High Frequency Trading: How Algorithms Trade in Microseconds

Beginner Investing • 6 min

In 2010, a firm reportedly spent hundreds of millions of dollars building a straighter fiber-optic cable between Chicago and New York — just to shave about 3 milliseconds off the time it took data to travel between the two cities. Three milliseconds is faster than a human eye can blink. That's how seriously high-frequency trading (HFT) firms take speed.

HFT uses computer algorithms to place, adjust, and cancel enormous numbers of orders in fractions of a second — often holding a position for mere seconds or less. These firms typically act as market makers, constantly quoting both a buy and sell price and profiting from the tiny spread between them, thousands of times a day.

Insider Angle: the whole business model depends on being microseconds faster than everyone else — because being first to see a price change and react to it can be worth real money, repeated millions of times a day. Critics argue this same speed advantage lets some HFT firms detect a large incoming order from a slower investor and trade ahead of it, a modern version of front-running that regulators have investigated repeatedly without a full resolution on how much of it actually happens.

Try This: Look at a stock's real bid-ask spread right now in a public options chain — that tight spread is largely a product of market-making activity, much of it algorithmic. Consider what that spread might look like without constant algorithmic quoting.

Try This — Live Data

A real current quote snapshot for SPY, one of the most heavily HFT-traded tickers.

Loading live data…

Quiz