Feed the algo trading bug with small capital

Do not commit more than 15% of your equity portfolio as even a small glitch can prove costly

As the stock market boom pulls retail investors into equities, a few of them are venturing into algorithmic trades. The rising interest of retail investors into computer program-based trades has prompted brokers to begin offering this facility to them. In the past few months, stock brokers such as 5Paisa, Zerodha, SMC Global Securities and Upstox have allowed retail investors to execute automated trades on their platforms.

“We have been receiving a lot of queries from retail investors for algorithmic trading since we recently launched the platform,” says Prakash Gagdani, chief executive officer (CEO), 5paisa. He says many of these investors either want to test the waters and understand how algo trades work or they have a proprietary strategy and want to use the platform to use it.

Backed by demand from brokers and investors, even the Securities and Exchange Board of India (Sebi) is planning to introduce rules for retail investors’ participation in algo trading. The market regulator is working on a white paper that would clarify the extent to which individual investors should be allowed to use trades based on computer programs, according to brokers. At present, there’s no framework on algorithmic trading for retail investors.

Faster trades and convenient: Algo trading, as it is popularly called, is a computer program that triggers buying or selling based on rules set by investors. If an investor wants to buy any aviation stock when oil prices fall, they can write a program to identify such opportunities. Algo trades are mostly used on derivatives trades for what is called as multi-leg strategies, where more than one trade needs to be executed at the same time. In a two-leg strategy, for example, the investor may want to buy a future and at the same time sell a put. Those who use technical analysis can also trigger trades based on charts and patterns. 

There have been programs earlier that helped investors identify opportunity. But, brokers didn’t offer platforms that could execute the trades on their own once the opportunity was identified. Brokers targeting retail investors now offer automated trades.

One of the biggest benefits of using algos is that it removes emotion from the trades. Experts say around 95 per cent of people lose money in day trading and nearly 70 per cent of newly opened demat accounts get closed within three months. “People start trading but when they make losses they stop trading. People get into the market, make some money, then make a huge loss, and then get out of the market. A lot of this happens because emotions come in the way,” says Ramakrishnan S, founder, Quantindia.in, a company that provides retail traders with algorithmic trading software.

As trades are automated, an investor doesn’t need to be in front of the computer all the time waiting for the opportunity to arise and then placing the trade. Program-based trading is also much faster and can execute simultaneous trades instantly. “As it’s the computer handling the orders, there’s little bias or emotion involved,” says Nithin Kamath, founder and CEO, Zerodha. The best part is if an investor has his proprietary strategy, he can back-test it to see how it worked under different market conditions.

Investors can also use algos to protect themselves from sudden market movements. For instance, the Nifty could crash 300 points in a day in as few as 15 minutes. If the machine is tracking the market, it will automatically exit his positions. But, if the trader is trading manually, his entire capital could be wiped out. 

Variety of offerings: Investors have the option to use either the exchange-approved strategies that are on brokers’ platforms or they can program their own strategy. For the latter, the investor would need to approach an exchange through the broker and get the strategy approved. Due to this, what brokers offer under algo trading slightly differ. 5paisa and SMC Global Securities have exchange-approved strategies on their platforms. To understand algo trading, new investors can use these popular strategies.

But, brokers like Zerodha don’t offer any strategy. It just offers a platform where its customers can write a program and automatically execute trades, once an exchange approves their strategy. As algo trades have caused market crashes a few times globally, an exchange essentially tests how the program can impact overall trades.

5paisa charges Rs 25,000 annually for the platform and to get strategies  approved from exchanges. If a person doesn’t know how to write a computer program, the broker helps with coding, starting at Rs 25,000.

Algorithms can also erode capital: In 2012, a US-based financial services firm Knight Capital Group lost $440 million in 30 minutes due to a trading glitch. While algorithms are the future of stock trading, a small glitch at any end — exchange’s, trader’s or broker’s — can lead to erosion of capital. That’s why it’s essential that investors spend time in understanding the technology that brokers offer. You need to have some technological acumen. If you commit an error, you should know how to deal with it.

As with any other investing style, an investor needs to spend time on algo trading to get the hang of it and fine-tune the strategy he uses. An individual may need to use the strategy for at least three to six months before he sees a meaningful profit. Don’t give up on algos in the initial few months if you don’t see returns. “Most platforms allow investors to back-test their strategies on historical data. Don’t use strategies that are not back-tested. If you are technologically savvy and can write your own codes, test the strategy against scenarios where there has been a market crash or the market has seen large, sudden movements,” says Rajesh Ganesh, 41, a Chennai-based programmer who is currently writing an algo-trading software.

As algo trading for retail investors is in the early stages, use only a small portion of your direct equity portfolio. Most brokers and analysts say investors should not use more than 15-20 per cent of their portfolio for trading in stocks based on algos.