiences that gave him some 'reward').
A recurring pattern in his writings or interviews is reasoning on the lines "most people believe not-X, so they don't do Y. but I do believe X, so I will go all-in on Y!" Examples: ignoring the Kelley criterion based on spurious reasoning, funding politicians, the initial Japan/Korea arb, the political trade in Jane Street, etc, etc. Any time there's a potential "edge" caused by the false beliefs of others, he will try to exploit it, *even when it doesn't make sense*.
There are many cases where this reasoning is spurious (or plain idiotic), and SBF's life story contains tons of them. Maybe other factors make Y a bad idea. Maybe there are unforeseen risks with Y. Maybe the total costs of Y are hard to quantify. Etc.
He is still using this reasoning while awaiting trial: "my lawyers tell me not to talk to the press, but they think from a narrow legalistic perspective -- so I'll talk to all the media!!" It doesn't actually make sense, doesn't accomplish anything, doesn't fit with any kind of coherent strategy -- but it flips the conventional wisdom, so why not go all-in on it?
His eschewal of the Kelley criterion is an illustrative example. In your article (https://sarahconstantin.substack.com/p/why-infinite-coin-flipping-is-bad) you point out that SBF's reasoning was faulty. His rationale was something like "Kelley assumes diminishing marginal utility of money, but I have linear utility, so I can ignore Kelley!!!" You pointed out that Kelley still holds even if you have linear utility. But I think there was a psychological component at play, too.
Where most traders see things like Kelley as useful guides to action, SBF sees them as arbitrary restrictions that he wants to hack. On hearing "you can't be a perfect expected-utility maximiser, because of assumptions A1, A2, A3", most sane people would think "do those assumptions hold for me? let me check carefully". SBF thinks "screw that, I can prove those assumptions don't apply to me!! and therefore I win!!!"
This is how he ended up spouting nonsense such as saying that he'd repeatedly toss a coin with a 51% of creating a second earth and a 49% chance of destroying humanity. Most EAs would take this as a reductio ad absurdum, and check their assumptions, but SBF simply wants to prove some weird point (that he's the perfectly rational utility maximiser, or whatever). In practice he wasn't remotely rational, he was a high-IQ idiot who rotted his brain with stimulants, gambling and dubious theories.
Best short summary I’ve read!
I was raised with parents who had a strong value system. They taught me that the most joy to be found in life was in service to a bigger goal, especially in helping others. That turns out to be deeply true, so when I make decisions I think about more than just myself and my goals. And I'm deeply content in my normal suburban life with my wife and my beautiful daughter, and don't want to endanger them.
But if that value system wasn't there? Hell yes I would borrow money from others, lie, cheat, steal, whatever, just for the thrill of maybe number get twice as big. The thought of a 51% double-or-nothing bet is exhilarating to my particular limbic system in a way very very few things are. Some folks are wired that way. Plenty of folks lose their whole lives to gambling addiction. Intelligence might impact how ruinous this impulse is (because smart people make better bets), or how many resources they're able to accrue, but I mostly see intelligence as orthogonal to how much someone will gamble.
And a gambling addict with sociopathy will gamble everything they can get their hands on.
Huh interesting!
I’m no altruist, and I don’t live a life of service, but I never even used a credit card until my husband talked me into it; the idea of losing money (or taking out loans I might not be able to pay back) is so creepy to me that it boggles my mind that people actually do things like margin trading, let alone SBF stuff. I guess there’s a whole dark matter universe of people who like to gamble out there.
I would build a model framing for thinking about this:
Different sets of people react differently to the same amounts of stimulus. This implies some people need a low amount of stimulus to be satisfied (people who like quiet cafes and calm music) and others need a higher amount of stimulus to be satisfied (parties, loud bars, sports games).
Different people have different risk tolerances. This is probably somewhat intertwined with the above, but doesn’t have to be for our model, so let’s separate it. Low risk tolerance people hate losing money, and hate paying interest. High risk people don’t mind losing money, as long as there is sufficient probable upside over time.
Sam comes across as a “high stimulus, high risk” sort of person, where he doesn’t mind gambling if the payoffs line up, and has an exceptionally high stimulus need (see the playing LoL constantly, always distracting himself with things). Other high risk high stimulus people would include people involved in most crazy adventure sports like free soloing, paragliding, base jumping, wingsuit skydiving, etc.
