1/n 🧵👉👉👉
2/ There was a moment back in May of last year when everything crystallized for me and my co-founders Craig and Rich.
Luna, 3AC, and all of that craziness was going on. And one of us (can’t remember who) asked a simple question:
Jan 19, 2023 • 4 tweets • 1 min read
Listening is among the most cognitively hard things to do well. 3 levels.
Not Listening
Waiting for them to stop talking so that you can talk.
Bad Listening
Thinking what YOU would mean if you said their words.
Good Listening
Thinking what THEY mean when they say their words.
In social situations we're usually not listening. It's more like a two person game we play, ChatGPT'ing our way through the situation.
In fact, in those spots it's actually a little unsettling when someone listens to what you're saying. They seem a little too intense.
Dec 12, 2022 • 9 tweets • 2 min read
1/ I was asked a question on DM and figured I'd write a short thread about:
Classifying trading days.
👉👉👉
2/ It matters a lot what we mean by "classifying". In particular, most people who talk about this implicitly mean:
"Look at a bunch of days in the past, and cluster them into some classes somehow."
Oct 26, 2022 • 9 tweets • 2 min read
1/ No-code "solutions" are the most frustrating thing ever.
You're solving the wrong problem!!!
Arcane programming language syntax wasn't created in some darkened room to keep out non-nerds (APL is the exception of course).
2/ Syntax exists to solve a very real problem:
Q: How do you translate what you want into a form that a system with ZERO semantic understanding does the thing you want?
A: Programming language.
Oct 21, 2022 • 7 tweets • 2 min read
The question of what trading should be allowed, and what trading shouldn't, is obviously a bit difficult to answer.
Yes, it's a bit of a game we all play to outwit the other. So who's to say what's fair-game outwitting and what's not?
The key concept to develop is the idea of a fair value.
"What *should* this thing be worth?"
I get that some trades and strategies don't care about that question. But for the purposes of understanding which trades are "good" and which ones are "bad/manipulative" it's useful.
Aug 25, 2022 • 4 tweets • 1 min read
Latest job ponzi I've discovered:
Companies penetration-testing their software suppliers.
Up and down the stack, companies are hiring and resourcing teams to penetrate their suppliers, while also having teams to protect against their customers pen testing them.
It's like watching the world computer evolve an immune system in real time. Slow by human standards, blazingly fast by wet evolution standards.
And we're the mindless B-cells, T-cells and neutrophils.
Aug 16, 2022 • 4 tweets • 1 min read
These are terrible terms and always have been.
The only question is: "What is the nature of the value you're providing the world?"
In finance, you basically have two choices:
1. Be the other side of the trade. 2. Provide *access* to the other side of the trades.
1. Because it's a very competitive world out there. Specialization creates the chance to obtain special skill. If you do both you'll be bad at both.
2. No one trusts or will do business with someone doing both.
Aug 4, 2022 • 4 tweets • 1 min read
1/ In trading, if you're wrong 80% of the time then someone can get rich off that. Just do the opposite.
Point is: "bad at trading" isn't the Washington Generals getting killed the Globetrotters every night.
"Bad at trading" is being a noise trader. Bleeding away fee $ forever.
2/ And maybe one time you oversize a position and the variance monster finally gets you. But even that just feels like bad luck.
So what does "good at trading" look like? It depends.
Jul 26, 2022 • 4 tweets • 1 min read
💯
But there's another critical aspect: you can't do this with middle-of-the-road devs. You need world-class people to make this work.
The problem is that most SaaS companies aren't nearly profitable enough (or interesting enough) to hire those world-class devs. They need to make do with 2nd tier people.
The illusion is that hiring more marginal devs will make up for not having great ones.
Jul 12, 2022 • 28 tweets • 5 min read
I've hit 1⃣0⃣0⃣0⃣0⃣ followers!
This would have been utterly shocking to me three years ago, so 🙏 thanks to everyone. As a thank you, here goes a thread on a question I get asked a lot:
➡️How do I come up with good trading strategies?⬅️
🧵👉👉
1/ Sadly, my answer is usually "I don't know."
And that's because I really don't.
A. I don't know your background.
B. I don't know what you're good at.
C. Most importantly, I don't know what interests you.
Jun 21, 2022 • 25 tweets • 5 min read
1/ A friend I know is months away from retiring early.
