FTX Trilogy, Part 1: The Prince of Risk

Editor's note: this trilogy was written in August 2021, fifteen months before FTX's subsequent meltdown. We have kept it up to show our thinking at the time, and the view of the broader market. We reflected on the fallout and what can be learned in a coda to the trilogy, "The Casino and the Genie."

‍This is Part One. You can find Part Two and Part Three here. 

A trader's job is to interpret risk.

In that regard, Sam Bankman-Fried seems to have something of a knack. The man who began his career at quantitative hedge fund Jane Street has carried that training into his most recent venture: FTX.

In less than three years, Bankman-Fried has taken the cryptocurrency exchange to an $18 billion valuation, propelling it at a pace that often makes it the fastest-growing player in the space, trading places with Binance.

That performance is no mean feat in a sector bloated with wealthy competitors, overrun with hackers and charlatans, and stalked by regulators. To get this far, Bankman-Fried has had to balance peril while moving at speed. He has to interpret and manage risk.

Of course, FTX is far from a sure thing. Few other sectors carry the hazards of crypto, let alone its rate of change. But for now, the company seems to have a suitable leader: a CEO wise to the hazards but willing to step on the gas. It is perhaps an added bonus that in his sleeplessness and idiosyncrasy, Bankman-Fried is an apt avatar for the crypto market in which he operates.

Today, in Part 1 of the FTX Trilogy, we’ll outline his origins, before peering into his mind.

The life of SBF

Origins

Sam Bankman-Fried (SBF) was born on March 6, 1992, in Santa Clara County, California. The son of two Stanford Law professors, Barbara Fried and Joseph Bankman, SBF was raised in a highly intellectual environment. That would have a significant impact on his later thinking, as we will go on to discuss. 

In 2010, SBF matriculated to MIT to study physics. As he tells it, he arrived as “sort of a math nerd.” During his first couple of years, he considered pursuing a career in academia, thinking of becoming a math professor. After realizing he took little joy in formal research, he expanded his horizons. Specifically, he looked for something that would suit his skills and merit his efforts: 

The one thing I did sort of know was that I wanted to find out how I could have the most positive impact on the world. I’d been into utilitarianism for a very long time and had recently started getting into effective altruism, which is basically this movement...you’re trying to figure out how to impact the world try and quantify things, try and figure out what the most efficient way of doing that is...

I was playing around with a lot of possible careers which were kind of all over the place. [I] had some conversations with people and basically, they said, “You can go and work for one of these charities or organizations you think are good, or you can donate to them. And frankly given your strengths and weaknesses, maybe you’ll be able to donate more to them than you’d be able to contribute working directly for them.” So anyway I thought about that and it sounded like a pretty plausible argument.

Convinced by the idea that racking up a considerable salary that could be redirected to philanthropy was his best path to impact, SBF spent the summer of his junior year interning at Jane Street Capital. A few friends had interned there previously and said “moderately good things.”

Founded in 1999, Jane Street had risen to become one of the largest and best-respected quantitative trading firms in the world. It has maintained both that reputation and scale: in 2020, the firm traded $17 trillion in securities. 

An enjoyable summer stint led to a full-time job after SBF graduated in 2014. It proved a good fit. SBF reveled in being surrounded by a “bunch of nerds” intent on surfacing and executing sharp trade ideas. He focused on international ETFs, a slightly exotic flavor that presaged some of his more eye-catching later work in the Asian crypto markets. 

Though he describes Jane Street as a sterling employer, after three and a half years, the young financier decided it was time to build something of his own. He remembered his rationale at the time: 

[I was thinking] I have a lot of things I want to try with my life. I don’t know what’s going to end up being the right thing. But at least one of them might go extremely well.

Alameda

However optimistic he might have been, even SBF must have been surprised at the scale and speed of his success. After leaving Jane Street in 2017, he spent time thinking through potential opportunities. Intrigued by the crypto boom gripping the markets at the end of that year, he turned his mind to the emerging ecosystem. 

As he began to study the market, SBF’s trading instincts began to hit overdrive: 

It had a lot of the hallmarks that might be a really inefficient system with a big need for liquidity. Which is, basically: gigantic demand all of a sudden, growing really rapidly, lots of volume, lots of retail users, and not a lot of time to build up institutions. Not a lot of time to build up liquidity…

It felt like the thing that was decently likely to have very large volume and price discrepancies.

That interest turned into an obsession when SBF recognized the arbitrage opportunities between US and Asian crypto markets. Because of differences in demand, currencies like bitcoin traded at much higher prices in Korea. This “kimchi premium,” as it was called,  sometimes reached 50%, presenting what looked like a once-in-a-lifetime opportunity. In theory, at least, an investor could buy bitcoin for $5,000 on an American exchange, then immediately flip it for $7,500 in Korea. 

While others spotted this inefficiency and sought to capitalize on it, SBF was quick to recognize its limitations. Because the Korean won was a restricted currency, the size of the opportunity was constrained. You could make money, certainly, but deploying hundreds of millions, repeatably, was infeasible. He looked for something bigger. 

Though less eye-catching, SBF realized that the Japanese market had similar characteristics to the Korean one. Net buying of bitcoin was high, resulting in a 10-15% premium on domestic exchanges. Critically, the Japanese yen was not a restricted currency, making it possible to put real money to work. That didn’t mean it was easy — it required a complex web of intermediaries and “looked like money laundering.”

Seeing an opportunity to make “10% per weekday” by purchasing bitcoin in the US and selling it in Japan, SBF sprung into action. Along with Gary Wang, a former Google engineer and MIT alum, and Nishad Singh, a fresh computer science graduate from Berkeley and high school acquaintance, SBF raised the money to start Alameda Research, a quantitative trading firm. It was, in essence, and ambition, Jane Street for crypto. 

The earnings from the Japanese arbitrage were just as ludicrous as SBF had imagined. With a fund that reached $100 million at its height, Alameda deployed and redeployed that money in the bitcoin markets, earning 10% — or $20 million — every day. 

But SBF was not content to count his winnings and ride into the sunset. Roughly a year after founding Alameda, he and his team began considering an even more significant opportunity: building a cryptocurrency exchange. 

The state of the art was poor, SBF felt. In running Alameda, he’d come to appreciate just how much room for improvement there was in the underdeveloped world of crypto. Indeed, despite what Coinbase and others might have marketed at the time, true institutional-grade trading was a ways off, particularly for those interested in more complex securities. BitMEX provided access to derivatives, for example, but like many others, it was prone to outages and other performance issues. With their hands-on experience, SBF felt the Alameda team could create an exchange made for professional investors like them. 

FTX

In late 2018, SBF, Wang, and Singh began work on FTX, an exchange “by traders, for traders.” 

There was a problem, though: the US was a hostile environment in which to build a crypto derivatives exchange. Domestic regulators had shown skepticism toward the emerging asset class, with particular enmity aimed at more speculative investments. The SEC, for example, had (reasonably) dampened the ICO craze in early 2018, and warned it was keeping a close eye on exchanges over the course of that year. 

As the team formulated a plan of attack, SBF made a decision: FTX would be incorporated in Antigua and Barbuda, and headquartered in the comparatively crypto-friendly city of Hong Kong. It represented a significant move, requiring relocation for Wang, Singh, and other vital players. 

It didn’t seem to slow development, though. In May of 2019, FTX officially opened its doors for trading. In the two years since it has grown into an $18 billion business, and one of the most popular exchanges in the world. 

While so much of that is down to the product itself as well as corporate strategy, SBF’s role can not be understated. To understand what FTX is, and what it might become, we have to not only know its founder’s biography but try and grasp what truly makes him tick.  

Understanding the man

A note on profiling

There is something nearly unconscionable about undertaking a profile. Not only does it require the writer to observe the subject with slavish obsession — breathlessly scrawling each stirring phrase, eavesdropping on every previous interview — it invites a nosiness that may only sometimes intersect with deep understanding. How much can we learn from a short conversation? How inquisitive is it reasonable to be in a first meeting? And what insight can we glean from life’s sediments? Is there value in the old tweet, bouncy CNBC section, long podcast, surprise poem? 

The ends only partially justify the means. 

After all, most of the time, the subject observes, too. Even if you are not in direct conversation with them, you may as well have been. Someone like you was, after all. Aware of that attention, the modern executive cannot help but direct the discussion toward talking points, the familiar rut of an established narrative. If the observer’s job is to sketch a portrait, the subject’s is to guide their hand. 

