How to Automate Your Crypto Investing Like a Pro

How to Automate Your Crypto Investing Like a Pro

This is a free to read, premium post. Subscribe to Stack for just $8.25 / month (billed yearly). You'll get access to our ranks, insights and easy to read reports. Plus premium subscriber posts like this one.

Note: We do not have an affiliation with any of the companies here, and there are no affiliate links in the post. Support more posts like these by subscribing to Stack.


Automated investing is having a coming of age moment. From simple robo-advisors, to full-blown trading bots, there are now a number of ways for ordinary investors to get started. No PhD in mathematics required.

It's a bit of a minefield, though. So I've put together an in-the-trenches, unbiased look at what's possible today for ordinary retail investors.

Why automate? 

Over $1 trillion is managed by quantitative hedge funds that build algorithms to remove human emotion from investing. 

Computers are better and calculating things and humans are notoriously emotional, which often leads to bad investment decisions.

The super-secretive hedge fund, Renaissance Capital, pioneered the quantitative investing approach. Take a look at the returns for their Medallion fund:

Renaissance Capital Returns

Over several decades, the fund has beaten Warren Buffett. Many famous investors like Peter Lynch, Ray Dalio, and George Soros all trail Renaissance in average annual returns,

It's clear that this approach can work, but up until now it's largely been out of reach of all but the most sophisticated hedge funds.

But wait, can't I just HODL Bitcoin? 

Avoiding big drawdowns 

I don't know about you, but I don't love seeing my hard-earned portfolio crash by 70%+

Simply buying and holding crypto over the past decade has been an incredible investment. But it's involved weathering these notoriously painful drawdowns that test investors. Buy at the wrong time, and you can be underwater for years.

Source - Visual Capitalist

Let's face it, HODLING is easier said than done. Automated trading strategies can offer a way to profit from crypto with less volatility by helping remove emotion from the process.

They also enable traders to execute strategies without being glued to the screen 24/7. 

Automation is not for everyone, but if you want to take a more active approach, then read on. 

Levels of automation 

Whether it's to improve trading returns or automate an investment strategy, what are your options? 

I've grouped the various platforms and tools by level of complexity to help provide a visual model of the space. The further down you go, the more difficult the tools become, though it's of course possible the rewards will grow as well.

Levels of crypto automation

Let's explore each stage. 

Automatic Dollar Cost Averaging (DCA)

Who's it for? 

Beginner investors.

What is it?

Arguably, the most basic way you can automate your crypto investing is to simply purchase a pre-defined amount of crypto on a regular basis. 

It's not particular exciting, but it can be an incredibly powerful, simple strategy to help avoid such painful drawdowns. 

By repeatedly buying regardless of market conditions, you ensure you purchase crypto at a wide range of prices, thereby averaging your cost basis.

This means you are guaranteed to avoid the dreaded "bought at the top" scenario right before an 80% crash. Instead, you simply continue to buy on the way down.

Platforms 

  • Almost all crypto exchanges allow you to set up recurring buys 
  • Use simple trading bots to set up more advanced DCA strategies, e.g. buy the dip. See below.

Pros

  • Very simple and easy to execute
  • No additional costs, though watch out for fees on some platforms
  • Guaranteed way to reduce overall drawdown percentage

Cons

  • Limited flexibility
  • Can underperform lump sum investing or more advanced strategies 

UI Trading Bots

As I mentioned at the top, you don't need a PhD to automate your investing anymore. You don't even need to know how to code to get started. There are now a wide range of user-friendly platforms out there. 

Who are they for? 

Serious crypto investors and traders 

What are they?

These platforms all have a user interface that enables you to design custom strategies, from simple "buy the dip" to more complex ones, even utilizing signals from Trading View. 

All of them tend to have a range of pre-built strategies and even marketplaces where you can buy winning strategies from other traders. 

Platforms like Cryptohopper also enable you to backtest strategies before you put them into action. Backtesting involves feeding the strategy historical data to see how well it would have performed in the past. Of course, there is no guarantee any strategy will continue to work in the future, but ensuring it worked well in the past is essential. 

If you're familiar with Stack's insights, they work similarly. When a new indication is detected, e.g., a candle or mean reversion indicator—we analyze the past performance of that indicator to see what the following trading period looked like.

