Q: "James, I want to start a crypto algotrading business. What...
Robot James 🤖🏖@therobotjames
50 views
Apr 18, 2025
~3 min read
Advertisement
1
Q: "James, I want to start a crypto algotrading business. What advice do you have?"
First, understand that you currently have nothing except an idea.
You have a vision of what you're going to need - which includes alphas, analytical capability, technology, processes etc.
First, understand that you currently have nothing except an idea.
You have a vision of what you're going to need - which includes alphas, analytical capability, technology, processes etc.
2
What you're going to want to do, most likely, is get your head down, get in your cave, and get to work making all that stuff.
This would be a mistake.
This would be a mistake.
3
Cos you don't really know what you're going to need.
You're guessing.
You're guessing.
4
And if you get your head down and start building what you *think* you need, then you're not going to be interacting directly with the market.
So you're not gonna get feedback on what you actually need.
And you're not gonna be making money either.
So you're not gonna get feedback on what you actually need.
And you're not gonna be making money either.
5
Also, things move fast.
Especially in crypto.
You might be right about what you need now. But everything takes time.
Edges die, new opportunities present themselves.
Will you want that stuff by the time you've actually made it?
Especially in crypto.
You might be right about what you need now. But everything takes time.
Edges die, new opportunities present themselves.
Will you want that stuff by the time you've actually made it?
6
So, here's the crucial thing.
You have to concentrate on making money NOW.
Making money TODAY, with the resources you currently have available to you, ALWAYS needs to be your primary consideration.
The future never comes.
You have to concentrate on making money NOW.
Making money TODAY, with the resources you currently have available to you, ALWAYS needs to be your primary consideration.
The future never comes.
7
"But hang on," you say. "I don't have any of the tech and processes and research and databases I need to research and run automated trading strategies."
And I agree. That's true.
It's not an excuse though. Suck it up. Get creative.
And I agree. That's true.
It's not an excuse though. Suck it up. Get creative.
8
What could you be doing *right now* that:
- gets you in the market with minimal tech
- has high-probability of making money
- won't take up ALL your time
- is aligned with the vision of where you ultimately want to get to?
- gets you in the market with minimal tech
- has high-probability of making money
- won't take up ALL your time
- is aligned with the vision of where you ultimately want to get to?
9
Start with the simple and obvious.
You want simple mostly-systematic strategies that:
- Exploit stonkingly obvious edges and are very likely to make money
- You can execute manually
- Don't take up all your time.
You want simple mostly-systematic strategies that:
- Exploit stonkingly obvious edges and are very likely to make money
- You can execute manually
- Don't take up all your time.
10
A very good example of this (from when I was starting out) is the short basis trade on perpetual futures.
As well as making money this would have moved you towards your goal, helping you:
- understand idiosyncratic supply/demand patterns for leveraged crypto exposure
As well as making money this would have moved you towards your goal, helping you:
- understand idiosyncratic supply/demand patterns for leveraged crypto exposure
11
- learn how to execute spreads effectively with minimal market impact
- start you on the path of building tooling for exposure/risk management and trade execution.
- start you on the path of building tooling for exposure/risk management and trade execution.
12
Another example: Cross-Exchange Statistical Arbitrage
As well as making money this would help you:
- understand idiosyncratic supply/demand patterns between venues
- learn how to manage x-exchange risk
- build processes, tools for consolidated exposure / risk reporting.
As well as making money this would help you:
- understand idiosyncratic supply/demand patterns between venues
- learn how to manage x-exchange risk
- build processes, tools for consolidated exposure / risk reporting.
13
Another Example: "Dirty Perp-Hedged Leveraged Yield Farming"
As well as probably making money, this would help you:
- understand the basics of decentralized finance
- understand the principles of AMM arbitrage, and how you might exploit it when you have automation capability.
As well as probably making money, this would help you:
- understand the basics of decentralized finance
- understand the principles of AMM arbitrage, and how you might exploit it when you have automation capability.
14
You gotta:
- get in the market day 1, or at least asap.
- prioritize making money *today* over other considerations.
- be conscious of trying to align today's money making activities with where you ultimately want to go.
- get in the market day 1, or at least asap.
- prioritize making money *today* over other considerations.
- be conscious of trying to align today's money making activities with where you ultimately want to go.
15
Don't build anything you *think* you might need.
Build the bare minimum for what you need today.
This stuff moves SUPER FAST and the best opportunities are at the bleeding edge of things.
If I built what I thought I wanted on day 1 I'd have wasted a ton of time.
Build the bare minimum for what you need today.
This stuff moves SUPER FAST and the best opportunities are at the bleeding edge of things.
If I built what I thought I wanted on day 1 I'd have wasted a ton of time.
16
Even if you have no technology at all, your primary consideration every day is: "How do I give myself the best chance of making money today?"
That's always number 1 priority.
If you don't prioritize that, you'll be forever preparing for a future that never comes.
That's always number 1 priority.
If you don't prioritize that, you'll be forever preparing for a future that never comes.
17
@threadreaderapp unroll
