Mechanical Trading Systems The Way to Trade?

[tommog]

Hi,

I was wondering whether people have a set mechanical trading plan such as a moving average cross or some other indicators and only enter if these criteria are hit regardless of any other factors in the market? Or do you trade by building up a picture of what is happening in the market and enter using a variety of techniques that you apply to the current situation?

When i first started trading i looked for mechanical systems but the more i trade i feel that although there are similar patterns that occur in the market, every trade should be treated as a unique situation therefore making mechanical systems a little too simple to really profit from the market.

What do you think?

Tom

 

[Anonymous]

Assuming one is not using a computer programme to resell, after a buy entry point a trader can do 3 things:

1. sell to take profit
2. sell to cut loss

or

3. move the stop as in trailing

Another method involves recognizing the situation in which there is a difference between the probability of winning and the odds, and all that's needed is a simple click of the buy or sell button with a limit close or stop loss if the initial position is executed, all done within seconds, but the waiting time could be minutes or hours.

Very mechanical when it comes to order execution.

 

[the blades]

Hi,

I was wondering whether people have a set mechanical trading plan such as a moving average cross or some other indicators and only enter if these criteria are hit regardless of any other factors in the market? Or do you trade by building up a picture of what is happening in the market and enter using a variety of techniques that you apply to the current situation?

When i first started trading i looked for mechanical systems but the more i trade i feel that although there are similar patterns that occur in the market, every trade should be treated as a unique situation therefore making mechanical systems a little too simple to really profit from the market.

What do you think?

Tom

Hello Tom,

I trade UK stocks on a mechanical basis. My system produces more stocks than my portfolio requires, so I apply a little discretion as to which to choose.

I trade the FTSE index and S&P500 stocks on a purely mechanical basis.

I love the lack of emotion and the ability to walk away and leave trades to run their course, unless I'm losing that is I also find as a part time EOD trader, mechanical systems require less time at the screen.

All IMHO.

Cheers,
UTB

 

[ZDO]

...
What do you think?

Tom

The best entry methodologies are the ones that UNIQUELY FIT YOU and you will only discover those in consults with your 'inner' councils rather than out here on boards, etc. No matter how 'profitable' a method is, you won't stay with it if it doesn't really fit YOU.

You discuss the 'randomlike' outcomes of your mechanical setups. Do you realize that the 'context specific' discretionary tweaking of setups you are considering will also have 'randomlike' outcomes?

Good entry setups and triggers are so completely overrated! Get contrary and work it backwards. Emphasize and understand your own 'psyche' as it relates to trading and timeframes (quantity of opportunities,etc.) first, master position sizing second, management of profits third, management of risk next, and finally develop setup and entry techniques.

You could spend many years and mucho dinero grail searching for (and finding great) entry techniques and still go nowhere without developing the foundations listed above. I maintain that if you totally ignored setups and entry techniques and instead really mastered this 'list', that your own personalized portfolio of setup and entry techniques (with the ideal personal proportions and cycles of mechanical and discretionary) would, as if by magic, present themselves to you...

All the best,

zdo

 

[mr_cassandra]

Hi,

I was wondering whether people have a set mechanical trading plan such as a moving average cross or some other indicators and only enter if these criteria are hit regardless of any other factors in the market? Or do you trade by building up a picture of what is happening in the market and enter using a variety of techniques that you apply to the current situation?

When i first started trading i looked for mechanical systems but the more i trade i feel that although there are similar patterns that occur in the market, every trade should be treated as a unique situation therefore making mechanical systems a little too simple to really profit from the market.

What do you think?

Tom

For my own trading, mechanical has been the way to go. My gut feelings stink, whereas I'm not bad at programming. I agree with the idea that there are repeating patterns which can be taken advantage of.

 

[andrewbee]

I too trade 100% mechanically (Dow futures), every hour that the market is open. I could probably make a lot more if I put in the time and work to learn trading from a master, but that's not what works for me. I want to be able to leave my time AND emotions out of it.

I am a professional programmer, and very analytically-minded, so it was easy for me to produce my own custom backtesting and automatic trading software. However, the reason for doing trading in the first place is to be able to not be tied to a screen all day. My goal is to leave the programming biz behind in one year.

Like a previous poster said, find the style that works for you emotionally. If you do something you love, proficiency will naturally follow. Feel free to PM me if you need advice on automatic trading.

