skip to main content
Uncategorized Posted by Team Topstep April 1, 2023

Systematic vs. Discretionary Trading

Before delving into the world of trading, a trader must decide on an approach to seek profits from the market. Traders will need to decide between a systematic approach—one based on specific rules that determine markets and entry and exit points—or a discretionary approach, which may include technical data and rules but where the trader maintains the final decision on a trade. Here we study whether systematic or discretionary trading styles yield superior performance results. In examining this question, we will discuss some of the basic characteristics of each and how they relate to return drivers, strategies, trade generators (fundamental or technical) and sectors. Additionally, we will analyze whether the benefits of diversification extend to styles.

There are numerous variations of both systematic and discretionary trading strategies. Many strategies incorporate both to varying degrees and could be defined as a systematic strategy with discretionary overlays or a discretionary strategy with systematic overlays. The one overriding difference between the two styles is that systematic trading strategies generate a definitive signal, whereas a discretionary strategy allows the trader to make the final call on price and time.

Most people’s perception of systematic and discretionary trading may have been more accurate 20 years ago. Nowadays, systematic strategies are not just simple strategies trying to exploit price and time data. Most are an amalgamation of trading rules that attempt to capitalize on momentum, mean-reversion, carry, value or other return drivers incorporated with well-defined risk management parameters.

Many systematic strategies now incorporate fundamental data; some are complex algorithms trying to capture noise, such as high-frequency programs, or even particular word groupings inputs from social media. On the flipside, discretionary traders are not simply traders trying to capitalize on a “gut feeling” derived from “divining” the fundamentals and/or price action. Most discretionary traders have well-defined trading strategies that assess and evaluate the supply/demand fundamentals. Those that have the wherewithal to survive for a reasonable or long trading history invariably have specific risk management parameters. Nearly all discretionary traders incorporate technicals to enter and exit the market, to implement risk management, and to varying degrees, to confirm fundamental evaluations. One caveat: A discretionary trader is more apt to exit a position prior to his technical parameters if the trade is not behaving to price action expectations.


In this analysis of systematic vs. discretionary trading we will examine the BarclayHedge Systematic and Discretionary Traders Indexes, the Barclay BTOP50 Index, the Barclay Agricultural Traders Index and the Société General (formerly Newedge) Macro Trading Indexes (Quantitative and Discretionary).

Let’s start with an examination of the BarclayHedge Systematic Traders Index and Discretionary Index. Not surprisingly, the number of programs has grown steadily in the period covered (February 1987 through May 2017) and now stands at 409 in the Systematic Traders Index and at 106 in the Discretionary Traders Index. BarclayHedge defines its Systematic Traders Index as programs that make decisions using automated systems in at least 95% of cases. Their Discretionary Traders Index includes programs that make at least 65% of their decisions in a discretionary way.

“Head to head,” (above) shows a VAMI (value-added monthly index) chart and a risk table of the two indexes. VAMI is the growth in value of an average $1,000 investment. VAMI assumes the reinvestment of all profits and interest income.

As one can see, there is very little difference in the returns of the two sub-indexes and 50/50 combination. This is somewhat surprising given the major differences in approach, the different trade generators, the numerous strategies and the number of markets traded by the different styles. What this presents is that the Discretionary Traders Index has significantly higher risk-adjusted returns as reflected by the Sortino Ratio. Given the similar returns, this difference stems from the greater volatility of the Systematic Traders Index, particularly to the downside. The Kurtosis numbers are huge and deserve some study.

Kurtosis and skewness are statistics that quantify the shape of a non-normal distribution. The higher the kurtosis, the fatter the tails and the greater the probability of a wider distribution. For now, let’s attribute the high numbers to the fact that these are indexes rather than individual programs/strategies. Skewness indicates a propensity of direction of the surprise (fat tails). What the kurtosis and skewness statistics indicate is that all three have extremely fat tails with a strong propensity to surprise in positive returns with the Discretionary Traders Index having the most extreme numbers for positive surprise.

The correlation between the two indexes is a relatively high 0.62. The degree of correlation is high enough as not to overcome the substantial difference in risk-adjusted-returns between the two. Consequently, 50/50 portfolio fell in between the two indexes in risk-adjusted returns but did yield the highest annualized return by a small margin.

Despite the similarity in returns, systematic traders have higher volatility as noted the risk metrics (see “Drawdowns, above). Broadly speaking, one could describe the Discretionary Traders Index’s returns stream as slower, smoother and generally less volatile that the Systematic Traders Index, though yielding similar net returns. The result being better risk-adjusted returns. The 50/50 Portfolio did produce lower drawdowns.

A Deeper Dive
As we mentioned above, all systematic strategies generate trades based upon technicals, even those that incorporate some fundamental data into their program. While the vast majority of discretionary traders generate trades based upon fundamentals, most incorporate technical analysis for entry, exit and risk management. Both styles are practiced throughout all time frames. Talking of particular strategies and sector foci, there are areas of overlap between the two where one style or the other is the predominant approach. The reasons for this stem from what return drivers – carry, momentum, value, etc. – the program is trying to capitalize upon, as well as what is the most efficient or pragmatic approach to do so.

