8 ms·

There’s this incredulous passage from Sequoia article that’s been taken down since. Make of it what you will At this point, mid-2019, SBF decided to double dow

There’s this incredulous passage from Sequoia article that’s been taken down since. Make of it what you will At this point, mid-2019, SBF decided to double down again—and scratch his own itch. He would bet Alameda’s multimillion-dollar trading profits on a new venture: a trading exchange called FTX. It would combine Coinbase’s stolid, regulation-loving approach with the kinds of derivatives being offered by Binance and others. He only gave himself a 20 percent chance of success, but, in his mind, SBF needed extreme risk to maximize the expected value of his lifetime earnings—and, therefore, the good his earn-to-give strategy could do. The fact that he was, by his own lights, overwhelmingly likely to fail was beside the point. The point was this: When SBF multiplied out the billions of dollars a year a successful crypto-trading exchange could throw off by his self-assessed 20 percent chance of successfully building one, the number was still huge. That’s the expected value. And if you live your life according to the same principles by which you’d trade an asset, there’s only one way forward: You calculate the expected values, then aim for the largest one—because, in one (but just one) alternate future universe, everything works out fabulously. To maximize your expected value, you must aim for it and then march blindly forth, acting as if the fabulously lucky SBF of the future can reach into the other, parallel, universes and compensate the failson SBFs for their losses. It sounds crazy, or perhaps even selfish—but it’s not. It’s math. It follows from the principle of risk-neutrality.

fwip 4y ago
Sounds like he forgot about the principle of diminishing returns. If you have $1B to your name, it's stupid to put it all on red, even if your expected return is positive.
- JohnPrine 4y ago
  Diminishing returns don't apply for SBF - his life's goal is to fund effective altruist charities, so each billion is as valuable as the last as long as there are good projects to fund. I don't think people are comprehending how tragic this whole situation is (acknowledging that it's SBF's fault). This collapse put the brakes on a powerful force for good, and lives that would have been saved won't anymore.
  - joosters 4y ago
    Well he failed miserably at that. If he had stuck with his original supposed $1 billion, he could have given almost all of it away and still lived a very comfortable life. As it stands now, his charitable foundation has pledged far less than $1 billion (at it remains to be seen just how much it pledged actually gets paid out). Perhaps if he had spent more time donating money (his stated aim) rather than inventing convoluted financial structures (FTX has over 100 related companies!), he would have achieved more, and cost people a lot less.
    - jefftk 4y ago
      There are lots of reasons to be mad and lots of red flags, but having over a hundred companies is pretty normal in this space. To operate a financial business legally across many jurisdictions you generally need to have subsidiaries in each jurisdiction. Ex: https://wise.com/help/articles/2974131/what-are-the-wise-group-entities https://wise.com/help/articles/2974131/what-are-the-wise-gro...
      - joosters 4y ago
        Not true - the FT took a look at the org chart of FTX, it's way more complicated than even Lehman Brothers - https://www.ft.com/content/c28e0570-d4c4-433c-b0a0-c99fba613822 https://www.ft.com/content/c28e0570-d4c4-433c-b0a0-c99fba613...
        jefftk 4y ago
        Lehman Brothers didn't allow individuals in many countries around the world to make trades on their platform, so I don't think they're a good comparison here. (Sorry for saying "financial" earlier when the reference class is really something more specific like "international retail finance")
  - coffeebeqn 4y ago
    It’s only ”good” if you believe the ends justify any means. What about FTX customers? People aren’t pawns you can play with for your “greater cause”
hnews_account_1 4y ago
The super stupidity of using math to dazzle people around you is my greatest lesson from this. In my highly technical field, I can use math without treating it like some precious scripture. But it’s good to know it is so unfamiliar to a normal person in the VC world, I can use it to justify even a startup doing genocide as a platform or something.
- mrguyorama 4y ago
  It's almost like people who don't actually know anything about math, statistics, or probability shouldn't be believed when they make poor arguments using the same. Maybe there's a reason it takes 4 years of extremely tough classes to understand the basics of "hard" math.
  - mjburgess 4y ago
    Four years? Most books, yt videos, jobs and academic press releases tell me all it takes is `import tensorflow` or `git clone github://lolcoin.git`
  - hnews_account_1 4y ago
    It’s not just that. But like if I made up bullshit like “the emerging market populations are growing hard and will strain earth’s resources. Our startup needs to kill them to save the planet, here’s a log Utility function to explain it”, some VC out there is an idiot enough to give me money. Maybe if I’m playing league of legends while saying it, I’ll get more cash than I even want to raise.
  - paganel 4y ago
    > Maybe there's a reason it takes 4 years of extremely tough classes to understand the basics of "hard" math. On top of that, any decent tech school also has an "ethics of engineering" course somewhere along the way, or at least I did about 20 years ago when I was a student (we had quite a stupid professor, but the intention was there).
    - mrguyorama 4y ago
beambot 4y ago
Quick detour: expected value alone does not optimally maximize your long term outcome. Bet sizing (a la Kelly Criterion) is just as critical. E.g. if you had a weighted coin that was 50-50 odds with 3-to-1 payout, you wouldn't bet your whole bankroll on just one flip.
- jeffreyrogers 4y ago
  SBF said he disagrees with kelly bet sizing because his utility function is closer to linear: https://twitter.com/SBF_FTX/status/1337250702104485893 https://twitter.com/SBF_FTX/status/1337250702104485893
  - beambot 4y ago
    I'm not sure I follow... SBF seems to be talking about the marginal utility of a dollar. True, most people think of this value as logarithmic. However, Kelly Criterion makes no such assumption: It's simply the optimal bet size (as a fraction of your bankroll) that optimally maximizes your returns over successive game plays. If you over-bet, it's much more likely that you'll blow-out -- as SBF mentions in your link (and ironically, occurred to SBF & FTX!).
- zen21 4y ago
  Why would a weighted coin have 50-50 odds? That sounds like a fair coin, and miscalculated odds.
streb-lo 4y ago
I this is exactly it. When you let utilitarianism run your life, you try and maximize your ability to do good, even if you have to bend a few rules on the way. But outcomes aren't certain and high EV plays can still carry a substantial amount of risk, which is exactly why we have finance regulations.