Source: 3Commas

Here are just a handful of platforms that I've come across, it's worth noting that there are a LOT of these crypto bots on the market.

I'd recommend sticking with one of the more established ones (see full warning below).

Platforms

What to be aware of

All of these platforms require access to your exchange account by sharing your API keys. Ensure that the API key doesn't have the ability to initiate withdrawals, and keep in mind that any bugs and errors with the platform may result in loss of funds.

Stick with a more established, well-reviewed platform and limit your exposure by only keeping the funds you want to trade with in the account.

Code Based Platforms 

So you know how to code? The good news is that there's now platforms that make building, testing and deploying quantitative investment strategies much easier. 

The bad news is that just knowing how to code isn't going to give you a significant advantage alone. A knowledge of the way markets operate is essential. 

You've probably heard quotes along these lines:

A computer lets you make more mistakes faster than any other human invention in history...with the possible exception of handguns and tequila. — Mitch Ratcliffe 

And this classic:

The point here is that computers can amplify our wins, and losses. 

Creating your own trading bot potentially offers more reward, but it also opens you up to a new category of risk.

Start by reading some books on market microstructure, the pitfalls of backtesting and algorithmic trading. 

Health warnings out of the way, let's dive in.

Who are they for? 

Serious investors who know how to code.

What are they? 

These are platforms that allow you to define your strategies in code, typically Python or C++.

Backtesting and live deployment is then taken care of, meaning you don't need to write custom connectors to exchange or bring in external backtesting libraries.

That opens up a lot more flexibility, allowing you to create truly unique strategies, employ custom machine learning models and any data sources you wish.

Quantconnect also has a large library of data that you can use to build algorithms with. Blankly is much newer, but the developer experience.

There are also some platforms out there that straddle the UI / Code division, allowing users to both define strategies in a browser based UI and then move to code if necessary. The ones listed below though are strictly code only.

Platforms

What to be aware of

Some of these platforms, e.g., Quantconnect and Blankly require you to upload your strategies to the hosted platform in order for them to execute. Like the UI based bots, that requires a degree of trust in the provider. Others can be self-hosted entirely, meaning you can keep full control of your code. You will however be responsible for deployment.

DIY Bots

If you want maximum flexibility and power, then building your own bot from scratch is the way to go.

Who are they for? 

Serious coders with deep knowledge of financial markets

What are they? 

No platforms here, just a collection of libraries and frameworks available to help you build a truly custom strategy. 

The Awesome Quant list on GitHub is, well, an awesome list to get started. You probably don't need to build your own backtesting framework, for example. 

Here are a few standout libraries that I've worked with:

  • CCXT - this library is an absolute godsend. It allows you to communicate with virtually every centralized exchange out there through a single, unified API. Ideal for trading across multiple exchanges. Better still, it's available for PHP, Python and Javascript. 
  • Backtrader - established Python backtesting library, though it hasn't been updated for a while. 
  • Backtesting.py - another popular python based backtesting framework, this one has more recent activity, so could be a good alternative
  • TA-Lib - available for python, there is also a PHP extension available

By cobbling together these open source libraries, you can kick-start your own solution. If you're looking to get a serious edge over the rest of the market, then this is going to be the best approach. 

What to be aware of

There are many, many more libraries on Github. I would recommend checking how frequently the library has been updated, though, as a lot of the list looks potentially abandoned. 

Building your own bot / trading system is going to take significantly longer than using a UI based bot, and quite a lot longer than using a code based platform. It could easily become a full time occupation.

Closing thoughts

  • There are now a wide range of platforms that everyday retail investors can use to automate trading and investing strategies.
  • It should go without saying, but automating your investing doesn't automatically guarantee higher returns. Treat the market with respect and remember you're competing with professionals.
  • Keep things simple to begin with and be sure to research, then backtest any strategy before taking it live.
  • Separate trading funds to avoid catastrophic losses. Use separate API keys with minimal permissions, then remove unused ones from your account.
  • Accept that losses are part of the game when trading. Keep them small to avoid wiping out.
Important Disclaimer: This material is for informational purposes only. None of the material or any material on the stack.money website should be interpreted as investment advice. Stack does not make any express or implied warranties, representations or endorsements whatsoever with regard to the material or related information. In particular, you agree that Stack and it's owners assume no warranty for the correctness, accuracy and completeness of the material.