Skip to content

Thoughts on System Design

September 12, 2009

In the past, I have shared with you most of what I have known, explored, investigated in my short but eventful career of a systems trader. I am sure, this will be one of those periods which I will forever reminisce wherever I be, whatever I do. It includes trading one of the biggest bear markets in the recent times, especially trading months like October 2008, when fear gripped “Dalal Path” and I riding NIFTY on the short side, on one of those days when everything seems going south. I rode bourses on the long side, on one of those days when Indian markets hit a continuous triple upper circuit. Somewhere in between I went bankrupt, came back, blew myself and then hit one of the better patches of my trading life

In between, prudent risk management saw me manage a unleveraged portfolio far better than I managed, initial periods of my trading career. The reason was simple. And not surprisingly has deep insights. But before I delve in, let me recount a story.

From the ancients to the modern

There was a company which started in pre World War II Japan, with its name inspired from the Japanese word for “fertile rice paddies”. Soon thereafter it produced its first vehicles A1 and G1 and was established as an independent company.Later, in 1937 it was changed to its present name, and it was formally named as the company we know it today.But whats brilliant is the philosophy which it developed from its root in ancient Japanese wisdom. It was instrumental in developing it into a full fledged management practice.Today we know the same modus operandi by its name and business leaders all over the world study it to increase profits, reduce wastage, improve production and in general increase the “happiness quotient” of their workers.

So what has this got to do with System Trading?

Hold on. I’ll be there.
Today we know this philosophy as “The Toyota Way” and trust me, it’s a far better way. Fashionably called Lean Management, The Fourteen Principles etc, it embodies some of the most thoughtful insights of production.

Let me recount out of those fourteen principles, some three principles which are extremely important, in my opinion.

1. Create a continuous flow to bring problems to the surface

2. Use “pull” systems to avoid overproduction.
Instead of systems pushing output to the next, it should pull from the previous, this will decrease overproduction and wastage.

3. Use only reliable, thoroughly tested technology that serves your people and processes.

And the grand underlying current in all these? Use the right processes to produce the right results.

This underlying principle is the most important flow which should be there in our lives, and yes why not trading?Consider a trader as a businessman, he is putting in trades, day in and day out. A repetitive process. And when there is a repetitive process, automatically by the philosophy, you must have a system which takes care of the entire process.

Doing it better,doing it good

This explains my reasoning, and bent towards why system trading is the way to go forward for anybody who wants to get started or who wants to improve his resources and reduce his wastage.

And strangely even, this explains why I had a very volatile slump when I blew up, came back and blew up again. It was last year. The reasoning is simple. I didnt follow a system. I didnt really know what to do, when to do, how to do, and finally what not to do. And I paid a price.

But what saw my unleveraged portfolio which I am handling so much better, was because of the outright presence of a system . That perhaps saved me ,a lot of sweat. I didnt rationalise when markets were going up. I didnt explain it. System just jumped in. I didnt wait to form biases, do fundamental research(which could have, or could not help me make right decisions), wait to theorize if metals are going to be the flavor or agri. The system just took in, as per its rules, exited as per its rules, managed risk prudently and created something being satisfied of. (I post my portfolio results in here, every quarter. occasionally I post a sneak peak of my portfolio holdings and performance on my twitter feed- @sohamdas). Didn’t I have biases?
Oh yes! I did.
So what did I do about it?

Now, going back to Toyota Principles. Rule Number says,create a continuous flow of problems and solve them. Something which I say proactive improvement. Don’t just sit around waiting for problems to pop up and go into your firefighting mode, but create a continuous culture of improvement and preparation. Something akin to principle of kaizen. The Japanese principle of continuous self improvement. Apply the same to your systems. Keep tweaking, keep improving, keep learning, keep bettering. And once it reaches a certain stage make it operational. Once it is operational, go back to the study room. Rinse,repeat, rinse repeat!

I believe, if complex systems like humans, totally unpredictable, irrational, having their own quirks can improve by a continuous pursuit of perfection, then why can’t systems have their own evolution cycle? Yes ironing out of many quirks, smoothening out of many faults is possible, both in humans as well as systems, but we all know, there is a limit. But that should not stop us from kaizen. Instead it just gives us another level to play along with. Develop other systems. For example designing other flavours of systems. Designing systems with various quirks. Leveraging on them. etcetera.

So this brings us to a much deeper insight. There must be a system to design systems to place trades. In other words, there should be clearly well defined steps, targets, roles and minimum base line tests in order to create efficient trading systems.

And this is something,which is singularly eye opening. At least to me.

I dont want to create systems of non standard value. Of non standard specifications. Which god knows what, gods knows when will break down. I should properly know what is its historical point of failure, point of break down, how often does it break down etc. so that I know when a system has hit a bad patch and when it has not. Seen from a superficial angle, you might have created a system with your own set  of biases and I have done it in mine. So can we swap the systems? Not until unless, we have our systems of designing the systems right. Else you might be testing on the basis of one statistical tests while I might be checking the trading significance based on something else.Two completely non standard procedures. While one proves to be  a dud, the other proves to be a hit.

Perhaps, we should make a science out of the art of designing systems. Perhaps that might help newcomers much better than peddlers and hawkers selling “NASA designed trading systems with rocket geekery involved and guaranteed to increase your equity curve by a gazillion percent”.



P.S: I want to know how do you go for designing systems, and what are the general steps you take. Consider this as the fees for reading this blog!

P.P.S: Do leave me some feedback, comments, or might be some love(would appreciate it well if you are of fairer sex), so that I know, how close to I am of having a marijuana smoking, free love type of cult. Okay almost kind of!

Now back to designing systems.

No comments yet

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: