They’re a different breed.
Today’s traders aren’t just people that know how to read or get the “feel” of the market and trade it, rather they’re specialists with entire skill sets. It’s not enough to graduate Wharton or the University of Chicago with a finance degree, instead one needs to study quantitative theory, data mining, financial engineering and the like.
“It is clear that the fundamental processes of the buy-side and sell-side are becoming more data-driven and there is no question that the value and necessity of data is growing,” said Jack Miller, Head of Trading at Baird in New York. And he, like others in the business, must be eady an able to digest reams of it in order to execute properly.
To better unerstand the sea change, please check out the below article – originally published in the October 2010 issue of Traders Magazine.
Upping the Ante
A Competitive Job Market Pushes Traders to Beef Up Their Analytical Skills
By John Hintze
The trader of tomorrow will be less a seeker of stealthy executions and more like a highly skilled quality-control engineer, finely tuning his trading platforms to adjust to changes in an increasingly complex market.In fact, steadily improving technology and a growing emphasis on quantitative analysis appear likely to transform the trader’s traditional job of executing strategies for portfolio managers and buyside customers efficiently and anonymously.
“This is the real problem for traders: They’re becoming the equivalent of specialized buggy-whip artisans-they have very finely honed skills, but those skills are becoming useless,” said Benn Steil, director of international economics at the Council on Foreign Relations and the principal of Efficient Frontiers, a capital markets consultancy.
Rather than working individual orders with the array of electronic tools now available (perhaps at times picking up the phone to find a natural match), traders in the future will spend their day processing real-time information about the current market and their own trades’ performance.
When performance dips or there’s a market-changing event, they will respond by tweaking their algorithms’ parameters, or even switching to new ones.
And so the traders of tomorrow will require a deep understanding of technology, not only for efficiency, but so that they can spot the next wave of instruments to put in their toolbox.
As today, they’ll need strong communication skills. But rather than using those skills to determine where a specific security or sector is headed, they’ll have to be fluent in programmer-speak and quantitative analysis, to understand in detail how their algos work-and differ from each other-and which of the latest developments can give them an edge.
Buyside traders will have to know enough about market structure and technology to grill sellside and vendor executives about their latest products and services. Sellside traders will need to understand more about a range of markets. And that’s just the beginning. But at the end of the day, a traders’ market feel, coupled with his ability to adapt and develop new skills will always be a requisite for success.
Still, if successful traders today hope to see their children follow in their footsteps, then their kids should start honing their quantitative skills now, so that they are grounded in mathematics and computer science as soon as possible.
Phil Krauss, director of trading at Harris Investment Management in Chicago, said he now looks to fill a junior position with someone who has a strong background in quantitative analysis. The replacement for an experienced trader, he said, must have a programming background, and preferably algo trading experience.
Krauss sees the skill set of a trader changing, and likely broadening dramatically, in the years ahead. The sellside is increasingly customizing its algorithms to match buyside customers’ specific trading strategies, and even to trade specific stocks. In fact, he said, algorithms may become so attuned to an institution’s particular investment strategy and trading needs that there may no longer be a need to distinguish traders from portfolio managers-at least at smaller firms.
“It’s going to get so automated that we will probably see an increase in the number of firms that have their portfolio managers and traders merge into one,” Krauss said.
That may be a ways off, but already professionals who previously held very separate roles at trading firms are intermingling much more. Programmers are stationed on trading desks at many sellside firms, along with desk analysts providing trading ideas. On the buyside, portfolio managers often sit with their traders.
If current trends continue, those professionals’ roles and skills are going to start to overlap, requiring traders to have a deeper understanding of how to program the algos they’re using, and programmers to better understand how markets function. If Krauss’ prediction comes to pass, many buyside traders are going to have to move beyond their technology-driven trading skill set and master the investment analysis skills of portfolio managers.
Traders have always been rapid processors and analyzers of information. In the future, however, the information they evaluate will be less about the condition of the current market in which they’re trading and much more quantitative and technical. Machines will continue to take over many of what are considered traders’ traditional skills; and traders will be tuning those machines to gauge where the price of a security is headed.
Quantitative skills are the common denominator among what Pat McCauley calls the “triangle” of traders, programmers and specialists in quantitative research. McCauley began trading at Susquehanna International Group in 1991 and now trains the firm’s new hires in a program he helped develop 16 years ago. He said that the triangle has grown much tighter in recent years, as the three groups’ duties overlap, and will continue to do so.
Michael Buek, Vanguard Group
“You’re going to need people who can do one or more of those three things, or at least understand and appreciate the other parts and be able to discuss them in detail,” he said.
Traders at Susquehanna are going to have to understand the technology and math underlying the algo models, and how they function in a trading context. “So that if something happens and the market changes,” he said, “the trader can change with it. I don’t think pure technologists can be successful at trading, and traders who don’t understand the technology and the math will have a very difficult time.”
At the front of the automation race that is changing the nature of trading are high-frequency proprietary firms, which have taken over the role of market makers. Prop trading firms are by definition insular, preferring to confront the markets on their own to avoid revealing trade secrets. They develop their own algorithms and quantitative research, with traders becoming an increasing part of those efforts.
Traditional buyside and sellside firms are scrambling to keep pace with them-unlike the proprietary HFTs, they must devote significant time tending to customers and communicating with brokers. Prop trading firms’ more traditional cousins on the buyside-money managers of one stripe or another-rarely have the resources to develop trading and quantitative tools in-house. They rely on sellside brokerages for much of their technology.
Consequently, traditional buyside traders’ quantitative and technological skills will unlikely ever be as comprehensive as those of traders at proprietary shops-but they will still have to work on building them.
The Same Language
Mike Buek, head of equity trading at Vanguard Group, noted that buyside traders were once more like administrators, assigning orders to brokers who traded for them. Trading is increasingly being done in-house now, though. And while the buyside continues to use brokerage’s technology, Buek said communication between the two sides has lessened.
“You still talk to brokers to understand how their algos work, what changes they’ve made to them and how they’ve been adapted to handle events like the flash crash,” Buek said.
Traditional buyside traders will still have to exhibit strong communication skills. But those skills are shifting away from contacting brokers for information. Instead, buysiders will have to query their sellside counterparts about the best tools to trade the security, how their technology differs from competitors’ and how it would handle unexpected events that could shake the markets.
And to ask their brokers and vendors probing questions, buyside traders will have to know the latest technology and the workings of modern markets. They will need to apply that knowledge, along with their technical and quantitative skills, to navigate orders algorithmically. And that will require knowledge about routing and order placement.
Large brokerage firms’ traders serving institutional customers will also have to hone skills similar to both prop traders’ and traditional buysiders’. Quant researchers and programmers are already sitting on trading desks at firms like agency-brokerage ConvergEx Group.
Their traders, including sales traders, have to understand existing and emerging technologies, not only to express their needs to the programmers and quant researchers developing the tools, but to explain the edge their tools provide.
In addition, buyside traders fully expect their brokers to be able to explain the ins and outs of different market centers-not just those run by the specific brokerage-as well as the impact of new market developments. Technology has also enabled trading in much greater volumes, often for multiple customers simultaneously. New technology will exacerbate those trends, requiring sellside traders to become more proficient multitaskers.
Craig Lax, head of electronic executions at ConvergEx, said his firm believes the markets are only going to become more complicated. Sellside traders, too, will have to become increasingly expert in technology, quantitative research and the markets. “It will be very hard to get involved in this industry without some level of programming, statistics, engineering and a good, solid understanding of how markets work,” Lax said.
Sellside traders in particular will have to increase their fluency in other financial markets, such as options and futures, and those overseas, he said. Clients will increasingly expect it.
“There’s a trend toward multi-asset-class clients. It will be hard to service clients properly if a trader only has a background in one asset class,” Lax said.
Still, it will be up to brokers of tomorrow to spot technological trends to keep ahead of competitors. Krauss said his firm still searches for natural block trades, but that those are now few and far between, and he expects the use of algos only to increase. And, he said, institutions will be operating those algos in-house more and more, rather than giving the business to a broker and paying full-service rates.
Consequently, Harris has been working since May with Quantitative Service Group, a vendor that analyzes the firm’s every trade, to capture its impact. That has allowed the QSG to rank the algos and brokers it uses by their effectiveness.
