Mechanical Trading ... Does it work?

Currently I trade a mechanical system. It is not computer automated, I just trade it strictly by the rules. I have been daytrading the morning session with the system for about 4 months now and have recently noticed that sometimes I do get a "gut feel" that the trade is going to hit my profit target or hit my stop before it actually does. It's not something I can explain but does happen.

So I think I have found the answer to my question. To create a set of rules which can be mechanically followed but to use descretion or "gut feel" as well. BUT ... don't try to use discretion until I get to know the system well by watching it over time.

Thanks all, your insights have been helpful.


Brutus

 

LOL
I really have a problem with this word.
well all traders should. I will never use it again ... sorry for my english, not native.

tntneo

 

Brutus:
My feeling is that a mechanical trader is just a computer-assisted discretionary trader, nitro said that you have to know how to trade already in order to build mechanical systems, I agree 100%.
But there are exceptions, Prediction Company research team built profitable automated systems without previous trading experience, but I don't doubt they paid their tuition as everyone else. Actually their investors paid their tuition, but they got it back later.
About mechanical systems not working forever, it's a fact of life, get used to it. The fix is building not one, but several, and then dropping those that shows signs of degradation. So you build not a system but a methodology, a process, that's why you have to know how to trade before designing mechanical systems.

 

Brutus

Would you be willing to share your mechanical trading system for the benefit of newbies?

I am thinking about a "mechanical system" as follows for day trade (long strategy) and welcome comments from all:

1. Select stock that has a "stable" trading range.
2. When it's traded at the lower range, get ready to buy.
3. when the entry point is triggered, enter your "buy stop" order at last trade price + 0.05. This way, if the stocks keep going lower, hopefully you buy order is not filled.
4. If stock goes lower, re-adjust the "buy stop" price to go lower with 5 cents buffer zone until your order is filled.
5. exercise money risk management for stop loss after the order is filled and bail if the trade goes against you and exceed your risk tolerance.

For example, PFE has been trading between 39.40 and 41.75 in the last 30 days, today's low is 39.65.

Apply the above principle, one would enter a buy stop order at 39.70 when it printed 39.65. The fill would be between 39.70 and 39.80 and the trade would be profitable.

Can a trading life be this simple? I would love to hear from more experienced traders.

 

Assume I wish to adopt your mechanical system and start trading it tomorrow morning. Alas, I cannot, for it is too general and allows for subjective decision making. What exactly is a "stable" trading range? Do I take a 30 day range like in PFE? Similarly, the entry and exit criteria are not clearly defined. Until you fully refine your methodology to completely eliminate all subjectivity, you will have a discretionary system on your hands. It is well worth the extra effort - you're nearly there.

 

1) Yes, it works; we trade 20+ systems at a time.

2) Yes, the development time and costs are lengthy
and expensive.

3) No, systems do not work forever.

 

Two things

To TNTneo: Thank you for your description. I don't want to know what your trading system is. Right now I'm teaching trading to a friend of mine and I was searching for some sort of idea how I could give him a good education and how he should begin. I think I'm now going to let him trade with my system and then after time I will show him the little details on what he can look at and further elements of becoming successful in trading like special news reports etc. Like you mentioned achieving the discipline is the main approach... and for this I think I have found a very good way to teach him that.

How long did it take you for the transmission from the objective systemtrader to adopt the good feeling in trading.



Alain
(Switzerland)

 

The other thing:

Some of you mentioned one of the main elements of systemtrading: Having different systems that work in the different market conditions. Then all that the trader has to do is define the market situations.

First, couldn't we build a system that defines the market situation in the BIG picture that then defines what system to trade?

Right now I only trade with a trend following system.. So what are the other market situation this system doesn't work. In a Trading Range... what other market conditions do you think to focus on to build a particular system.


Alain
(Switzerland)

 

Hi Alain, I was not sure why you were asking the question. Now I know and I like it. It's about learning not asking the 'magic' way.
back to your new question, making the transition is probably very different for each trader. It takes a 'long' time to really understand what people and the market are telling you. For too long what you think is the 'truth', not what you see. Anyway, it took me a bit less than a year.
I still love systemtrading, and if I have a new idea, of course I start coding first, not trading first. Several have written this big truth : backtesting does not make sure the system will work in the future. however a failed backtesting is sure to show you the system won't work in the future.
but once the idea is proven I will quickly switch to my pseudo discretionnary mode. (usually after the forward and live testing).
Some systems are not compatible with me in freelance mode. Then I don't use them, others are.

tntneo