I am particularly worried about this in studying Sam Bankman-Fried. This is no slight against him. He is an uncommonly accessible, friendly executive — within minutes of chatting on Twitter, we arranged a call. He has been generous with time and information. Rather, it’s an admission of the fact that prior coverage has contoured his story, set a narrative. Every article discussing the man flits through the same three-step: he is young and very rich! He sleeps on a bean bag in his office! He cares about the world!

These are all true or seem to be. And we will predictably speak of them, too. But beneath (and beyond) these recitations of SBF, is the sense of a vast hinterland. Of someone with a deep intellectual and emotional life that is hard to tap, perhaps even for those close to him. 

I have done my best to play the filching voyeur. But I have no definitive explanations for you, no magical secret that unlocks SBFs psyche. Just an attempt to understand, however flawed the mission. In particular, I’d like to investigate four traits I believe shade his character:  

1. Chaotic good moral alignment
2. Rapid processing speed
3. Suspicion of conventional wisdom
4. Rare ability to toggle perspective
   

I will look to assess the tradeoffs of these characteristics, along with their strengths. 

Chaotic good moral alignment

Given his interest in fantasy games like League of Legends, I feel reasonably confident hazarding that at some point in his life, SBF has dabbled in Dungeons & Dragons. For those that have yet to while away an afternoon, evening, and night engrossed in the role-playing game, the basic idea is to create a character of one’s own, then undertake a quest alongside the rest of your party. 

In devising a character, players are asked to choose a race (human or elf, for example), an occupation (rogue or wizard), and an alignment. This final decision requires the participant to determine the moral and ethical outlook of their character along two axes: lawfulness to chaos and evil to good. The options look like this: 

Someone like Captain America is a classic example of a character with a lawful good alignment. He plays by the rules and is motivated by philanthropic desires. Someone like the Joker, by contrast, is an example of chaotic evil. He just wants to watch the world burn. 

SBF may be the most chaotic good founder on the planet. It’s clear that he’s driven, in some respects at least, by altruism. But the manner in which he operates FTX demonstrates a willingness to play fast, just within the boundaries. 

Let’s unpack his positioning on both axes, beginning with the good. 

As mentioned, SBF has shown a sincere interest in altruism that predates his wealth. Before starting Alameda, he spent a couple of months serving as a director at The Centre for Effective Altruism. Today, FTX commits to giving 1% of its net fees to charity, a sum that has surpassed $10 million to date. 

SBF’s grounding in moral philosophy can be traced back to his parents. A keen interest in ethics is apparent in the work of both academics, though it is particularly prominent in Barbara Fried’s. In one of her most notable papers, “What Does it Matter?...”, she dissects the limitations of one of moral philosophy’s memes: the trolley problem. Though it would represent too much of a detour to fully unpack Fried’s argument, the work harmonizes with our discursion in its discussion of probability in ethical decision-making. There is a symmetry here: just as her son has proven adept at interpreting the intangible world of financial risk, Barbara Fried illustrates expertise in untangling the haziness of morality. 

In our discussion, SBF noted his interest in the work of Peter Singer, and early utilitarian Jeremy Bentham. SBF’s decision to become a vegan echoes utilitarianism’s analytical approach to morality: 

It’s a chicken being tortured for six to eight weeks so we can spend half an hour eating it. That makes no sense.

SBF’s ethos is not merely abstract or disconnected from the world in which he operates. He has also demonstrated a sense of duty to the crypto ecosystem, best shown by his willingness to step into an interim leadership role at SushiSwap. When “Chef Nomi,” the creator of the decentralized exchange fled the project with tokens in tow, SBF stepped into the power vacuum and steadied the ship. When I asked him about this he referred to it as a “time of need type of thing.” He added: 

The community was going through a hard time with the Nomi escapades. I was just helping to get it in the right place. To get it to stand on its own. I’m super excited by what the team has done.

It’s easy, in the face of his good-spirited acts, to overdetermine SBF’s motivations. Many media outlets cast his story as a quasi-messianic quest situated within a capitalist framework. But that is a disservice to the reality of company building. Though having a higher mission helps, the long hours and constant stress are hard to sustain if not driven by genuine interest. While SBF is admirably morally aware, we don’t need to burden his company with the holier-than-thou trappings favored by nonsense-peddlers like the founders of WeWork or Goop. FTX is a business, and it is run by a businessman.