 

[DDI]

I dont have the expertise to program my criteria and leave it to trade automatically but I do base my trades on technical criteria such as MA x over, ADX, MACD etc....

I find my system is ok and over long term i do produce profits, although i have gone through long losing streaks.

And with experience I am finding listening to financial news and other analysts throws me off track. for eg I head some analysts that USD will strengthen by 5-7%, and was short EURUSD from 147, but didnt take my proft at 14530 (when my system gave bullish signal)!!!

I really need to block myself off Bloomberg tv, cnbc tv and the FT.

 

[grantx]

Mr Cassandra,

My guts likewise (maybe it’s dietary).

Which is the simplest programme to implement a mechanical system, eg buy when this, sell when that? My Excel is OK, vba is just below my gut instinct. You’re going to suggest something like C+, aren’t you?

Grant.

 

[TickCOM]

Hi,

I was wondering whether people have a set mechanical trading plan such as a moving average cross or some other indicators and only enter if these criteria are hit regardless of any other factors in the market? Or do you trade by building up a picture of what is happening in the market and enter using a variety of techniques that you apply to the current situation?

When i first started trading i looked for mechanical systems but the more i trade i feel that although there are similar patterns that occur in the market, every trade should be treated as a unique situation therefore making mechanical systems a little too simple to really profit from the market.

What do you think?

Tom

That depends on the platform you use. There are a lot of market players out there who thinks they can make money this way - they are most propably taking your money in this very moment. Look at this video from Reuters:

YouTube - A revolution in trading

I dont think automated trading will take over manually trading as lined up in the video - but rather a new segment forming.

Till lately, there has been no platforms available to make profitable automated trading, except if you are programming the system your self, and you are using several timeframes. But the market is moving, providing new advanced tools using GUI in the development of algorithms ( which all mechanical systems needs ).

There is a post in another section on trade2win about a fellow who is living on developing algorithms for automated trading - so its possible.

 

[TickCOM]

But what excatly do you mean about mechanical trading ?

A charting software giving you a ping when it advices you to trade - or a full unattended trading system in the basement running 24h a day ?

 

[mr_cassandra]

yes, i do

I developed and used my own mechanical system some years ago. I need to disclose up front it is currently a paid service and I am not advertising by answering your post.

If you want a free obsolete copy of my program to get ideas from just email me at [email protected]. Please state in your email that you understand I'm not offering investment advice and that the version you receive is no longer in use.

It uses several indicators to make buy-sell-short decisions and is composed of visual basic commands in a microsoft excel file.

Hi,

I was wondering whether people have a set mechanical trading plan such as a moving average cross or some other indicators and only enter if these criteria are hit regardless of any other factors in the market? Or do you trade by building up a picture of what is happening in the market and enter using a variety of techniques that you apply to the current situation?

When i first started trading i looked for mechanical systems but the more i trade i feel that although there are similar patterns that occur in the market, every trade should be treated as a unique situation therefore making mechanical systems a little too simple to really profit from the market.

What do you think?

Tom

 

[mp6140]

Mp --- Robots Or Humans ?

Hi,

I was wondering whether people have a set mechanical trading plan such as a moving average cross or some other indicators and only enter if these criteria are hit regardless of any other factors in the market? Or do you trade by building up a picture of what is happening in the market and enter using a variety of techniques that you apply to the current situation?

When i first started trading i looked for mechanical systems but the more i trade i feel that although there are similar patterns that occur in the market, every trade should be treated as a unique situation therefore making mechanical systems a little too simple to really profit from the market.

What do you think?

Tom

======================================================================

ALL "decent" systems work well a lot of the time, although what you describe is not a "mechanical" system per se.

the problem with most is that they are late to get in and often even later to get out because of the inherent nature of a ma x-over. Additionally, you must utilize extremely good money management techniques because these systems work best on the longer timeframes, which unfortunately can often resemble a drunken person, walking down a street, what with the ups and downs, dips and weaves !

while originally VERY MUCH AGAINST THEM, i think theres no better way for a person to learn (learning the markets takes MUCH longer than you can imagine) then by watching the ups and downs of the market as the price seeks your profit.

having said this, you must understand that even with a "simple" system, you must be concious of stop losses (i dont use them but do not advise you do such until you have a WHOLE LOT more experience), only using a small amount of your capital so as to avoid the dreaded "margin call" and NOT overtrading, which is simply HAVING to trade another position, especially if your present one does not look good (when you overtrade, you start building up the use of margine, and sooner or later like a house of cards, everything falls down on you.

needless to say, demoing a few systems till you find one that works with your particular mindset is the most important thing.

sit down, learn slowly and be prepared to learn a whole new world, with its own rules, culture and bandits.

Fortunately, there are some good guys out there -- look for the ones in the white hats who carry silver bullets and ride a white horse. Having an indiginous native sidekick doesnt hurt either !

HI O SILVER - - - - - - - - - - AWAYYYYYYYYYYYYYYYYYYYYYY!

mp

 

[yongp]

Hi,
I'm new in futures and got burned badly letting my emotions interfere with an open position. I've read that the best way to start is by having a mechanical system and later on mixing it with intuition towards a discrete system. Would like to seek help on what resource can i get to begin exploring a mechanical system to start with? i.e. any good book to recommend? a classical chart pattern versus price action versus technical indicators - which is better?
thanks heap!

 

[Chorlton]

Hi,
I'm new in futures and got burned badly letting my emotions interfere with an open position. I've read that the best way to start is by having a mechanical system and later on mixing it with intuition towards a discrete system. Would like to seek help on what resource can i get to begin exploring a mechanical system to start with? i.e. any good book to recommend? a classical chart pattern versus price action versus technical indicators - which is better?
thanks heap!

Hello Yongp,

Like others on here, I too, develop my own mechanical systems. The main reasons for choosing this approach were:

1. Emotional element is "almost" removed. You will still experience emotions when it comes to monitoring your Equity Curve, Drawdown, etc but at least on a trade by trade basis, this element is eliminated.
2. IMO Mechanical systems are those which are "hard-coded" and as such can be Backtested. This is VITAL if one is to commit real money to a strategy. One needs to know in advance, that their idea has been proven to have a positive expectancy. There's no guarantees that a strategy will continue to work in the future, but knowing that it has worked in the past is definately a step in the right direction.
3. By carrying out thorough backtesting, one can begin live trading with a set of metrics such as MaxDD, No of consecutive Losing Trades, etc. These metrics can then be used in real trading to monitor the system to ensure it is behaving as expected.
4. As already pointed out, screen time should be reduced. I am more than happy to spend time looking at charts for strategy ideas but only on my terms. I do not want to be tied to the screen for the whole trading session.

Regarding what approach is better, I would say that this is an individual decision. Personally, I only rely on price action, as I find that indicators contain lag. However, if I need to calculate something and a corresponding indicator already exists then I may use it so as to reduce the amount of coding.
I should stress that this is a personal choice though. I'm sure that one can probably design a strategy around only indicators but I would suggest that the system's metrics such as drawdown, timeframe chosen, length of trade, etc would need to differ from one designed around Price alone.

Books I would recommend (in no particular order):

Design, Testing and Optimization of Trading Systems by Robert Pardo
Mechanical Trading Systems by Richard Weissman
Trading Systems by Perry Kaufman
TradeSim Manual

The last one is a manual that accompanies the TradeSim software, which IMO is vital software if one is serious about developing & backtesting Systems.

Hope this helps,

Chorlton

 

[mr_cassandra]

machines versus emotion

Morning, I need to tread carefully in responding so I'll state up front I'm a system vendor.

Now thats out of the way, I had the same experience as you and I wrote my own mechanical trading system because I needed (urgently) to cut ME and my emotions OUT of the loop.

I trade the sp400 index and this has worked very well.

Bear in mind as you pursue this that NO system or method will be perfect and don't waste any time seeking perfection.

Instead focus on consistent imperfection whoses gains have a decent upward march over time, warts and all.

Stick with it, be very persistent and it can be done. The boards are full of quasi-gurus and learned pronouncements why this and that can't be done. Most of them have never tried.

Drive AROUND them.

Regards, Steve

Hi,
I'm new in futures and got burned badly letting my emotions interfere with an open position. I've read that the best way to start is by having a mechanical system and later on mixing it with intuition towards a discrete system. Would like to seek help on what resource can i get to begin exploring a mechanical system to start with? i.e. any good book to recommend? a classical chart pattern versus price action versus technical indicators - which is better?
thanks heap!

