Understanding Crypto - Algorithmic Trading

Times really have changed. The first publicly traded company was The Dutch East India company. To boost the sale of shares, the company created the Amsterdam Stock Exchange in 1602 and for the first years their shares were the only ones traded there. If you wanted to invest in a merchant trip, you had to be right in that exact building and look for someone who wants to sell shares to you. While many aristocratic men sent their servants to buy and sell shares, only the rich could afford to do that. Not only to invest, but spend their time reading about stocks instead of working in the docks.

After a while telephones got created and you could actually call in and place your order, making it easier, soon to be replaced by the personal computer and internet access in everyone’s home. Now everyone could play and even do it while working. Check the market from your office computer and make changes as you wish.

We have now entered the next generation of trading. Crypto markets never close, you can buy assets instantly and only you decide how much you are going to invest. At the same time you do not have to go to sophisticated parties and rub elbows - you get the same information about the company on the internet as most people do. No entry barrier at all - you could register in an exchange right this moment and start trading. 10 minutes and 10 dollars later you would be a trader and able to take profits. So one thing is clear - the power has been taken away from those we stereotypically consider to be “businessmen”, which raises the question... What is the next thing the common folk can take from the conservative trading elite?

What is Algorithmic trading?

Big companies that deal with stock trading have huge departments full of programmers. Their task is to create tools that will do the trading for them. Tools that notice the price changes, calculate fees, calculate different currencies and, if they deem it to be profitable, buy the coin. And it is not done in a second, as a second would be way too long. We are talking about microseconds here. That means that a regular trader like you? The second you notice a good price you are already too late. A computer has done everything before you.

That is why the common trader has almost no place in scalping or small trades like that. They tend to make bets for a longer period of time and base those orders on the overall sentiment around the exact asset. Basically they make smart moves when they expect the price to rise. The problem is that it is not all there is. There are a lot of other things one could do to earn in the market.

What can computers do?

The algorithms have grown quite complex, so it is difficult to explain everything they can do. Because of that we will mention just some examples to give you a feel of what a computer can do better than the average trader can.

1) Arbitrage. We already wrote about it here. It is a way to earn from the differences in exchanges. Computers help you not only find these anomalies, but also create complicated networks and systems to create a single trade from many small trades, at the same time calculating the fees in a mere second.
2) Microscopic changes. Imagine you buy 10 000 coins for 100 cents. You would probably wait until they cost about 120 before you sell. A computer doesn’t have to do that. It can trade 1000 coins the second the price jumps to 101. And buy them back when the price is 99 cents again. A person would not even be able to notice the change when a computer would already have profits. Just be careful to find an exchange without trading fees.
3) Technical analysis. There are a lot of cycles in the markets, a lot of correlations. Sometimes they can appear in places where they are not expected. A person would not imagine to look. But you tell the computer to look for them and it will scan everything, eventually finding a good couple of stocks to trade.

These are just some of the things computers can do better. They are fast, all-seeing, precise. No computer ever has gotten too happy or too excited and made a mistake. They are as good as they are told to be. Speaking of…

Where to start?

There are two main choices if you want to start using a tool to help you. You can either buy it or make it yourself. Of course, buying it is easier. One would suggest you try several tools and see what exactly you are looking for. There is no need to buy the most expensive tool in the market only to realize you are unable to understand 70% of the options that it offers. As always, start small, let the needs grow organically, then start investing to ensure better gains. There are several free options too, that will allow you to make simple algorithms.

The other option is to make one yourself. While some might think that it is pointless to make something that has already been made, that person has probably not met a really enthusiastic programmer. It is not only a tool they are making, they are learning, gaining experience and enjoying themselves. If your tool is not fast enough for the real pin precise scalping, it can work better as a tool to find correlations or predictions. It is your own tool, you decide what it will do. Of course, understand that you are literally gambling on your ability to code everything correctly.

In either case, you choose your own strategy. Before making it as an active one and placing real money in it, you can always see how it would have worked in the past. Take a look at bitcoin last month or last year and implement the strategy there. Is the algorithm correct and you got “imaginary” money or did it sell at the dip and buy at the peak? Only when it is tested like this, it can start working with small sums and then work up to be placed at more important tasks.

Cons of human-less trading

Yes, computers are fast, have no emotions and make no mistakes. And that can be a real problem, as they also do not read the news. If you have set an algorithm, you have to consider that it will always be active. If you get the news that your favorite asset is being discontinued, the first sellers will activate your algorithm and when everyone will be in a selling frenzy, it will buy the now useless asset. So you are visiting your parents, you hear the news on the radio and know that all the time you are driving from your parents to your computer, it is buying up everything others throw at it.

There was the strangest market crash on May 6th, 2010. While it lasted just for 40 minutes, it had a great impact. New regulations against fast trades like these were implemented and many practices outlawed. The most interesting thing in all of this is that the main culprit was named to be a single man for this trillion dollar crash. With the help of a modified software he managed to place and cancel huge orders in mere seconds, thus disrupting the market and triggering other mechanisms that all continued to deepen the crisis. It is hard to predict where everything would have landed without any human intervention.

You need to understand that this changes the whole feel of the game. You are no more the smart investor placing big bets on things you “feel” are going to rocket. You become a mathematician, a coder. Someone who doesn’t even know the name of the coin being traded or the project under it. All you know is that the asset has done this and this in the past and there is a chance it will do this in the future. So you tailor your algorithm, add functions, place safeguards, test it and launch it. Without any feel of the culture or hype. Just numbers.

Thank you for reading our blog. If you have any questions about the topic or want to suggest a new one, please write an email to [email protected].