The second part of SBF’s “choatic good” alignment, of course, is the preference for chaos. This manifests on both an individual and organizational level. 

SBF seems to thrive in disorder. On the day I spoke with him, he mentioned that I was his sixteenth meeting. He ran me through them: 

1. Media interview
2. Media interview
3. Discussion with a potential venture investment
4. Discussion with potential FTX Pay partner
5. Catch up with an investment bank
6. Internal meeting on licensing issues
7. Internal meeting about regulatory issues
8. Internal meeting about potential NFT products
9. Interview with a potential new venture partner
10. Discussion with lawyers about regulatory issues in another jurisdiction
11. Internal debrief about regulatory issues
12. Catch-up with a company being considered for acquisition
13. Internal meeting with legal and compliance
14. Discussion with a potential venture investment to bolster FTX Pay
15. Discussion with a sports franchise about an FTX endorsement
16. Interview with The Generalist
    

That’s four legal discussions, three interviews, two investment assessments, one potential acquisition, and a smattering of other matters. 

As he told me, SBF likes this level of activity. “That’s intentional,” he said, “There’s a ton going on constantly.” He manages this disarray thanks to his natural gifts as well as a worrying work schedule. Part of SBF’s hagiography is that he sleeps as little as two hours a night, collapsing into one of the bean bag chairs positioned throughout the office. From the corner of our Zoom call, a blanket sat strewn across one of these beds. As one of his investors told me, “he responds to my texts 24 hours a day.”

Putting aside the durability of this lifestyle, SBF’s preference for chaos feels especially well-suited to the market in which he operates. Just as he illustrated in executing the Japanese cryptocurrency arbitrage with Alameda, he is at his best when things are destabilized. As the rest of the world seeks to regain its balance, floundering in the process, SBF sees clearly. He has a talent for assessing the melee, picking a path through it, and moving quickly. 

Just as SBF is content to live his personal life at the edge, he’s also shown a willingness to run his companies the same way. As he noted in our conversation, "Most of life lives in the margins." Presumably, SBF believes business opportunities co-habits with it. 

From FTX’s inception, he has been willing to act controversially to achieve his goals. 

In preparing for this piece, I spoke with someone familiar with both Alameda and FTX’s early days. They highlighted the clear conflict of interest posed by the relationship between the two organizations. Wouldn’t FTX’s exchange favor the market maker founded by its CEO? How could other traders be sure they were receiving a fair shake? 

The truth is they may not have. The individual I spoke to explained how vital Alameda had been to FTX initially, providing liquidity for the new market. What’s perhaps less clear is whether Alameda received something in return. 

The source noted that FTX drew strong traffic partially because it quoted such tight markets. It seemed impossible to this person that a market maker could actually make money on such spreads, which suggested one of two things were going on: either, Alameda played on a level playing field with others and lost money (essentially subsidizing FTX), or the firm received some kind of advantage. This might have been a first look at the exchange’s trading volume, for example.

None of this is illegal, but it’s thorny territory. So far, it has proven worth the risk. SBF has driven both organizations to new peaks. As time passes, Alameda and FTX seem to be growing increasingly independent, perhaps diminishing potential conflicts of interest. But the willingness to operate at the limit of acceptability tells us something about SBF’s orientation and his desire to win, however turbulently. 

As this might suggest, the chaotic good alignment comes with foibles, of course. 

SBF’s exertion levels may be necessary for the moment, but it does not seem sustainable. Only genetic anomalies are able to consistently operate in a sleep deficit. Is SBF one of them? Old friends and family will know best. Given coverage of this issue seems to stem exclusively from FTX, I would assume he is not among the insomnious few. (Ed: SBF followed up after this piece to share that he has "in fact been an insomniac for about 15 years." He describes his relationship with sleep as "somewhat complicated.") 

Though overly macho investors often insist founders grit their way through such pesky physical limitations, our biology demands tend to catch up with us. The stress of building a generational business serves only to exacerbate the potential for burnout. 

