Technology Can’t Fully Replace Human Judgement and Intuition (by Bruce Fador, Wall Street Horizon)10.17.2016
Technology now comprises a big part of today’s trading landscape, perhaps most of it. Beyond just automating what used be to manual tasks, technology has encroached well into the “thinking” parts of a trader’s job.
The key question for firms is where to draw the line. How much of the human touch should they outsource to machines?
It’s a tough call because technology delivers many advantages. With automated trading systems, for example, investors can build very precise market entry and exit rules into their models, and then use their computers to execute and monitor trades. No emotion, just “set it and forget it.”
So the more automation, the better, right? Well, not exactly.
In many corners of the market, lucrative buy and sell opportunities reveal themselves with things that aren’t technical indicators, and aren’t easily captured and fed into models. These little facts, occurrences, and nuances show up outside the realm of hard-core market data. Instead, they float around on the periphery, and often just briefly. To spot these things and connect the dots to formulate a buy or sell move, the human touch is required—the eyes, ears, and brain power of a human trader.
Corporate events data is a good example for showing both sides of this question. Algo shops use corporate event data feeds, such as those provided by Wall Street Horizon, combining them with other data, and feeding it all into their models where their algorithms go to work looking for correlations. Algorithms can process far more data and do more complex calculations faster than humanly possible. But they’re entirely lacking in judgement and intuition. They can only do what they’re programmed for – nothing more and nothing less.
Then there’s the flip side. Continuing with our corporate events example, lots of interesting things happen in this data realm. Many of them are difficult to capture in a feed and nearly impossible for machines to assess properly. When there are unique aspects to a situation, humans can call upon their intuition to make assessments; machines can’t.
Maybe it’s a company delaying its earnings announcement until late Friday afternoon, a CFO who sounded more confident than usual on an investor conference call, a suspended dividend, an added conference presentation, or any number of other subtle clues put out by companies. Technology generally won’t pick up on these items, let alone discern what they might actually mean.
Yet with access to the right information, it’s easy for a trader to anticipate what these events mean for stock and options pricing. They see one of these subtle clues, evaluate it in context, and quickly determine if it presents a trading or risk opportunity, and if so, act on it. Sometimes traders don’t even know what they’re looking for until they see it.
Wall Street Horizon tracks all forward-looking corporate events—from earnings to dividend dates to participation in industry conferences and trade shows. They also archive their history. There are over 40 types of events covered, with more categories being added, such as movie and video game release dates. The information is available in two formats: raw (for the quants) and a nice graphic interface—available on the web or on Thomson Reuters Eikon (for the traders). The quants know in advance exactly how they’ll be using the data; the traders know some of what they’ll use it for, but the rest. . . well, they’ll know it when they see it.
Human intuition and judgement are key assets in the investment community. Machines and automation are great for some things, just not everything. Providing traders with the right data will give the machines a good run for their money every time.
Due diligence process has not kept pace with the proliferation of algorithmic trading.
By Greta Zhou and Andy Cheung, APAC AES, Credit Suisse
Broker's new platform generates execution strategies that leverage client-specific trading data.
Trading platform addresses the market's need for performance, transparency and control.
Forensic analysis makes ’black box’ algorithms more transparent.