Back to Blogs

Is Algo Trading Dead?

Published on Apr 01, 2016

How many people you know have really made any or much money, in the long run, using Algo trading systems? I have been around for almost seven years now and have met 100s of players in this field, but know only a handful of them, who have been successful!

Is Algo trading dead? Why am I writing its epitaph just when you thought Algo trading is the next big wave, which would replace any manual methods we know in trading and investment?

To put it in perspective, I would say that Algo trading is dead for people, who have either burnt their fingers badly trying to reap maximum benefit in the shortest possible time or who think that Algo trading is more like a Federal Bank machine, which will keep printing money.

First things first – An Algo system is only a pattern, which occurred in the past. Unlike a deterministic pattern such as a Google Page Rank mechanism (How many times have you got an exact match on the page you are looking at in Google? ‘Quite often’, you would say!), markets are governed by probabilistic space and hence, Algo systems are themselves prone to failure given the probabilistic component of the market. What it means is that the patterns of the past might not occur in the future with 100% certainty. If they did, they will be quickly crowded around by smart guys, till they stop working!

Hence, Algo trading ends up becoming a data crunching exercise for most professionals, who become hindsight billionaires by making tons of money in their back-tests. Algo trading finally ends like a dead dream for such professionals where the rubber meets the road and they run their systems in live markets with little or no success. I have met more people who belong to this category than otherwise!

There are less than 5% patterns, which repeat in the market with a high degree of certainty. Even with these patterns, risk management and getting the basics right sometimes becomes a challenge. Of course, there is no secret recipe on what these 5% patterns are. But I have usually noticed people, who apply their systems to varied asset classes like equities, commodities, currencies, fixed income and realize the system giving results with some amount of accuracy. People belonging to each of these classes are usually more successful.

Systems, which work very well in say commodities but are a massacre in equities, are likely to face turbulent times more than a system, which on an average has performed OKaish on both the asset classes. The reason is not hard to imagine, as commodities sometimes do lose their volatility and become like equity trading in a very narrow range and vice versa. Hence, I would put my money on systems of the second kind, where I get average returns on both commodities and equities rather than making a fortune on commodities and losing my shirt in equities.

Algo trading is the most challenging subject today in the finance world, when a system, which is supposed to be predictable and consistent in nature meets uncertain market environments on a day-to-day basis. Algos, which weather the storm, make riches and glory for its maker, while others become dead – only to be talked about in books and financial papers for academic purposes!