Perhaps the greater risk is that SBF flies FTX too close to the sun. The fear here is that in his desire to succeed, SBF guides the company over some real or perceived boundary. When I asked one crypto investor about FTX’s regulatory risk, for example, they responded, “Will they face penalties at some point? Probably.” They continued by remarking that it would make little to no difference to the company’s trajectory, in their estimation. 

While every executive is capable of a misstep, it’s clear SBF is aware of his business’s risk. That makes it hard to visualize a truly catastrophic gaffe. Though he may be chaotic, SBF is also calculated.

Don't miss our next briefing. Our work is designed to help you understand the most important trends shaping the future, and to capitalize on change. Join 53,000 others today.

Rapid processing speed

Part of the reason SBF has been able to manage this level of input is that he seems to have elite processing power. I get the sense that part of the reason he is happy to have sixteen meetings in a single day is that his brain almost...needs it? While the rest of us might be glad to have a couple of projects on which to direct our attention, depending on our mood, SBF appears to crave a dozen or more. 

As it stands, he is the head of three organizations: Alameda, FTX, and Serum, a decentralized exchange spun out of FTX. While he noted that he is less active at Alameda these days, it is nevertheless an incredible amount of context switching to manage. This is without even including his stint at SushiSwap, of course. 

SBF has spoken about his abilities in this area, also articulating its tradeoffs. In a tweet thread from February of this year, he shares a little of how his mind works: 

According to this simpleton's understanding, there are two types of memory in computers: RAM and hard drives.

RAM is fast to access, expensive, and small.

Hard drives are slow, cheap, and vast.

The computer I'm using right now has 64GB of RAM and 500GB of disk space.

Also, RAM empties out every time you restart your computer; your hard drive persists, and is what saves state.

Anyway, I've found this to be a useful way to think about how *I* remember things.

And, in general, I think I have tons of RAM, and a relatively small hard drive.

The trade-off is not that SBF has a “relatively small hard drive,” by his reckoning. Indeed it only becomes clear later in the thread. In explaining why he often plays online battle game League of Legends, SBF notes: 

I play a lot more than you'd expect from someone who routinely trades off sleep vs work.

Why?

Well, there's one answer, which is the obvious one.  The single most universal thing about LoL is that everyone who plays it says they wish they didn't…

And maybe that's just the answer. Maybe not, though…

[W]hen I'm really, truly exhausted, why do I sometimes instinctively open up League?

Sometimes I'm physically tired, and I sleep.

But sometimes my exhaustion is mental.  My mind will be spinning, my RAM full of everything important to me.

Because I don't have much disk space, and don't trust it much.  I live my life in my RAM.

Even if I wanted to empty it out, I couldn't.

For better or for worse, those thoughts valuable enough for my active memory don't leave it.

For most people, short-term memory is reserved for things you'll soon forget.

But I stuff my mind full of things to remember and do and think about, and those thoughts linger for a while.

Sometimes forever.

Because once they transition from RAM to hard drive, they're mostly gone.  Worlds alive only as long as I remember them.

So, anyway, sometimes my mind is too full, or full of things demanding and exhausting.  I want it to calm down.

I'll try to lie on the beanbag, but it doesn't help.  My mind is still spinning.

Clinging to its looping thoughts.

I lie awake, an insomniac once more…

And so I'll open up League of Legends.

And, without thinking, I'll enter a game, and draft, and start…

There's no room left to think about anything else.

And so my mind shifts to a new, very different loop of thoughts, obsessing over last hits instead of responsibilities.

And the old thought loop--the exhausting one--is forced out of my active consciousness, left to spin alone.  Biding time.

It'll be back.  In a few minutes super minions will overrun our nexus and my mind will discard its League thoughts, welcoming back the old ones.

But I've bought myself peace of mind for 30 minutes, and given myself time to do what generally happens during sleep:

time to give my mind some rest, and respite.

Time to process my thoughts, and consolidate them, and make peace with them.  At least for a bit.

And then, it's back to work.

It is an almost beautiful text when read in another context, away from the dopamine bait of social media. Barbara Fried, an accomplished poet as well as professor, would find music in her son’s arrangement. But it’s clear that though SBF has supernatural abilities when it comes to processing speed, it may contribute to exhaustion. Again, the imperative here is that investors and leadership ensure sufficient resources are allocated to reduce the strain on FTX’s founder. 

