7 Myths About Systematic Investing

The world of investing is split into 2 big schools (1) Discretionary and (2) Systematic. As an investor, you are either in one or the other. Discretionary investors makes all the investment decisions. Systematic investors, on the other hand, leave the calls to the models they developed. And then, there are the “hybrids” who uses models, but retain the discretion to make the final call. Technically, I consider “hybrids” discretionary as well.

Everyone knows what discretionary investing are. If you invest, there is a good chance you lean towards this camp. Many greats such as George Soros and Paul Tudor Jones are discretionary traders. Systematic investing, however, is not well known prior to 2008, even though they have been around for decades. The most prominent systematic figure I can think of is Jim Simons. He is the founder of Renaissance Technologies which runs the world’s most successful systematic fund — Medallion. The Medallion fund has delivered a stellar net annualized return of close to 40% for almost 30 years. This is despite charging a 5% management and 44% performance fee. If we reverse out these fees, it would put their raw performance closer to an incredible 80% per year.

If there are systematic funds with such a fantastic track record, then why are so little known about their approach? For the most part, many such funds are tight lipped about what they do. And a good number are run by engineers, scientists or mathematicians who can’t really pitch a sexy story. So as of today, many misconceptions still abound on systematic investing. Let me try to clear up the air and bust all these myths and fiction.

Myth 1 — Systematic Investing is Black Box Investing

Systematic investing simply means letting your models make the investment decision. It says nothing about the models you are using, and these models are not necessarily black boxes. They can be rule based models that uses well defined fundamental or technical criteria. They can also be quantitative models that exploit certain mathematical relations in relative valuations; or the more common factor based investing grounded in regression theory. There are many more. But each of these operate within clear and well understood boundaries. Given the inputs, you will know what drives the model to give you the outcome. These are not black boxes.

But yes, systematic investing also encompass black box approaches such as algorithms that make use of data mining and machine learning techniques like neural networks. In these models, you are indeed not able to rationalize how they arrive at the outputs. However, these comprise only a small part of the entire systematic approach.

As a side point, it is always interesting to me that while we label machine learning, which is modeled after human cognition, a black box. Humans, on the other hand, are not. Of course, we can always argue that we are still far from being able to replicate human thinking. But by the same argument, we are also far from understanding how our own brain works. So why are humans any less of a black box?

A lot boils down to trust in intuition. But for me, intuition is another word for “black box”. You can present the same set of conditions to a person more than once, but he can end up with a different decision each time. And if you ask him for a reason, he is unable to articulate clearly why. More often than not, you can only attribute this to a gut feel.

Myth 2 — All systematic strategies is 100% automated

Not all systematic strategies are automated, but they can be automated, and many are. The beauty of systematic is being able to distill the strategy down to a clear set of rules or mathematical expressions. When you have that, automation is really just a matter of choice. There is no rule to stop you from executing the trades manually. But of course, there are times when it is necessary to automate, such as when you are running a high frequency trading strategy. That is a game of speed and there is no way humans can match up to machines.

There are other advantages to automation. First, it frees up time which can be better spent elsewhere, say on research. Second, it reduces the chance of human errors or unplanned interventions. But it is not without drawbacks. With humans there are human errors, with system there are system errors. You may also become excessively detach from the market as time passes.

Myth 3 — Systematic strategy is complex

Does everything we do not know about have to be complicated? Who says systematic strategy have to be complex?

Just because no one tells you how they build their models does not mean they are complex. Of course, there are complicated ones, but strategies can also be formulated based on simple concepts.

As an example, I can have a basic rule to buy any US stocks that opens down more than 3% from its prior low, and then close off the positions at the end of the session. This is certainly not rocket science and almost anyone can implement it. (Disclaimer: I am just shooting off the cuff and I have not tested the rule) Perhaps, there is a general perception that simple stuffs do not work because everyone have already thought of it. This is not unfounded, but it is always possible to adapt or extend a simple concept into something that has a slight edge.

Myth 4 — Systematic is 100% science and Discretionary is 100% art

Investing has elements in both science and art. It does not matter which side you belong to. Discretionary is not all art, and neither is systematic all science.

Discretionary investors can also be guided by rules which they have internalized through their experience. This part is science. But these rules are often not as well-defined as their systematic counterparts, not to mention that these rules can change. This leaves a lot of room for “art” to bloom. This “art” element is shaped by each individual’s experience, knowledge, personality and is unique from the rest. This is why it is so difficult to impart discretionary investment or trading skills to another person. You can teach 10 people how to trade using your set of rules, and end up with 10 different results.

For systematic, there are also many considerations in designing a strategy where the end decision depends on the creator’s experience, preference and common sense. For example, when would you start to deploy a new strategy, and under what conditions would you retire one? How do you minimize tail risk, or do you just ignore it and blindly rely on conventional risk measures? There are no right or wrong answers and everyone’s approach can be different.

Myth 5 — You can completely decouple emotions in systematic approaches

The system or strategy has no emotions. That, I can guarantee. Even in the face of a falling mountain, it will not stray from its path. But what about you? Will you feel nothing when the strategy doubles your money? Can you remain cool when your strategy loses half your assets?

I know I am exaggerating, but the point is, you are ultimately a human. You cannot escape from emotions. So even with systematic strategies, one can never rule out “unplanned” meddling from its creators. But whatever the reason, such meddling from a systematic perspective, is undesirable. Because you are introducing an unknown element (which is yourself) into what should have been a well-tested strategy. That defeats the whole purpose of adopting a systematic approach. Hence, all systematic investors should exercise restraint and discipline in sticking to their plan.

Meanwhile, feel free to let your emotions run. Winning and losing is part and parcel of investing, and feeling happy or fearful as a result, is just being human.

Myth 6 — Systematic investors are losing all their control to machines

From the perspective of a discretionary investor, a systematic investor may appear to cede and lose all control to his models. But if you see it from the viewpoint of the models’ creator, it is entirely another picture. He built the models according to what he wants. He knows down to the nuts and bolts of the models and how they are going to behave under different market conditions (except for black box type strategies). To put it simply, these models are just doing what he wants them to do. So how can we consider that ceding or losing control? It is more appropriate to regard the models as an extension of the systematic investor himself, but one that is pure in focus, disciplined and devoid of emotions

Myth 7 — No strategies last forever, so all systematic traders will eventually fail

Does an evergreen strategy exist? I seriously have no answer. But it is prudent to assume that nothing last forever. Market regime can shift, dynamics can change, behaviours can alter. Competition can also erode profits and the effectiveness of current strategies. It is not uncommon to see funds that shine in the past gradually fade out of business as their performance falters. Of course, it can be due to other reasons like capacity or tail events, but a deteriorating strategy is definitely one of them. So to some extent, the first part of the statement is true. But it does not mean that we are destined to fail.

Post 2008 presents some of the worst periods for hedge funds. An unprecedented era of money printing created one of the most challenging environment for them. Correlation starts to break down and volatility vanishes. Some big hedge funds bite the dust. But with their fall, you see the rise of others — hedge funds that crowd source strategies, hedge funds that uses AI techniques for trading, and hedge funds that focus on cryptocurrency trading. But whether they are hype or the real thing, only time will tell.

Both discretionary or systematic investors face the same issue. That is why it is important to adapt. In order not to be obsolete, you have to keep refining your approach. You will have to monitor old strategies and be ready to retire those that are no longer effective. At the same time, you have to constantly research and develop new strategies to keep pace with the ever changing market. If life is as easy as finding a single method, and that it will work for eternity, then we will all be living happily ever after.

Hope you have a good read. Feel free to share the article!

Originally published at investmentcache.com on September 28, 2018.