fwip 4y ago

Sounds like he forgot about the principle of diminishing returns. If you have $1B to your name, it's stupid to put it all on red, even if your expected return is positive.

JohnPrine 4y ago
Diminishing returns don't apply for SBF - his life's goal is to fund effective altruist charities, so each billion is as valuable as the last as long as there are good projects to fund. I don't think people are comprehending how tragic this whole situation is (acknowledging that it's SBF's fault). This collapse put the brakes on a powerful force for good, and lives that would have been saved won't anymore.
- joosters 4y ago
  Well he failed miserably at that. If he had stuck with his original supposed $1 billion, he could have given almost all of it away and still lived a very comfortable life. As it stands now, his charitable foundation has pledged far less than $1 billion (at it remains to be seen just how much it pledged actually gets paid out). Perhaps if he had spent more time donating money (his stated aim) rather than inventing convoluted financial structures (FTX has over 100 related companies!), he would have achieved more, and cost people a lot less.
  - jefftk 4y ago
    There are lots of reasons to be mad and lots of red flags, but having over a hundred companies is pretty normal in this space. To operate a financial business legally across many jurisdictions you generally need to have subsidiaries in each jurisdiction. Ex: https://wise.com/help/articles/2974131/what-are-the-wise-group-entities https://wise.com/help/articles/2974131/what-are-the-wise-gro...
    - joosters 4y ago
      Not true - the FT took a look at the org chart of FTX, it's way more complicated than even Lehman Brothers - https://www.ft.com/content/c28e0570-d4c4-433c-b0a0-c99fba613822 https://www.ft.com/content/c28e0570-d4c4-433c-b0a0-c99fba613...
      - jefftk 4y ago
        Lehman Brothers didn't allow individuals in many countries around the world to make trades on their platform, so I don't think they're a good comparison here. (Sorry for saying "financial" earlier when the reference class is really something more specific like "international retail finance")
- coffeebeqn 4y ago
  It’s only ”good” if you believe the ends justify any means. What about FTX customers? People aren’t pawns you can play with for your “greater cause”

hnews_account_1 4y ago

The super stupidity of using math to dazzle people around you is my greatest lesson from this. In my highly technical field, I can use math without treating it like some precious scripture. But it’s good to know it is so unfamiliar to a normal person in the VC world, I can use it to justify even a startup doing genocide as a platform or something.

mrguyorama 4y ago
It's almost like people who don't actually know anything about math, statistics, or probability shouldn't be believed when they make poor arguments using the same. Maybe there's a reason it takes 4 years of extremely tough classes to understand the basics of "hard" math.
- mjburgess 4y ago
  Four years? Most books, yt videos, jobs and academic press releases tell me all it takes is `import tensorflow` or `git clone github://lolcoin.git`
- hnews_account_1 4y ago
  It’s not just that. But like if I made up bullshit like “the emerging market populations are growing hard and will strain earth’s resources. Our startup needs to kill them to save the planet, here’s a log Utility function to explain it”, some VC out there is an idiot enough to give me money. Maybe if I’m playing league of legends while saying it, I’ll get more cash than I even want to raise.
- paganel 4y ago
  > Maybe there's a reason it takes 4 years of extremely tough classes to understand the basics of "hard" math. On top of that, any decent tech school also has an "ethics of engineering" course somewhere along the way, or at least I did about 20 years ago when I was a student (we had quite a stupid professor, but the intention was there).
  - mrguyorama 4y ago

beambot 4y ago

Quick detour: expected value alone does not optimally maximize your long term outcome. Bet sizing (a la Kelly Criterion) is just as critical. E.g. if you had a weighted coin that was 50-50 odds with 3-to-1 payout, you wouldn't bet your whole bankroll on just one flip.

jeffreyrogers 4y ago
SBF said he disagrees with kelly bet sizing because his utility function is closer to linear: https://twitter.com/SBF_FTX/status/1337250702104485893 https://twitter.com/SBF_FTX/status/1337250702104485893
- beambot 4y ago
  I'm not sure I follow... SBF seems to be talking about the marginal utility of a dollar. True, most people think of this value as logarithmic. However, Kelly Criterion makes no such assumption: It's simply the optimal bet size (as a fraction of your bankroll) that optimally maximizes your returns over successive game plays. If you over-bet, it's much more likely that you'll blow-out -- as SBF mentions in your link (and ironically, occurred to SBF & FTX!).
zen21 4y ago
Why would a weighted coin have 50-50 odds? That sounds like a fair coin, and miscalculated odds.

streb-lo 4y ago

I this is exactly it. When you let utilitarianism run your life, you try and maximize your ability to do good, even if you have to bend a few rules on the way. But outcomes aren't certain and high EV plays can still carry a substantial amount of risk, which is exactly why we have finance regulations.