Hacked Exclusive: “The algorithm knows something you don’t.” An Interview with One of the World’s Top Algorithmic Crypto Traders

When Alexander Gordon-Brown first heard about Bitcoin back in 2010, he thought it sounded mathematically interesting — and financially impractical.

“It seemed like a passing craze”, he tells me, with no small sense of irony. Though his views weren’t uncommon at the time (or even today), Alex went on to change his outlook as completely as any person can.

In 2016, as he watched the cryptocurrency boom from his office at a proprietary trading firm in London, he realized his initial judgment had been mistaken. The math of crypto still fascinated him, just like the math at his day job — but now he saw it as an “unusual and extraordinary opportunity” to trade an asset with no historical precedent.

Alex wasn’t the first finance professional to go all-in on crypto. Bankers like Arthur Hayes have been exploring Bitcoin since well before the recent boom. Still, Alex had one skill that set him apart. He’d spent the last few years learning to create financial algorithms — code that watches the markets and generates trades to take advantage of every passing trend.

Alex is humble to a fault. Every word he says feels carefully calibrated to avoid exaggeration or self-aggrandizement. But the facts of his brief career in crypto speak for themselves. Though he started with a set of small, conservative bets, Alex’s accounts now log tens of millions of dollars in daily trades. He may be the single most prolific algorithmic trader in the markets — but he’s also the first to insist that the “most” in that title is impossible to prove.

Still, whether he’s first or second or third, this kind of success — built from thousands of tiny profitable trades, rather than a few lucky guesses — doesn’t come by chance. Alex describes his particular talent as follows:

“In terms of understanding how these markets fit together, and how opportunities arise, especially on short-term timescales — I suppose that I qualify as an expert.”

I’d go a bit further and call him a world-class trader: Someone who, time and time again, has managed to wrangle money-making rules from some of the world’s most chaotic markets.

Until now, he’s also been someone who never mentioned his success in public. But last week, he agreed to sit down with me and discuss his move from standard stock trading to the frontiers of finance.

Aaron Gertler: You majored in mathematics at Cambridge. What led you from math to your first job at a hedge fund?

Alexander Gordon-Brown: “I became interested in trading at university, because I thought it might cater well to my strengths. Building mathematical models comes naturally to me. But even more important for trading is quick analytical reasoning activity. It’s pretty similar to an intense strategy game, and I really enjoy those.”

AG: You were doing very well as a trader in your previous job. How did you decide to make such a dramatic switch, into an asset with which you had less experience?

AGB: “I had built up some savings from work, and I wanted more control and flexibility over my time, priorities, and the nature of my work as a whole. I was browsing Facebook one night, and someone had linked to an article about inefficiencies in the crypto market, so I decided to take a closer look. After that, I just tried out trading, first with money that I could afford to lose, then with larger amounts as I got more confident.”

AG: Did you start out well, or was there a steep learning curve before the trades began to work out?

AGB: “I started out with Bitcoin futures, nothing too complicated, and I think that my initial ideas were well-conceived. I generally profited from the beginning. That said, I did make quite a few mistakes starting out, which were more ‘missed opportunity’ than ‘losing money’. I expected to see certain efficiencies in crypto that would have existed in any other market, and later discovered that those didn’t yet exist.”

[Throughout his trading career, Alex has specialized in “market-making” — offering to buy and sell a wide range of assets, making a profit on the spread, and creating liquidity in markets, giving other participants greater flexibility with their own trades. Smaller markets have less liquidity, and therefore offer greater opportunity for market-makers like Alex to profit.]

AG: You probably won’t tell me, but I have to ask: What’s the secret to your success? How have you been able to scale up so quickly?

AGB: “I focus on doing a large volume of trades but keeping my positions small. Many of my trades make small amounts of money and make up for their size by sheer numbers. I don’t want one bad call on a larger position to overwhelm the upside from the many small trades.”

AG: What does a day in your work life look like, given that you manage algorithms rather than making trades yourself?

AGB: “It’s a continuous cycle. I spend some of my time keeping an eye on the positions I’m holding, to make sure the algorithms don’t run off and do something unreasonable. With automated trading, it’s easy for a small mistake to become a big mistake rather quickly.

“I also try to keep up with the wider world of crypto. It’s been a busy few months! A lot of the news that’s come out recently has had effects on pricing, so I find myself checking the news cycle continuously. At my previous job, I worked mostly when the London markets were open, but crypto markets are always open. The hard thing is finding the discipline to switch off for a while and stop working, since I know I’ll just have to catch up again.”

