Quant trading is taking over Wall Street. Gone are the days of discretionary traders manually entering trades. According to Knight Capital Group’s 2015 annual report, algorithmic trading has grown 29% from 2014 to 2015 amongst the 25 largest U.S. asset managers. Average daily U.S. equity share volume went from 43% from the 25 largest U.S. asset managers to 72% in 2015 (see “Algos rule,” below).
Algorithmic trading will account for almost a third of total currencies trading in 2016 according to GreySpark Partners (see “Swapping traders for software,” Below). Some analysts say that the rise in forex algorithms is due to new government mandated oversight globally. This transition was ushered along due to currency manipulation cases from New York to London starting in 2013. These cases shined a light on the over-the-counter world where despite the movement in regulated markets toward electronic trading, they operated as a call around market well into the 21st Century (see “Put down the phone,” below).
Futures have also seen added activity, so that now algorithmic and quantitative trading account for a growing portion of overall trading activity (see “Trading plan,” Below).
It is in this atmosphere that recruiters for high frequency proprietary trading shops and banks must find tomorrow’s trading all-stars and build the next wave of trading innovation.
Aside from steering away from human corruption, part of the allure of algorithmic and quantitative trading is the elimination of emotion and irrational, yet natural, human behavior that is brought on by discretionary and manual trading. This is where algorithms come in. They provide an opportunity to eliminate emotion from the trading decision-making process. Orders are often preset and executed based on statistical arbitrage triggers, historical price points and spread relationships between what sometimes may look as uncorrelated securities to outsiders.
As a result, there are many competitors in the race to build meritocratic, quantitative and algorithmic funds and platforms. Two such companies are Quantopian and Quantconnect. Quantopian, the larger of the two in both resources and membership, is now partially backed by Steven Cohen after his $250 million investment. Quantconnect was started by three founders and also provides backtesting features along with free tick data. This small startup has grown from three founders to a pool of more than 30,000 quant/algo participants. Both models follow an interesting talent recruitment concept where there are limited barriers to entry. Basic computer programming skills and trading concepts must be attained by participants and novices are helped with tutorials and through the ability to adjust publicly shared strategies in order to progress into profits and viability.
Anyone, self-taught or college educated, can create an account and compete for capital allocations and retain 10% of profits while being fully funded by Quantopian.
Aside from Hedge Funds and quant meritocracies, other participants have risen to cater to this growing phenomenon, which has blossomed from the days of Richard Dennis and Bill Eckhardt’s Turtle Trading programmers as far back as 1983. Quant participants such as consultants, analysts and talent recruiters can help Institutional traders refine their strategies and also recruit new talent.
One of the most renowned Quant consultants is Haim Bodek, CEO of Decimus Capital, who first became intrigued with trading in 1996. While working on credit card fraud forecasting he began to think about applying those same predictive models to finance. Armed with a curious mind and a bachelor’s degree in Mathematics and Cognitive Science from the University of Rochester, he was able to land a job at Hull Trading, one of the top option firms in Chicago. Over time, Bodek learned that statistical processes needed to be molded by trading thought processes to be effective. In other words, basic statistical methods must be refined by consultants or traders so that big data is not wasted on ineffective trading models. Data and programming alone cannot account for a winning strategy.
For this reason, Bodek advises that job seekers/students out of college, should learn statistics, and have a strong background in computing.
“Generally, in this day and age even if you are planning to go down a traditional trader track, I recommend two things: One, make sure you are being schooled in some quantitative discipline. If you are not going to go into quant trading directly, where I recommend a stronger background in computing — you are going to want to have minimally, a minor in statistics or a [related discipline]. You are going to want to have a strong enough background in technical fields. And the second thing is, and this is probably more for the quants, is to trade your personal account,” Bodek says. “Think of it as half of the learning experience. A lot of quants come right out of a PhD program, they say they want to work in finance and they have never traded an option. You know, you can read so many books about it, but if you walk into an interview and you’ve never put a single dollar behind some the most basic products out there — futures, options, single stocks — it basically shows.”
“Let’s say it’s a sign of a person who is really interested in becoming deeply involved in the business. Most people who like the business and thrive in the business are people who started trading before they had their first interview.”
Bodek is currently advising and consulting institutional trading companies while providing useful insight into how one can start the journey of attaining a job in quant trading. Bodek highlighted the importance of strategy fit and also culture fit.
