Skip to Content

Randomization and Optimization on Electronic Stock Exchanges

A trading strategy revealed in two new patents illustrates the high speed game of cat and mouse played by automated traders.

It is difficult to overstate the extent to which modern stock traders have become, in essence, cyborgs. Since the dawn of electronic stock exchanges in the early 1990s, the speed at which trades occur has allowed traders to exploit ever tinier movements in the price of a stock. In the 1980s, for a broker not on the trading floor, “immediately” meant a half hour. Nowadays, trades happen in microseconds.

Automated trading systems are ubiquitous. Automated traders have their own industry publication and - this is how you know you’ve arrived - industry-specific editorial cartoons. A patent published on an automated trading system just a few weeks ago by Trading Technologies International, based in Chicago exemplifies how these automated systems work.

System and method for prioritized automated trading in an electronic trading environment” describes a software algorithm that decides the order in which to put through a list of trades when more than one of those trades would normally be triggered after a single condition is satisfied–say, that the price of a stock drops to a pre-determined level. The patent describes what might be considered a fairly basic function of all automated trading software, which is relieving the burden of prioritizing a batch of trades that were previously queued up and are waiting for the right conditions before they’re sent to the exchange.

Prioritizing trades is inherently a deterministic process, unless you’re trading on a so-called “dark pool” where trades are invisible the details of your trades are available more or less instantaneously to everyone watching the market. If you are moving a sufficient volume of shares and another trader can predict what your automated system is going to do next, they could swoop in and take advantage of whatever directionality you’re providing in the up or down movement of a stock.

Hence a second approach, published in a patent also issued to Trading Technologies International just a few months earlier. “System and method for randomizing orders in an electronic trading environment” allows a trading strategy that is seemingly at odds with deterministic prioritization of trades. This patent describes a process that seeks to make your trades indistinguishable from the background noise of other trades so that no one can predict what you’ll do next, or where the price of a stock might move as a result. If it works as advertised, there is the possibility that no one would even know you are trading that stock.

These technologies could be seen as a great leveler. Traders employed by financial institutions and large funds are increasingly trading stock in dark pools to avoid disclosing to the world the large trades they are making. Using software algorithms like the ones described above, even everyday traders buying and selling on public exchanges like NASDAQ and the NYSE have similar power to both optimize and disguise their trades.

On the other hand, automated trading algorithms also represent a further step away from trading on a human time-scale. If day traders exploiting the tiniest movements in stock prices are something worth worrying about, what about software algorithms whose behavior in chaotic markets is ultimately unpredictable?

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.