Suspicion of conventional wisdom

One of my favorite quotes comes from Machiavelli’s The Prince: 

A prince who is not himself wise cannot be wisely advised...Good advice depends on the shrewdness of the prince who seeks it.

SBF exemplifies this approach. You get the sense that every opinion has been constructed from scratch, rather than borrowed offhand from some prior teacher. 

At one point in our conversation, he mentioned how, for example, he was ill-convinced by the mantra, “it’s better to be safe than sorry.” 

Is it? 

As SBF asked, rhetorically, “Are you adding value in the world with that philosophy and that advice?” 

It’s a small anecdote but it exemplifies the way SBF thinks and problem-solves. He shows a healthy skepticism for conventional wisdom and devises from first principles. This was a trait highlighted by one of FTX’s investors, Kyle Samani from Multicoin Capital. Samani reiterated at several points that SBF was exceptional in this regard, demonstrating an obsession with “first order correctness.” On that subject, Samani remembered a conversation in which SBF considered building Serum on Solana rather than Ethereum. 

Over the course of just a thirty-minute call with Samani and Solana founder Anatoly Yakovenko (remember that rapid processing power), SBF thought through the needs of the new system, analyzing factors like optimal throughput and acceptable latency. By the end, he’d reached his decision: Serum would use Solana. 

Almost all great entrepreneurs exhibit some predilection for thinking from first principles — innovation often requires looking at a problem set considered static and remixing different elements. If there is a trade-off to this approach, it’s usually speed — reframing an issue from the ground up takes time. Clearly, that doesn’t seem to be a problem for SBF. 

Ability to toggle perspective

Perhaps the highlight of my conversation with SBF was when he disappeared down a rabbit hole about perspective. We had been talking about thinkers he liked, a question that didn’t sit altogether comfortably: 

[There’s not just one person.] There are bits and pieces I try to patch together from different people. In so many cases, people will have really important insights that I missed. Then they’ll also have a lot of really dumb thoughts. 

He spoke, with cheery frustration, about how people often give advice when they are ill-positioned to offer it. They might assume they have a grasp on the situation and its various intricacies but actually lack the context to be useful. He noted this was one of the impediments to running an aligned team: sometimes an employee will become fixated on a problem they believe is being ignored, but has actually already been considered. 

That SBF finds this irritating tells us something about him. Whenever he listens to someone, SBF seems to be mentally toggling between their perspective and a few dozen others. 

Again, Samani called out this ability as a particular gift. Like no one he’s ever met, SBF is capable of jumping between the nitty-gritty details, then leaping up to macro perspective. He understands the details of every type of trade possible on FTX’s platform, while also playing pan-national chess with competitors. He appears equally able to dance between the functional perspectives of team members.

As with the previous characteristic, this talent might lead to slow decision-making in others but it doesn’t in SBF. Perhaps a more plausible risk would be that SBF expects a similar panoramic vision from other members of his team and grows frustrated when that is absent. It would be hard for anyone else in the organization to have his level of insight, particularly across different entities. Could that lead to unrest? It seems unlikely to present a meaningful problem. 

It’s hard not to end a study of Sam Bankman-Fried without feeling impressed. Across my interviews with investors, crypto experts, and the man himself, I was struck by the sense that this was a special executive, and an unusual human being. One source said, entirely seriously, "Sam is the most capable human I've ever met. And that should be interpreted as broadly as possible."

Simultaneously open and enigmatic, practical and intellectual, ethically principled and operationally flexible, SBF demonstrates a rare set of characteristics almost ideal for the sector in which he works. Like few others, he seems at home amidst chaos, able to calculate and calibrate to risk’s ever-changing shape. 

In FTX, he has found his cathedral. Crypto’s fastest-growing exchange is impressively creative and remarkably daring. That isn’t to say it hasn’t stoked controversy. In Part Two, we’ll explore the company’s triumphs and its vulnerabilities. For now, we can be sure it has a fitting ruler.

Read Part Two and Part Three, next. 

Don't miss our next briefing. Join 53,000 readers and get powerful business analysis delivered straight to your inbox each Sunday. No spam, no noise. Unsubscribe anytime.