“Well, many of the different firms have different approaches. The firm I started out with was very much about building statistical models that could pick up the small amount of edge per transaction and the founder, Blair Hull, was legendary,” Bodek says. “He started out counting cards in black jack before he moved over to options. That kind of approach really kind of colored the trading culture. One of the competitors was a pit-like trading firm; they had more of a “show ‘em down” type of approach. Their trader training cinsisted of going to Vegas and they were literally all about poker and bluffing. The thing about it is that a lot people get into the business having some kind of presumptions of how things operate and really, you have to adapt and find a culture that fits your personality.”
Choosing the right company to work for is a tremendous factor in a trader’s success and development. Most of the time traders go on interviews hoping they get selected, however, traders must keep in mind that if they are truly confident of trading for a living, it is they that must do the choosing.
“Some people do well at firms that are much more black box, other people do better at firms that are a little bit more inspired by trading culture,” Bodek says. “And the culture and the strategies definitely do tie together. There are different ways to trade the market and there is a lot of diversity out there and that’s pretty much how it all works.”
Another market participant who is unique and valuable in their approach to this new wave of quant and algorithmic trading is ViableMkts, founded by Chris White. QiableMkts is a strategic advisory group with a focus on financial technology and fixed income market structure.
“We have clients who come to us who are looking to explore quantitative trading techniques in fixed income, and we can give some insight into what it takes to apply some of those techniques in the fixed income space,” White says.
White stresses that knowledge of market structure is extremely important in his line of business.
“That’s really important, especially on the programming side where you’re on the search for the next automated market opportunity or automated trading opportunity. Being able to figure that out is really grounded in your understanding of market structure,” White says.
Laks Narayan, CTO at ViableMkts, is an extremely valuable asset to ViableMkts’ clients. His extensive knowledge in programming languages along with experience in building trading platforms from scratch can help their fixed income clients innovate and outsiders understand how the technology works together. Narayan advises that Excel, C++ and Python are all necessary for traders and consultants as important technology platforms and programs.
“When I hire people into technology, how well they understand finance, financial libraries and how they are built across different backgrounds [is vital],” Narayan says. “For example, you write a financial library for pontification of the brief on equity options. Now that can be called from the front end, the web-interface, from a C++ based trading platform [or] from Excel.”
He points out that Excel is a very important and underrated skill that people need to understand. “It’s really the computation of calling the financial libraries, keeping the complexity under the hood, this is really important. For this task, C++ now [and] a lot of Python is being used,” he says. “Not as much C, but C++ and Python are being used. When it comes to banks, building their platforms, there is a lot of Java that’s still being used. So Python will really be coming to the smaller business space, the larger banks will still hold a lot of Java. A lot of older systems are C++.”
Polly Chan and Justin Cheng from GTP Talent Search provide further insight into recruitment and what banks and other financial institutions are looking for in traders. A recurring theme amongst all three groups was the necessity of C++ and Python. Cheng provided some clear insight into what exactly high end trading recruiters are looking for.
“Most of our clients are looking for talents that have STEM (Science, Technology, Engineering and Mathematics) backgrounds, preferably with PhDs or Masters and also top GPAs. Proficiency in at least one programming language is almost a must, with a high preference towards Python and C++ for research and development purposes,” Cheng says. “Most trading companies use Python, R or Matlab as the programming language for strategy research purposes. For software / strategy development, most companies use C, C++ or Java.”
Much like Bodek, Cheng and Chan advised that knowing different trading companies’ strategies and organizational setup is helpful to the potential employee. GTP Talent Search also emphasized intangible qualities that a candidate must possess if they wish to be successful in their job search.
“Candidates must have a passion for trading and the financial markets,” Chan says. “Ability to handle stress and make quick decisions in an ever-changing and fast paced environment are essential qualities. Excellent communication skills are also a necessity, as problems are solved through the collaboration of highly motivated and intelligent people from around the world.”
From Quantopian and Quantconnect to job searching with experts such as Chan and Cheng, new entrants into the field of Quantitative and Algorithmic trading have many choices and options. For more experienced candidates, opportunities arise with consulting and advisory firms such as Decimus Capital and ViableMkts. The continued growth of this automated, computerized field depends, ironically, on the people in the field along with their ability to create and program alpha.