AG: When you’re checking in on your algorithms, how do you know when it’s time to tweak something?

AGB: “When a position starts to build up quickly for no apparent reason, or when an unusually large volume of small trades pops up — again, for an unknown reason. Whenever I see a major algorithmic decision, I need to try and match it to some change in the world. Usually, there’s a clear reason — a coin’s price has increased, for example — but if I can’t find the reason, that’s a sign something strange is going on.

“Fortunately, most instances of weirdness are actually false positives. It looks like something strange is going on, but the algorithm is justified, because it knows something you didn’t.”

AG: Do your algorithms mostly change themselves according to automated rules, or are you generally changing them by hand, in response to your own beliefs about the market?

AGB: “It’s closer to the latter. I try to write code so that the general structure is very flexible, and that I’m mostly just changing parameters. I’ve set things up so that I won’t have to write a lot of additional code every time the world changes.”

AG: And when the world does change, how do you choose what gets your attention? What kinds of news are most important to the way you work?

AGB: “I try to pay attention to the things I think other people will pay attention to. Those pieces of news usually signal times when trading will be particularly good, or when it might be a good idea to shift my sleep schedule so that I’m trading when there’s a lot of market activity. I may have my own beliefs as to which types of news and opinion are most reliable, but if lots of other people who trade decide the news is important, it will be an important day.”

AG: Given the huge boom in ICOs recently, how do you approach the release of new coins you might want to trade?

AGB: “I generally haven’t been too keen on trading newer cryptocurrencies in any big way. I want to stick to things I think are relatively well-established, and where I understand what’s going on. I won’t continue to totally ignore them forever, but right now, they’re not something I pay much attention to.”

AG: Algorithmic trading gets a lot of bad press from people who see it as unfair, or somehow taking advantage of human traders. But in an interview about your previous job, you mentioned that you see algorithmic trading as a positive thing, since it helps markets become more efficient. Do you think the same thing about crypto markets?

AGB: “I think that [arguments about efficiency] actually apply more strongly, since cryptocurrencies are even more inefficient. In crypto markets, I see both the opportunity for algorithmic traders and the value for other traders as being greater.

“Most people who want to trade Bitcoin don’t want to get set up on every exchange so that they can sell on whichever one gives the best price at any one time. They want to sell on any given exchange and get a reasonable price that isn’t too far out of line. And that’s what algorithmic trading allows them to do. Without it, you need a human to pick up every one of those trades.”

AG: Given that people with your background and talents are present in cryptocurrency markets, would amateurs be better off not trying to trade at all? How would you recommend someone approach the market if they don’t have the time or programming skill to follow your path?

AGB: “If you just want to invest in cryptocurrency for the long term, just buying some set of them and forgetting about it seems like the right plan.

“But if you do want to trade every day, focus on the things computers are bad at, and make sure you really understand what you’re doing before you stake any money that you can’t afford to lose.”

AG: What are computers bad at, when it comes to crypto trading?

AGB: “Algorithms are still fairly bad at processing news. You need to choose which stories to trust and how much to trust them, and that’s an area where algorithms struggle, even in stock markets that are much more mature.”

AG: Do you have any idea how common algorithmic trading is within crypto markets? Do you know of anyone else who operates similarly to you?

AGB: “I suspect that other people like me exist, but I don’t know anyone in particular. If you watch these markets behave, once you know what you’re looking for, you can see that other people are doing some amount of algorithmic trading.

“For example, you might see an order of an exact size being executed the same way every second. There might be someone sitting in a chair and pushing the same button very precisely, but… it’s probably a computer.”

AG: You’ve grown your trading to an impressive volume after a relatively short time. Are you still seeing rapid growth in how much you trade? Do you expect to hit a natural limit, as you pick most of the low-hanging fruit you can reach with the types of trades you make?

AGB: “It’s possible that I’m currently at a local peak, especially since some of the ‘frothy’ interest in cryptocurrencies from three months ago has disappeared, and overall trading volume has decreased.

“But a couple of exciting-looking events in the crypto world could raise volume a lot, as people come back to the markets. There have been other Bitcoin crashes, but that’s never been a permanent trend. I don’t expect it to be permanent this time, either.”

Featured image courtesy of Shutterstock.

Author:
Aaron Gertler is a freelance writer and cryptocurrency dabbler. When he's not writing, he spends his time advising philanthropists on their charitable donations and reading everything under the sun. You can find him on Twitter or on his personal website.