But it also turns out gambling and betting is an extremely rewarding stimulus, especially when you win. The high from winning is intoxicating enough that the signal can erase / blunt the feeling of losing, especially if you are predispositioned to not minding losing. So high risk / high stimulus people are more wired to chase the from the high of winning and don’t feel the pain of losing as much. And the stimulus from the actual betting process and the stakes at risk is hard to match with basically anything else (except for high adrenaline adventure sports as mentioned above).
One thing this comment points out that didn't occur to me in my original comment: SBF literally never lost, at least not in a way he'd recognize as losing. So while I think intellectually he thought of his risks as "positive EV value bets" or whatever, emotionally it probably felt the same as a night of craps where you win on every shoot.
And it's equally hard for me to imagine a person who can't even imagine a person who enjoys gambling. Vegas makes a lot of money.
I think Blackjack (appropriate username) has it right - different people have different dopamagenic needs, and different risk tolerances, and when high dopamagenic need meets high risk tolerance there's a danger of gambling addiction.
Psychologically I have a need for a lot of stimulation, but I have a low risk tolerance because I'm essentially content with a lot to lose. I'm stuck playing $5 buy-in internet poker for my fix. :P
Of course I don't steal or defraud people out of money to gamble, which would greatly increase the amount of risk I could take. But if I were a sociopath or narcissist, and didn't care about others, or was able to rationalize everything I did as "for the greater good" simply because I did it, defrauding people out of money to gamble would be a reasonable M.O.
Where does Patrick McKenzie suggest that? In the tweet linked in this post, he suggests that Alameda's alpha was from laundering / legal risk. This seems like a pretty reasonable hypothesis, but it's about Alameda, not FTX. If this were true about Alameda, what would it tell us about FTX? I don't see the connection to stealing money. Why would he steal money to give to other people? Did they blackmail him to create FTX? Why not just not? Why not come clean now?
It's such a red flag to start a startup and have the founder retain all of the stock. Either you only hire inexperienced people, or you unethically mislead them. Tricks like, the offer letter only describes the number of shares, and it turns out theres 1000x more shares outstanding than they verbally told you. Either way is setting your company up for disaster.
One lesson here that I wish more people would take is, don't join a startup where the founder is being secretive about how much stock you're getting. Every time a friend of mine has been in this situation, a year or two down the road they discover that 1. they were only given a tiny amount of stock, not worth anything near as much as the founders hinted, and 2. the general quality of people at the company is really low.
Obviously in a tech startup the stock might be worth nothing. But you should know what % of the company you own (with some flex for things like size of option pool, amount of future dilution, etc).
Yep. I actually did have one case where I was strung along for a while and then later treated right, but the cause of the stringing killed us in a different way anyway. So yeah.
For ModelBot, I wonder if the trade was some kind of market manipulation (e.g. pump and dump or similar) rather than arbitrage. That would explain the difficulty of backtesting it and also of running it with a smaller stake.
I wonder if the decision to name it "ModelBot" suggests that there was something more than simple arbitrage going on. "Model" is not the way I'd have described arbitrage.
> In elementary school he’d read the Harry Potter books over and over. By the eighth grade he had stopped reading books altogether. “You start to associate it with a negative feeling, and you stop liking it,” he said. (505)
I found this amusing due to doing a similar thing myself. Harry Potter grabbed me as an early understandable long-form book with lots of content to chew through. Then I reread it multiple times over the next years.
Picking out books to read became harder and harder. I did end up associating going to the school library with a negative feeling, of having to use cover art and limited statements to discern what a book might be about and whether it would be any good. I was picky, and in reality none of my tastes were being properly appealed to (Harry Potter was just familiar and dense enough along some dimensions). I read other fiction, but at times I just gave up and picked something I knew I liked, and once you reread the first Harry Potter book it is far easier to continue onto the next. That would then make the next week or two of reading during class free-time a breeze, though a predictable one.
Also what's wrong with his commentary on Shakespeare? He is ignoring the possibilities of 'being in a time with lots of pressures just makes you a better writer' and 'his writing is perhaps just in fact that good, update on that'.