He’s been very successful at work and wants to spend more time with his young family. But he’s heard horror stories about smart driven people who also retired early and then decided to unretire.
What’s going on here? 🧵👉👉
2/ WARNING: This is some real bargain-basement philosophy and sociology here.
It’s literally just some random dude trying to make sense of the world. Trying to figure it out in real time. Ok here goes…
May 26, 2022 • 7 tweets • 2 min read
Alpha leak of the day: improving beta models is alpha.
Finding signal (alpha) is really hard because it's buried under tons of noise. Beta models are a de-noising technique. A way of saying "this stuff doesn't matter".
Better betas = a better lens to view markets.
So here's a stylized example of what I mean.
Let's say you think there is momentum in TSLA returns. That 9:30 -> 10:00 returns predict 10:00 -> 12:00 returns.
I'm not saying that's actually a great trade (probably not), but it's a clean example.
May 25, 2022 • 4 tweets • 1 min read
This is actually pretty riveting.
If anyone wants to see a real live example of long-entrenched interests talking their own book with a succession of word-salad statements and whataboutist hypotheticals, you can't do much better than this.
"Certainly the National Council of Farmer Cooperatives is pro-innovation."
What follows are 5 minutes of homespun tales about (clearly overlevered) hypothetical farmers.
May 25, 2022 • 10 tweets • 3 min read
1/ There's a lot that crypto could learn from a somewhat obscure Saturday Night Live sketch from 1995.
Yes, 27 years ago... 👉👉👉
2/ Grayson Moorhead Securities
May 7, 2022 • 4 tweets • 1 min read
p. 49: "Sophistication and specialization, caracteristic of most of the Roman world, were fine, as long as they worked: Romans bought their pots form professional potters, and bought their defense from professional soliders...
amazon.com/Fall-Rome-End-…
... From both they got a quality product -- much better than if they had had to do their soldiering and potting themselves...
Apr 26, 2022 • 6 tweets • 2 min read
Alpha leak of the week:
If you don't have a small heterogeneous group of people to talk about trading with, stop what you're doing and build one. (h/t @KrisAbdelmessih) Here's how:
1. Your public twitter presence (reading/following/posting) is not a substitute. Nor is a discord channel with 100 people. You can't be truly honest and vulnerable to a hundred people, and it's hard to interactively teach/learn with them.
Mar 20, 2022 • 39 tweets • 8 min read
1/ A thread about why Moneyball is the best movie about trading ever made.
Yes, we're going scene-by-scene here... 👉👉👉
2/ Opening: Billy is listening to his A's lose to the Yankees in the playoffs. He's trying to maintain an outward appearance of calm. But he can't and eventually he loses it.
"What thing did you sink way too much time into before realizing it wasn't going anywhere?"
My personal (partial) list... 👉👉👉
2/ (this doesn't mean these things don't work, it's just that I couldn't make them work well enough to justify the time spent)
- Any paper by Didier Sornette.
- Hawkes processes.
- Rough vol (8 years ago, maybe it's better now).
- GANs for marketdata
Now it's YOUR turn!
Jan 26, 2022 • 9 tweets • 2 min read
1/ It's kind of amazing how many SaaS startups are basically just playing Settlers of Catan.
Just roll some dice, pick up some cards, and build some stuff. All thinking is tactical.
Settlers of SaaS 🧵👉👉👉
2/ It's the complete commoditization of everything.
CEO: "I want to build a software feature."
Me: "Tell me about the feature."
CEO: "Oh it doesn't matter. Just a feature. Some database, some backend business logic. Some UI."
Jan 16, 2022 • 8 tweets • 2 min read
1/ From my DMs: "Sometimes I get existential dread [...] thinking I'm not contributing anything to society simply by trading as a job."
My thoughts...
2/ It is generally very difficult to make money in ANYTHING if you're not providing value to someone else.
The main exception is when you're *so* good at sales that you can make money by screwing others.
Jan 14, 2022 • 6 tweets • 1 min read
Every day, a thousand recruiters post a thousand LinkedIn requests like this:
"Hi network! Looking for [incredibly niche skillset possessed by almost NO ONE but that you could train ANYONE to do in 3 months]. Please reach out if interested or know someone to refer!"
🤦♂️ 🤯
Seriously, this is such a massive source of alpha that companies are just burning away.
If you need some niche skillset, train them for it! The recipe isn't *that* hard:
