What is Algo Trading ?

In today’s environment, everything going automated based on machine learning, thus allowing work to be completed so fast that human is not able to accomplish. Same goes to stock market trading with help of a new podium called Algo trading. Algo trading stands for Algorithmic Trading also called Black Box trading, algorithmic means set of rules and algorithmic trading simply states automated trading. Now let’s figure it out how it works.

At the retail level, if we do trading in stock markets we need to manually examine which stock we expect to rise or fall in very next minute, hour, day or week and based on this we execute our trade. But doing so, takes lot of time of a trader in performing the task of scanning and executing the trade and by the time trade gets executed the opportunity wiped out and he left with losses. Algo trading helps us in dealing with this, in algo trading, a trader writes down his trading strategies in a computer programming language with the help of a programmer, these strategies could be in the form of Quantitative, fundamental or technical analysis. It is just like coding in VBA macros in Excel to execute the task much faster. Once the trader successfully wrote down his trading strategies, he back-test his strategies on previously available stock market data and check as if he would have been used these strategies in past with algo trading, would it be successful. If yes then he deployed the same algo trading machine into a live market and the algo software start scanning the entire stock market based on the strategies fed into the computer program and start generating profitable trades. Believe me, this entire process goes completed in milliseconds and the algo keeps on scanning the stocks and making the trade until it does not reach its prescribed limit. The power of algo trading is limitless it can make even thousands of trades in just one second.

Algo trading was not allowed in India till March 2008, afterwards, SEBI allowed the same in India. You might be surprised to hear that around 50% trades in India are executed with the help of Algorithmic Trading and the major players who use algo trading are Investment banks, Hedge funds, Asset management Co. etc. The algo trading is a very sophisticated process and is not reachable to hands of the retail traders because it involves a good amount of cost. This is the reason the retail traders fail to generate profit in trading on a consistent basis. You might be heard from a retail trader that “when I buy stocks it goes down and when I sell the stock it starts going up” this is simply because the majority of stock market trading is captured by a sophisticated algo trading computer programs and the retail trader are fighting with these machines to make the profit. Obviously, the retail trader will lose. One of the reasons why retail traders lose in the stock market is that they trade with emotion and sentiments, a retail investor does not off-load his position when he starts getting loses and immediately square-off when earned a small profit, letting his losses to get wider while narrowing the profit. On the other hand, these hard-core algo machines do not trade with emotions they strictly trade as per the strategies written in their computer programming language, these machines also play with retail traders’ emotional behaviour and generate profit. For example, in the recent past on 18th Dec, 2017 the day of Gujarat election result, market made a gap down opening and goes all the way 867 (SENSEX) points down just with the fear of BJP losing the election. Therefore all traders even investors start selling stocks based on their emotional sentiments and algo trader started making money because there were no human behaviour involved and if you check the BJP won and the market closed above 138 points after recovering the 867 points. This is how Algo trading works and generate profits. Indeed algo traders also lose, in fact, if you check, their loss happens to be in millions but it happens not because of algo trading inability but with the wrong programming fed or perhaps not back-tested with the past data.