Also, people do definitely ask the question of 'who the best writer is'. Though I think the quote is more trying to gesture at the (classic?) point that there's at least a hundred writers as good as Shakespeare or better, just that he gains a lot of weight by by being early and taught in all the classes.
I think the primary issue with his Shakespeare quote is that it's completely disengaged with the subject matter/reality. As a method of art-criticism it's essentially equivalent to saying "discard all art before the industrial revolution on a bayesian basis". His comment judges the art without any reference to the enjoyment from viewing it, the influence on society, the addition of words/phrases to English and simply says "well he played against a weak field so it's probably not very good".
Putting it another way: Do we discard the art of Michelangelo (dead about a month before Shakespeare was born) b/c the Bayesian priors on sculptors, painters and inventors are so weak?
Being even less kind to SBF: If I ask someone to judge a book, they ask when it was published and tell me it's crap based on the year of publication, what I've learned is that person is terrible at modeling art judgements (and possibly terrible at knowing emotions exists).
It seems analogous to many of his decisions, where he comes up with a novel idea, decides he is done, and discards all other relevant considerations.
Which is actually a reasonable thing to do as an academic, and he comes from an academic family.
In what world is it reasonable for an academic to ignore prior work, and still expect to get published in a peer reviewed venue?
I think of it more as a 'you should be skeptical that it is really as good as all that, and start thinking about alternate hypotheses like self-feeding fame and an aura of historical mystery'. I would say that much art does actually fall into this category. Various household name artists are legitimately good! There's pieces that are centuries old which I think capture a certain feel better than anything else I've seen. I just also think there's simply better artists (so if for quality, focus on them), artists with more information on how they painted (so if you're trying for inspiration), etc.
(The classic response is that art culture is about signaling, which is in part true)
There's certainly reasons to read it purely for effect on society. However, for a lot of English classes (for example) it would potentially be better to focus on some other short stories - unless the English class is specifically focusing on important historical aspects (but many of those simply don't). I agree that completely throwing it out is the wrong move.
> Being even less kind to SBF: If I ask someone to judge a book, they ask when it was published and tell me it's crap based on the year of publication, what I've learned is that person is terrible at modeling art judgements (and possibly terrible at knowing emotions exists).
I'd come in with a high prior that the book will be worse along various axes that I enjoy. There are certainly books from hundreds of years ago that I've enjoyed, like I enjoyed Faust as the first that comes to mind, but they often had focuses and writing styles that weren't as evocative as the best writers I've seen are. Also they tended to not include story elements which I'm not as interested in, which isn't a mark against them since a lot of pieces I like simply were nonexistent or minor.
These are also from the ones selected for lasting the ages. The real issue I'd say with the quote is ignoring the selection by people finding Shakespeare/Michelangelo/etc's works good, which gives it significantly more bits than random writer/artist from that time period.
As well, the quote isn't saying that the book is necessarily awful, it is saying that you should roll to disbelieve that he's among the greatest authors.
As Taleb argued, Shakespeare has survived time, which is much more than you can say for modern books.
In other words SBF does not understand all the variables.
...or have a soul, or much of any kind of affect at all, which is pretty key to forming a judgement on art...?
The murder part sounded to me like it was gesturing at the idea that calling something murder isn't significant. There isn't some label hidden on the act that says 'murder' and thus you should get more worked up about it. That the problems with murder come from the effects, not whether you term it murder or not, ranging from the person dying to a person being willing to kill others.
I'm kinda mostly only commenting this because I think (though I'm operating without the surrounding context of the rest of the book here) the way the post takes these quotations more aggressively because of what he became, rather than them being something that would be too objectionable if they were in some random post on LW.
I am fully aware of what I did with the framing here, and yes I do see the real and important point that is part of it, but one should not ignore the Yikes, either. And it echoes so many common themes throughout.
What he said is also bog-standard Academic Moral Philosophy about why murder is bad. If you ask a real philosophy professor what's bad about murder, you'll get pretty much the same speech SBF made.
Yeah, in an earlier time, I read The Lord of the Rings six or seven times because (1) my parents' preferred punishment was to send me to my room, and LotR was on my shelf,* and (2) I was bored a lot of the time, we didn't get a computer until I was 15 or 16, and re-reading LotR was likely to be more enjoyable than a random book off my parents' shelves. (Although I did find some winners - thanks, mom and dad!)
* In hindsight, maybe they "punished" me more than I recall? To be fair, they would send me to my room for an hour and I would start reading, then stay in there for four.
Being a complete layman financially, my guess for the reasoning to not activate ModelBot is that it would have potentially leaked the details of how the bot worked to competitors, based on transaction histories. If it barely works, someone has the opportunity to refine it before you (you have one 100x programmer, they have 101 programmers) or simply copy you. No chance to test it in public without a loss, even at small volumes. Otherwise I agree, it makes no sense not to set expense limits on the bot and calll it a day.
As I understand it the arb was across exchanges, which means you wouldn't reveal much that was useful other than you were taking good prices, and everyone knew such arbs were possible. I doubt this was a concern.
And if leaking information is for some reason a concern, just build a simulator for the exchange APIs and try it in a sandbox. There are lots of ways to build confidence short of running something at 100% in production without supervision.
Have you read Number Go Up? If so, how does it compare?
I started it but it didn't grab me the same way. Not sure whether to continue when things get less busy.
Minor nitpick: do you *actually* believe that electron suffering is something EAs care about or was that a tongue-in-cheek remark to prove your point? Because there is a forum post on it, but it is an April Fools joke.
Almost all EAs are not so foolish. But choosing a "more real" example that to me was almost as absurd - of which there are several that I am very confident are real - seemed like a counterproductive distraction here.
https://reducing-suffering.org/is-there-suffering-in-fundamental-physics/
This is not evidence for the EA community believing in electron suffering. This is just Brian Tomasik being Brian Tomasik.
Sure, it's impossible to prove that every single ea cares about this.
However, everyone who cares about this is an EA.
Thank you for writing this. More than anything else recent (recent enough you can send to a normie and tell them to read it and they don't got "but it's from 2 years ago!") it's a fantastic argument in favor of "maybe people should actually do some thinking and thinking about thinking".
I would add, to the unanswered questions section: How is it impossible for someone so good at games/puzzles to be in Bronze in League of Legends? This is like hearing someone is a competitive-at-world-champs level chess player, but really struggles with tic-tac-toe and checkers. There's probably some good SBF mentality questions to mine here.
My guess is SBF was in the habit of being massively distracted and thus losing matches? In Brawl you often get 30-60 second breaks and can somewhat move them around, in LoL not so much? Or the micro--optimizations the game is all about simply bored him, so he never bothered doing them right, and that kept him at bronze.
Or perhaps he was intentionally losing to maintain lower ranking.
I do find 'why did he keep playing that game instead of switching to another' to be a good question. But I suspect that being punished for distraction was a bug rather than a feature - the book describes the constant pressure to act as something he loved.
That's a fair point. In league, if you stop paying attention for 60 seconds coming back and winning becomes extremely hard (when you combine demoralization effects to your team plus actual game state changes).
EDIT: on further reflection him being in Bronze makes a lot of sense given the information available. He joins a game expecting to play, his EV/attention calculus shifts and he afk's for 30-60 seconds and his team has a conniption while falling behind. When a more normal person rejoins they are put off by the external negative effects and chat consisting people telling them to jump off a bridge (hypothetically, in a video game). SBF discounts this to 0 (self assessing that he had to act has his EV fluctuated wildly by the moment). But, he also gets to enjoy a lot of fun gameplay as bronze people are terrible so as long as his mechanics are remotely decent he can probably personally outplay them and still have fun (even if he loses b/c his team is playing terribly after tilting off the face of the earth).
High intelligence is helpful but neither necessary nor sufficient to be good at League.
This is not the kind of game that you can completely solve intellectually from the first principles; there are too many moving parts—it's practically intractable for a human mind, even if you do remember all the mechanics and interactions (that's a huge if, btw). So you have to make educated guesses. And while intelligent people are usually good at that, it also requires a lot of domain knowledge.
