Automated trading uses an algorithm to follow predetermined entries and exits to trade faster than human traders, but it takes more than a computer to make it work.


The first feature-length science fiction film, 1927’s Metropolis, had an ambivalent view of technology. On the one hand, the visual effects of the futuristic, industrialized city run by automated machinery were stunning. But on the other hand, the film explored the anxiety of robots eventually replacing people in a not-so-flattering light.

Nearly 100 years later, technology continues to evolve and substitute some of the ways we work. Still, not all of these changes come with the ominous sci fi tone. In fact, automation in trading can be helpful to increase liquidity in the market. And although a robot completely replaces one woman in Metropolis, automated trading systems still rely on trading skills and technical knowledge from humans. That means the technology can’t completely replace people (. . . for now).



Automated trading is when an algorithm follows a defined set of entries and exits as set by a human trader. Also called algorithmic trading, mechanical trading, or algo trading, this type of trading can trade faster and more frequently than manual trading. Automated trading has become increasingly more common in the past few years.

The rules that the algorithm follows are typically based on price, timing, and a technical indicator. For example, a trader can set the program to buy a number of shares for a stock once it passes a moving average, and then sell if it falls below that moving average.



As convenient as it sounds to press a button and let the algorithm do all the work of trading for you, automated trading systems still require some expertise. To start, using an automated trading system relies on at least a basic knowledge of trading. You should be able to identify your risk tolerance and form a trading plan that works best for you before setting the entry and exit criteria.

There are several different strategies traders use with automated trading. Some of these include:

  • Trend following (also known as trend trading), which is when the trader buys an asset when the price trend goes up and sells when it goes down. Because this strategy doesn’t involve making any predictions or price forecasts, this is one of the simplest strategies for automated trading.
  • Mean reversion (also known as trading range), which assumes the high and low prices of an asset are temporary and will revert back to the average prices. Trades programmed for mean reversion happen when the price of an asset goes above or below the mean price.
  • Volume-weighted average price (also known as VWAP), which tells the algorithm to break down a large order of stocks into detailed smaller orders and forward them into the market at the targeted points near the VWAP. 
  • Arbitrage opportunities, which direct the automated system to find and purchase an asset that is priced low on one market and then resell that asset to another market at a higher price.



Human traders can be meticulous and proficient, but even the most diligent traders can’t afford to sit in front of their screens all day. On the other hand, algorithms can watch the markets 24/7, meaning automated systems can generate orders the moment the markets meet the predetermined trade criteria. Between the amount of time the algorithm is monitoring the markets and the speed with which it can generate orders, this pace can make a noticeable difference in a trade’s outcome.

Automated trading can also offer backtesting, which allows traders to download historical price data from exchanges and see if their strategy would have worked in the past. This means you can experiment with your trading plan without risk. However, past performance does not guarantee future results.

Finally, the algorithm is free of one of the most human traits: emotion. An automated trading system can stay disciplined more easily without the temptation of revenge trading or adrenaline-fueled decisions.

Automated trading does come with some disadvantages, though. Since it works on a predetermined set of parameters, the algorithm can’t make instant market decisions based on new information. 

Moreover, much like anything else performed by technology, automated trading systems are still subject to technical errors. If there isn’t a person monitoring it, the system could lose internet access and stop working without the trader’s knowledge.  

And the algorithm can only be as good as it is programmed to be. A coding error caused Knight Capital to lose $440 million in 30 minutes back in 2012, proving the confidence to rely on a poorly tested system can be devastating. Scrutinizing your software and developing your algorithmic skills can minimize costly errors. 

 One more notable disadvantage of automated trading are the high costs that represent a barrier to entry. Programming, hardware and low latency data lines can be expensive for casual retail traders.

Depending on your goals, automated trading can be a helpful trading tool. But even the most advanced algorithm depends on research, planning, and technique from human traders.