The next step is to get analysis in real time, allowing traders to adjust strategies on the fly. “Business is going to be much more real-time,” Krauss said.
Market Feel Counts
I think of the modern trader as a cyborg, using a lot of technology to harvest data, but with a human understanding of the markets and how to manage that technology,” said James Angel, an associate professor at Georgetown University’s McDonough School of Business.
Despite the growing importance of technical know-how, Angel said, the ability to multitask and, perhaps most important, an understanding of how markets work will remain vital. He worries that market experience may be lost in the age of automation. To combat that, he suggests reading not only popular books by the likes of Michael Lewis, such as “The Big Short” and “Liar’s Poker,” but classics like “Reminiscences of a Stock Operator,” Edwin Lefevre’s 1923 account of renowned speculator Jesse Livermore.
For its part, Susquehanna International Group, a long-time market maker, continues to put its new hires-typically straight out of school-through the rigors of a simulated open-outcry trading floor.
McCauley believes learning to establish prices and to trade in a seemingly antiquated environment is vital. “You have to aggregate information, process it yourself, update your prices and trade against prices in the mock trading crowd. That has a lot of value,” he said.
And because information on which to base trades is nearly always incomplete, traders’ intuition will still come into play when a quick decision is needed about which tool to deploy. As Vanguard’s Buek put it: “I know what happened the last five times I used this broker or algorithm, and now I have to make a call.
“We’re not talking to as many people, but traders are watching what’s happening in the market to make judgments based on experience,” Buek said.
Traders will increasingly have to monitor market conditions real time and adjust algos appropriately-or perhaps switch to an algo with a different strategy, or a different financial market, such as futures, he said. “It’s not like traders will set the algos and forget; they’ll always be monitoring and tweaking them.”
McCauley, too, was quick to point out that though SIG’s trading-floor simulations are important, the nature of the information and methods used to determine prices and trade against other market participants has changed dramatically.
Fifteen years ago, exchange floors were the main repositories for information. Now data comes from 60 or more market centers. Also, more market participants than ever are participating in the price-discovery process, including individual traders using sophisticated, fast systems from their brokerages.
So now the data prompting trading decisions comes from myriad sources and is harder to get-and those trends are only likely to intensify, McCauley said.
“At the core, we’re still a market-making firm that’s good at decision-making, but the tools have changed a lot,” he said. “The rigor of the quantitative and technological skills required for traders to be successful is much higher than it used to be.”
Susquehanna hires a few experienced traders but mainly grabs graduates from engineering schools such as the Massachusetts Institute of Technology and the California Institute of Technology.
S.P. Kothari, a deputy dean at MIT’s Sloan School of Management, said students entering its finance programs at both the undergraduate and master’s levels already have strong programming skills-which are not even taught at Sloan.
George Papa, SpiderRock Trading
“Many are already electrical or computer-science engineers,” he said, adding that the biggest banks-including Goldman Sachs, Morgan Stanley and J.P. Morgan-regularly solicit MIT finance graduates to fill trading positions.
George Papa, a managing partner at Chicago’s SpiderRock Trading, agrees that ever-stronger math and quantitative skills will be key for traders looking ahead. He sees high-frequency trading firms taking two paths: Either they’ll rev up their technology to trade even faster-an approach that, as with a Ferrari engine, may well result in frequent breakdowns and expensive repairs. Or, firms will focus on building smarter algorithms.
SpiderRock has chosen the latter approach. Four members of the firm’s 12-strong staff are traders with sophisticated math skills, another five are programmers, and the rest are operational staff. Papa sees a “layering effect” in capital markets developments, starting with building computer networks in the 1970s and 1980s and moving to DMA in recent years, with each layer becoming a part of the next. DMA, he added, is becoming fairly standard; next will be algos that usurp many of the skills traders use today.
“Instead of being directly connected to the market, traders will connect into market centers through layers of algorithms, which will give them tremendous power to do many things at once, instead of focusing on individual trades,” Papa said.
What will traders do then?
Search for large-scale trading ideas, he said. “If you have a class of algos that allows you to fluently execute with the market,” he said, “then you’re going to be able to think about the markets in much more of a macro sense than people do today.