 

[Chorlton]

Morning, I need to tread carefully in responding so I'll state up front I'm a system vendor.

Now thats out of the way, I had the same experience as you and I wrote my own mechanical trading system because I needed (urgently) to cut ME and my emotions OUT of the loop.

I trade the sp400 index and this has worked very well.

Bear in mind as you pursue this that NO system or method will be perfect and don't waste any time seeking perfection.

Instead focus on consistent imperfection whoses gains have a decent upward march over time, warts and all.

Stick with it, be very persistent and it can be done. The boards are full of quasi-gurus and learned pronouncements why this and that can't be done. Most of them have never tried.

Drive AROUND them.

Regards, Steve

Hi Steve,

Can I ask what parameters/metrics you focus on when deciding whether a strategy is viable or not? ie. MaxDD, CAG/MaxDD, Sharpe Ratio, Net Profit, etc, etc

Thanks,

Chorlton

 

[mr_cassandra]

parameters

My work is completely outside the standard box of commonly accepted beliefs on mechanical systems. I only had one goal, to far exceed what a good broadbased mutual fund would deliver. Viable in my opinion is a moderately steady march upward over time with drawdowns which don't put you out of the game but are not inconsistent with your gain rate. IE if you're gaining 30% a year, 20% drawdowns are horrific. If you are gaining 100%+ per year, you will find that same 20% drawdown to be more proportionate.

I wrote my own program in excel using visual basic commands. I started by targeting the major selloffs usually seen in the fall, in the u.s. market. The code then began to find the same thing in mar-april of many years. After that I targeted smaller mini trends within the overall as well as shorts.

I made many mistakes and had many setbacks. I have learned from them and gone on.

I have to be honest and say that I FEEL many books and commonly accepted evaluators are from people who have never fielded a successful system. All of my work ended up with a viable system and truthfully I never considered all the academic recommended statistics. Those are not helpful in creating a system but yet, if your work is good, you'll end up satisfying those parameters.

I want to give you a very strange analaogy but consider this. The investing world is focused on sending you to read books xy and z for everything you want to do. In my years at Church I used to watch people jump thru hoops doing bible studies. When they were done I'd ask them about the meanings of what they'd read. Most of the time they had absorbed no meaning , they were just robotically 'studying'

In investing and religion its not the rote studies that matter, its the meaning and appication.

Hi Steve,

Can I ask what parameters/metrics you focus on when deciding whether a strategy is viable or not? ie. MaxDD, CAG/MaxDD, Sharpe Ratio, Net Profit, etc, etc

Thanks,

Chorlton

 

[Chorlton]

My work is completely outside the standard box of commonly accepted beliefs on mechanical systems. I only had one goal, to far exceed what a good broadbased mutual fund would deliver. Viable in my opinion is a moderately steady march upward over time with drawdowns which don't put you out of the game but are not inconsistent with your gain rate. IE if you're gaining 30% a year, 20% drawdowns are horrific. If you are gaining 100%+ per year, you will find that same 20% drawdown to be more proportionate.

I wrote my own program in excel using visual basic commands. I started by targeting the major selloffs usually seen in the fall, in the u.s. market. The code then began to find the same thing in mar-april of many years. After that I targeted smaller mini trends within the overall as well as shorts.

I made many mistakes and had many setbacks. I have learned from them and gone on.

I have to be honest and say that I FEEL many books and commonly accepted evaluators are from people who have never fielded a successful system. All of my work ended up with a viable system and truthfully I never considered all the academic recommended statistics. Those are not helpful in creating a system but yet, if your work is good, you'll end up satisfying those parameters.

Steve,

I find your comments interesting, as I like the "thinking outside the box" approach which you seem to have taken.

Apart from the link to your Yahoo homepage, do you have a blog, etc?

 

[grantx]

Yongp,

I would suggest studying - and perfecting - basic bar charts and developing your strategy from there. There are hundreds of technical indicators and thousands of books which alone will absorb one lifetime.

The only TA book I can recommend is Murphy's Technical Analysis of the Futures Markets.

Steve,

I'm glad we now have a name. Two excellent posts.

Grant.

 

[new_trader]

Morning, I need to tread carefully in responding so I'll state up front I'm a system vendor.

So why do you hide the fact? I thought site guidlines clearly stipulate that if you are a vendor you should have a Vendor icon.