And then you also have to make these decisions fast, because the game is not waiting for you. Which you can't without good reflexes and/or good foresight. You need a lot of experience for both, and intelligence only helps you so much here (you can learn more efficiently if you are smart, but there are limits, and it's also not automatic; you have to actually make decisions to train specific skills to take advantage of that). I know quite a few not-all-that-intelligent folks that are quite good at making split-second decisions in league simply because they've seen most standard situations in the game enough times for their brains to successfully pattern match the situation with optimal response. You simply can't do that with intelligence alone; you will be too slow.
Now I don't know if he had the necessary prerequisites for developing the kind of skills required to excel at League or how much he played, but frankly, it does not matter.
There are many things necessary to be successful in League, and the right mindset is a big part of it.
You cannot have the kind of views and values that SBF apparently had and be good at League. It is a game where your bad bets bite you in the ass really reliably and really fast. If you keep making high-risk calls, that is guaranteed to cause you to lose, often. And you can't lose 100 games, but then win 1 and come out positive like he could in crypto.
"If you keep making high-risk calls, that is guaranteed to cause you to lose, often. And you can't lose 100 games, but then win 1 and come out positive like he could in crypto."
I think that's the key understanding here.
He was playing while in meetings and stuff, right? Where his attention was (intentionally?) split? I don't get the impression that he was playing to win, so much as playing to keep himself from getting bored.
I was also, to some degree, friends with Sam when I lived in NYC. At least on the level of occasionally talking. We also talked about him leaving Jane street. I wish I wrote down what exactly he said. I don't remember him telling me he was going to run or help run anything. He just told me he was going to take a job. This seemed kind of dumb but 'take a job at an official EA org regardless of other opportunities' seemed like the kind of thing EAs do all the time. I just figured Sam's EA job quickly didn't work out for some reason. I'm still not sure if he was blatantly lying or the EA job just fell through. But I don't have any notes. Definitely feels like a bit mysterious to me.
It's more gonzo than gonzo. I knew the outlines of the story already, but the various cognitive train wrecks kept me glued to the entire review. And not just SBF. The entire cast is extremely smart in some ways yet makes dumb decisions that could be trivially avoided, throughout. And these people are all rationalist-adjacent at least. "The first principle is not to fool yourself – and you are the easiest person to fool."
I saw a trial quote that the thing where they bribed chinese officials wasn't in the charges. Why wasn't it? Seems like a blatant crime for which there's a lot of evidence, no?
Yes, it is a direct violation of the Corrupt Foreign Practices Act, and everyone involved is super guilty. As I said, of all the crime. What, they should charge all the crimes no matter how small?
My recollection is that it came out after he agreed to the extradition agreement so it will be tried separately in the future.
> At this time, they were borrowing at 50% (!) interest in order to trade
here’s a Patrick McKenzie thread explaining why this statement from Lewis is most likely wrong
https://x.com/patio11/status/1709271113962852747
Regardless of whether you want to call it a 'hack' the MOBL exploit is at minimum commodities/securities fraud. Its roughly the same exploit that landed Avraham (another rationalist) in jail and will almost certainly cause him to serve prison time. This sort of exploit is absolutely not ok even if some people's intuitions differ.
TBC: I directly mentioned Avraham with exactly this legal take, in case anyone is confused on my position on that. Yes this is illegal.
Is it unethical? That is for the philosophers. If done in the NYSE I would say yes.
the ethics dont seem very complicated to me! I'm all for ignoring unjust laws, but your argument for doing so needs to be a lot better than 'enriching yourself'. Im happy to even accept personal autonomy as a reason for doing things like using illegal drugs. But wanting millions of dollars, which you will be moving from other people to yourself, is an extreme weak justification.
Appreciate this. What a fantastic review! I also read it and agree that it is an awesome book.
But I never believed Lewis was earnest. He was SBF's guest and Lewis surely knows that the crimes he described alone are enough to convict SBF. Emphasizing this in is unnecessary and imo turns the book into something less interesting -- a polemic on SBF. By writing the book in this way he:
1. Ensures future subjects will be happy to let him tag along and speak candidly with him
2. Keeps the focus on the big picture that you end with (although he never specifically states it kind of like the Alchemist)
3. Produces what's probably the first widely-read description of how powerful EA is becoming.
I have to wonder if Lewis is touting SBF's innocence in the hopes of selling more copies of the book to people wondering what the proof of his innocence is.