For instance, there are practical reasons why the majority of trend followers are diversified and systematic. As mentioned earlier, technical analysis is the trade generator for systematic strategies. Trend followers focus on exploiting momentum. Many programs may have contra-trend or mean reversion components but are predominantly trend following. In general, markets do not trend for the bulk of the time. Trend traders typically take many small losses hoping to catch a few good trends that more than compensate for the multitude of small losses.

Statistically, the odds improve when applied to a greater number of markets. It would be difficult to have a deep understanding of, analyze and monitor the fundamentals—whether supply/demand or macroeconomic factors—of 25 to 75 markets. Hence, a systematic approach to trend following in a broad array of markets is the most proficient way to capitalize on the return driver of momentum.

On the other end of the spectrum are discretionary fundamental traders. These traders immerse themselves in the fundamentals of select markets — doing deep dives on a wide array of factors. Discretionary physical commodity traders are a good example of this. Like other fundamental traders, commodities traders analyze a wide variety of fundamental factors. The majority of discretionary hard commodities traders seek to exploit the return drivers of carry and relative value as well as momentum.

Frequently, these are opportunistic in nature. A major reason that these markets lend themselves to this approach is the more complex cost of holding these types of asset classes. In addition to a financing rate – the cost of holding a financial asset — there are the costs of storage, transport and insurance. Additionally, the benefit of holding the asset — the convenience yield — varies between market participants. Many discretionary fundamental traders are niche traders and focus on just one market or market sector.

Jesse Blocher, Ricky Cooper and Marat Molyboga explain why physical commodities lend themselves to the fundamental discretionary approach to trading in their 2015 white paper, Performance Persistence in Commodity Funds.

They describe how commodity traders have the opportunity  to gather more information from diverse sources than equity traders and can apply that expertise. They note how skilled traders who understand the nature of data can exploit overreaction to early production forecasts and limited weather information by exploiting more sources. They can also find data on seed and fertilizer purchases that provide them an informational edge.

The paper states, “In some ways, commodity analysts may actually have some advantages over equity analysts. Publicly traded equities have significant requirements for disclosure, but often the most important information about customer demand and supplier capacity is labeled ‘inside information,’ on which it is illegal to trade. In contrast, commodity funds are legally able to gather important information about supply and demand, and deploy it in profitable trades.”

The paper describes a number of fundamental market factors in commodities markets that underscore the importance of being able to tap into a vast flow of information. Consequently, physical commodities provide unique opportunities for a trader to gain a competitive edge via fundamental analysis to capitalize upon carry and relative value. The carry trade is affected by the interaction between supply/demand and the cost of holding the asset. As opposed to a value play in equities, relative value seeks to exploit mispricing in relative values rooted in geography or related commodities, WTI vs. Brent crude oil, for example, or Soybeans vs. soymeal.

Big Systems vs. Small Ag Traders
Given the value of fundamental information in trading physical commodities, let’s compare the Barclay BTOP50 Index and the Barclay Agricultural Traders Index versus the same period (see “BTOP vs. Ags,” above). Neither index has specific parameters for style, but the BTOP50 is composed of the 20 largest Commodity Trading Advisors (CTAs) by AUM (two-thirds are systematic diversified trend followers). The bulk of the 38 programs currently in the Barclay Agricultural Traders Index are discretionary fundamental traders.

The spread of the returns is greater than the Systematic and Discretionary Indexes but still minimal. The BTOP50 Index returns are somewhat greater with very similar volatility levels resulting in slightly better risk-adjusted returns. The kurtosis and skew statistics for the BTOP50 indicate moderately fats tail with the moderate propensity to surprise with positive returns. The same for the Agricultural Traders Index indicate very fat tails with a high propensity for positive surprises.

The correlation between the two indexes is 0.23. The 50/50 combination of the two resulted in the return falling in between the two but with the risk-adjusted return significantly higher. The possibility for positive tail events fell much closer to the Ag Index, so a strong possibility of positive returns exists.

After reviewing the drawdown data from the BTOP50 and Ag index (see “Big vs. small,” above), one can see that the drawdowns of the Agricultural Index are longer than the BTOP50 and moderately deeper as well. Run-ups are also greater and of slightly longer duration. The 50/50 portfolio significantly reduced drawdown depth and duration relative to the two individual indexes.

What is interesting is that the risk metrics are dramatically improved by combining these two indexes. This likely due to the lower correlation as not all of the components of the BTOP50 trade ag markets, and those that do trade them from a systematic technical perspective as opposed to the vast majority of programs in the Ag index, which trade from a fundamental discretionary perspective.

In the second part of this series we will look at the Société General indexes and analyze what we have learned from all of the data.


About the Author

 

Mike Dancey, CAIA, is VP of Institutional Services and Head of Research at Managed Account Research Inc.  He consults on CTA selection, due diligence, portfolio construction and integration.