Systematic Trading vs. Discretionary Trading

Hello Forum,

I thought it would be interesting to start a little discussion about the two groups of traders.

I am myself more a systematic trader, as I am a believer of clear rules. However I must admit my trading style carries some discretionary elements!

I think successful Traders all follow rules, whether they are systematic Traders, or discretionary Traders. The bad systematic Trader has rules which are no good. The bad discretionary Traders seem to favor the entry caused by greed and an exit caused by fear, which is similiar to having no rules, or closely related to gambling.
The good systematic Trader has a good set of rules, however the good discretionary Trader is more flexible and able to adapt to their new environment.

So at the end of the day, I seem to think it is all about developing the right rules? Maybe this could be our next discussion...

Regards
Paul

I hate to be so broad in my statements, but I have come to believe that any system that has discretionary elements (even minor ones) is destined to fail over the long term. The number of people that can exercise enough discipline and self control to make them work is very, very small.

In addition, I believe that the best traders use systems that are very long term, very boring . . . and very profitable

I wouldnt be so quick to make believes like that after not even 1 year in this bussines.

I'd have to strongly disagree with that seeing how the "big dogs", the traders working at banks usually utilize discretionary trading styles depending on fundamental factors. I'm not talking about algorithmic traders or even those who deal with derivatives like options, etc. But spot forex traders usually base their trades on longer term fundamental views and only enter in at tech. signals.

As for these best traders who trade longer-term, I doubt many of them are willing to place a long term position without at least a minimal idea of which direction the fundamental trend is.

You need to rethink this statement buddy!!

Longtobefree's response is not untypical of somebody who is trying to find their way (albeit he seems to be unhindered by any form of false modesty), he is right about some things and not about others. What it reminds me of however, is the amount of time during ones journey one forms a set of beliefs about what works, only to have them pulled down brick by brick by the markets. I know I have arrived at such final conclusions many times only to meet with disappointment. Perhaps this is a subject for another thread.

To the subject of the thread, I've only seen concrete proof of the long term profitability of a handful of independent traders, and none of them post here.
From what I know of what they do, they think in ways completely orthogonal to what most people think of trading, for example, most don't consult charts to make trading decisions, think about that for awhile...

My opinion on what makes a trader successful is that he or she has an idea of which direction the market is most likely to go (most likely, not definitely. after all, this is a game of probability) irregardless of what is happening in the charts. This is why some traders in big banks can trade profitably long-term without even a glance at technicals except maybe price levels.

Now when you can predict market direction with reasonable accuracy (as in above 50% factoring in the spread or commission), then technical triggers and money management will come into play to maximize your profits. However, the key is in figuring out the market direction and then using technicals in conjunction with that, not the other way around.

How do you predict market direction an accuracy of above 50%? Fundamentals. Interest rate differentials, economic outlooks for both countries in a currency pair, market momentum, etc. In other words, discretionary outlooks based on fundamental factors.

i second this. ive seen these guys in action. i've even talked to a few of them about systems and they sort of chuckled at me and gave me that annoying patronizing smirk (freakin pissed me off lol). but yeah, they either trade based on short term things like interest rates or whatever, or they take a long term fundamental direction and trade short term within that. heck, i've seen some guys buy a currency just because they felt like it. if that's not discretionary, i have no clue what is.

but those are the big guys. they either have years of experience or are trained by people with years of experience. normal retail traders dont have that luxury, so for a retail trader, having rules is a good bet until you have been trading for 5-10 years. at that point, you should pretty much know the market and you can trade however the heck you want to.

as for my vote on the subject, i have to lean to systematic. thats just the way i am; i'm a more logical, reasoning type of person. a couple of the guys i work with are true discretionary traders tho. they use no rules at all for some of their trades (scares the crap out of me). so i dont think there is one specific way that works for everyone. go with what works for you.

I've stuggled with exactly the same issue in developing my trading method, and found that there is value in both a systematic approach to entries/exits, and also great value in discretion when it comes to trade management.

First, entering and/or exiting a trade. I found the greatest success by using a "robotic" mind set for entering and exiting a trade. Assuming your enter/exit mechanism has an edge on the market, it should be robotic...enter on each and every entry signal, exit on each and every exit signal. This gets you into an out of the market on a very consistently profitable basis.

Now, trade management is a whole other animal. There are people (like me) who like to milk, encourage, coach, begile, and seduce their trade to move to higher and higher time frames, while simultaneously moving stops along to lock in profit along the way.

From my experience, I've never seen any "systematic" approach to managing a trade that actually works well...like entry/exit mechanisms. Instead discretion plays a big role in looking at a larger picture and deciding on a level to place your stop to as to work your way from a 15 min time frame to a daily time frame.

Also discretion plays a BIG role in "knowing" when NOT to trade. Deciding that the market just isn't going your way, or that you've made your projected pip goal for the week or month and now you'll wait until next week, or next month to look for trading opportunities again.

In short, entries and exits should be as robotic as you can make them...I use a bot that I wrote myself. But, trade management and the decision to trade at all is definately discretionary and in my opinion a successful trader cannot have one without the other.

I too use both.

I analyse the market direction then launch my EA to work in only that direction. That way all I have to do is make sure my direction is the correct one

That way I am not emotionally attached to the trades, I just babysit it everynow and again to make sure the market is still heading in the intended direction.

Thanks

Hey guys,

thank you for your input.
My first post was an introduction to this topic, now lets look at another difference between the two. Systematic trading also includes backtests, Empircal evidence of the strategy. Now I personal backtest all my strategies before I trade them, as I believe it is a important aspect. I think a lot of people create none true empircal evidence because of the lack of quality or length of data, or due to over optimization.

On the other side of Empiricism there is Rationalism, which is the foundation of many discretionary traders, long-term fundamental traders. This is done more by logically reasoning. Although I am a systematic trader, my strategy has to be supported by "Rational Evidence". Rational Evidence would not be that when the 23 SMA crosses the 51 EMA, this is clear buy signal, but that Traders are driven by emotions; fear & greed, and a trend-following approach is the right way to go. This is just an example.

At the end of the day I think both approaches are vital, and are best when used together. Like in science, it is probably best to join the two approaches from John Locke and Rene Descartes...

Althought I am going more into philosophical ideas, I think it is import to discuss this idea, as it connects very much to trading.

Regards
Paul

I think this explains the combination of the two pretty well.
You discretionary analyse / trend filtering, is done "by hand", although it could probably be programmed.
But either it is too complicated to be programmed, or you want to remain flexible or adapt to the environment....
Regards
Paul

PokPok,

Although I whole heartedly agree with your logical and systematic approach to trading and gathering data. There are some holes that cannot be filled in this approach...I am just saying it to make you or others aware not that I dsiagree with what you are doing.

Discretion whether rational or irrational cannot be quantified nor can you see it emperically (except as a measure of your account balance). When demontaz decides to trade only in one direction he is using his analysis of the market from HIS point of view. Another trader doing the same thing may chose the opposite direction, still another may chose NOT to participate.

Each trader has these same three choices, and further, each trader may chose differently under what appear to be identical situations in the market. This kind of variable and the decision making process that occurs in the human brain cannot be measured, programed, or quantified even by the best experts. Multiply this by the number of people (yet another variable) that participate in the market at any given time.

I dare say that not matter how far you back test your data and "know" your system works, there will come a time when you use your discretion to enter, manage, or exit a trade that does NOT fit into those cute set of rules we all like to refer too.

Therefore, this discretion will skew your "real" data. While it is good to check your system for an EDGE in the market, don't rely on it too heavily. The human variable must be accounted for and that is the one variable that cannot be quantified or repeated.

Instead practice making "good" decisions...paper trade your method with a demo account. Then put "real" money on the line using micro lots and then use your discretion to make the "correct" decisions each and every day. When you are confident you can make good decisions from a "neutral" mindset EVERY time you sit down at the computer, then you are ready to start trading larger lot sizes and make ever increasing sums of money.

great post, gspajon

another thing to add about discretionary trading, when you arent trading inside a set of rules or watching a set of indicators, you are relying on your own skills and intuition. when you do this, you learn how the market works. this then builds experience and you now have something to fall back on when your indicators or systems fail. its always a good idea to have something to fall back on when everything else falls away. my personal opinion on the subject is that you should learn the market first, then set up your rules to follow what the market does.

Black Sun,

First of all thank you for your compliment, I very much appreciate it. I too agree with the above premise, "First learn the market". However, for many of us (and I count myself among these) just looking at the market and watching just price action, seems like random noise. For me that only thing that helped me recognize patterns in market behavior were indicators. I too agree that relying solely on indicators can be fatal to your trading account, as it was to mine. However, there are those of us who are still struggling to "feel" the market flow and indicators give us a rational way to recognize behavior patterns in the market.

Once we are able to "see" the market behavior we can then concentrate on finding repeating patterns that have a high probability of success. This process while very long indeed, also introduces us to the concept of human behavior in the market as we watch some patterns errod, while others hold true. During this process one immutable fact becomes increasingly clear...there is NO pattern to human behavior, only probabe outcomes...like the "science" of psychology.

Sometimes there is absolutely no explanation for aberant behavior, (ex: Jeffrey Dahmer), yet it still occurs, and more often than we want to recognize. It is here that our discretion plays a huge role, because the probability is leading us in one direction, while that "gut" feeling says something completely different. The answer? RISK MANAGMENT.

When ever I get that "still small voice" in the back of my head, I start reducing the risk in my trade. Tighten stops, or enter with a smaller position AND a tighter stop. So far it has served me well. Again this too is not perfect and it is certainly one of those things can cannot be measured or programed. It just IS.

This I believe is one of the most valuable skills one must develop as a trader, and once it is fully developed, YES, you don't need a chart to "know" when or what direction to trade....I'm still working on that.