Mechanical Trading systems

[tradias]

Dear traders,

I am new to forums but have been a trader for some time. I wanted to understand how traders in general perceive purely mechanical trading systems and how many have really consistently made money trading currencies with those so called Pure mechanical systems. What i mean by mechanical are those that generate buy sell signals based on the traditional technical indicators.

I look forward to a healthy interaction

 

[£10kLoser]

I believe they can indeed produce profits but they are greatly helped by some discretion on the part of the trader. For example, mechanical signals may fail at new release time or during the dead period between the end of the US session and when the Asian session becomes active. They also often fail when there is obvious sideways movement.

 

[Adamus]

Ed Seykota and Richard Dennis both made their fortunes with mechanical trading systems, although they operated in the futures markets, not forex.

I'm trying the same approach and all I can say is that there's a lot to learn, and a lot to unlearn, too.

 

[nile_croc]

Mechanical systems are great, however, you won't get rich trading with them. These strategies are great (i.e. the turtle system) and teach all the basics (position sizing, scaling, diversification, effective use of stops) however you need to be very well capitalized (and patient!)to trade them the right way. The way that people such as Richard Dennis, Ed Seykota, John Henry, etc. make money, although they are very good at what they do, is by managing other people's money for a fee. These managers also have the capital base to be able to diversify across dozens of markets and sustain large drawdowns, which are necessary to implement long-term trend following systems.

Keep one thing in mind: just about all the great ones, Soros, Rogers, etc., are discretionary traders and use their analysis as a means to gain an edge over the market. Most of the great money managers and traders are people who do lots and lots of homework and gain an understanding as to what is going on, and their trades are a result of their knowledge. They're not always right, and are frequently wrong, but they definitely know how to make big money when they are right and hold on to it!

Just my two cents.

 

[stevenzed]

AutoTraders

Hi,

I believe they can work if you develop your own and do not share it with 1000's.
I believe the market is sophisticated enough that if many of us are doing the same think they eventually quit working. Nevertheless, I am still searching for the holy grail.
I am currently looking at the forex-ai system which claims to work differently for each user.

Steven

 

[Adamus]

Just the format of their website is enough to put me off forex-ai. Make sure you test it in multiple markets and multiple market conditions to see how it holds up in markets other than the ones they tested it on.

 

[4xleader82]

Forex-AI AutoTrader is an Artificial Intelligence based Expert for MetaTrader 4.
An Automatic Trading Robot which can generate consistent profit every month!

Unlike many other systems available on the market, this is not a system works as a back test only. They have live statements available directly from our broker account and it gets refreshed every 15 minutes. You can see how the system trades. Live

I had to google the term forex AI to recall what it was.. I've been on the MQL4 site and someone did make this system and is offering it for free..

In my opinion, a system that can not be backtested and can only run live (demo: obviously) is risky business. We all know that the market is not predictable. (this is the reason why we use indicators, support, resist, etc to HELP us out with our trades. AND even with the best indicators, we still make mistakes because We do not make the market. Good luck with the A.I. and I'll have to inform myself more before having a solid opinion.

 

[rtsurvivor]

I've seen a lot of "Pure Mechanical Trading Systems" come and go over the years. One thing I've noticed about all of them is that a fairly large trading account is required.

A "Pure Mechanical Trading Systems" is one that generates a buy/sell signal and the member/client takes the trade; no questions asked. That is walking on a tight rope for a small account; the draw-downs can clean out a small account.

A good mechanical trading system , IMO, is one that allows the trader to use a bit of judgment. I'm checking out a few that may just work.

A system that uses the "Trigger-Lines indicator caught my eye and I'm about to run it thru some real time prices. In reality, back testing doesn't really prove anything.

The attached chart shows the system. I'm using the "Mechanical System" loosely but the "Trigger-Line" indicator says when to buy or sell. It does require watching but the thinking part is reduced quite a bit. :whistling

The jury is still out on it but I'll post some results on it a bit later. :smart:

Good trading to all
RT... :clover:

 

[neil]

Trading methods

I've seen a lot of "Pure Mechanical Trading Systems" come and go over the years. One thing I've noticed about all of them is that a fairly large trading account is required.

A "Pure Mechanical Trading Systems" is one that generates a buy/sell signal and the member/client takes the trade; no questions asked. That is walking on a tight rope for a small account; the draw-downs can clean out a small account.

A good mechanical trading system , IMO, is one that allows the trader to use a bit of judgment. I'm checking out a few that may just work.

A system that uses the "Trigger-Lines indicator caught my eye and I'm about to run it thru some real time prices. In reality, back testing doesn't really prove anything.

The attached chart shows the system. I'm using the "Mechanical System" loosely but the "Trigger-Line" indicator says when to buy or sell. It does require watching but the thinking part is reduced quite a bit. :whistling

The jury is still out on it but I'll post some results on it a bit later. :smart:

Good trading to all
RT... :clover:

Here is another method - Go Naked:
http://www.trading-naked.com/library/LBR_Scalp_setups.pdf

A list of methods\;

Library - Articles & Reprints

No Brainer Trades - Live Trading and Analytics for the Forex Market

 

[Chorlton]

Dear traders,

I am new to forums but have been a trader for some time. I wanted to understand how traders in general perceive purely mechanical trading systems and how many have really consistently made money trading currencies with those so called Pure mechanical systems. What i mean by mechanical are those that generate buy sell signals based on the traditional technical indicators.

I look forward to a healthy interaction

Hello All,

If I may add my 2cents.

I am personally aware of a large number of successful traders who trade purely mechanical systems (by mechanical I refer to any system where no discretion is used... ie. The signals are followed without question). A small percentage of these systems trade forex, while the remaining ones trade equities.

As someone who also develops mechanical based systems, I believe they are an excellent approach but only if they suit your own personality / style of trading.

For example, with any trading approach after a string of losses it can be difficult to "pull-the-trigger" on the next trade. With a mechanical system this emotional problem is reduced as if one has faith in their system then one should simply continue to keep trading as long as the results achieved to date, fall within acceptable previously backtested limits.

Another advantage (not discussed very often IMO) is that a mechanical system will tell you when to trade and more importantly when to stand down and instead sit on the sidelines. The emotional need to keep trading by some traders is removed by following the signals without question. For example, During the 2nd half of 2008, a lot of the systems traded by those I know switched off automatically due to mrkt conditions and moved to cash. In contrast those who were trading similar strategies but in a discretionary way continued to trade. From what I have read from them during the last 6mths, the discretionary traders were finding it very difficult to make money. (Note. These systems were trading Equities and not forex)

Where many go wrong with mechanical systems (including myself at the beginning) is that they believe that a particular system will continue to work forever!! This is rarely the case for many reasons. However, As long as one continuously monitors their systems performance, one has a better chance of identifying when something has changed.

Another problem is that some system developers try to develop a system which works in all mrkt conditions. This is also rarely possible IMO. If, for example,you have a strategy which looks for trends, then one shouldn't be surprised if the performance of it drops off during range-bound mrkt conditions. Afterall, it wasn't designed to trade this type of mrkt.

regards,

Chorlton

 

[rtsurvivor]

You are right on most points chorlton. However, what a number of the systems don't allow for is drawdown for small accounts. I refer mostly to commodities and even the FOREX; I never cared that much for stocks.

I can site one example that I followed for a while that used Ellot Waves to call the trades in the lumber markets; drawdown was not factored in. I made money following the trades but that is only because I could withstand the DD problems that a number of the members could not do. {And We paid $300 p/mo too!} Lumber is a thin market and can make some wild swings. The guy that ran the system was very good at what he'd developed, but again, the DD turned into a real commodity trader killer.

What bothers me most are the ones that "HYPE" their "Pure Mechanical Trading System." to the new FX , or commodity traders. They're led to believe that a small account is enough to make their fortune if they sign-up for greatest trading system in the universe; just pay $xxx each month! It wouldn't take long to eat up their account.

That, of course, is just my opinion.

It may sound like I'm on a soap box but I've been in this game for a tad over 25 years and I've seen a lot of things come and go. I guess it just sticks in my craw a bit to see the new traders getting the short end of the stick all the time.

Anyhow, good trading to all.

RT... :clover:

 

[Chorlton]

You are right on most points chorlton. However, what a number of the systems don't allow for is drawdown for small accounts. I refer mostly to commodities and even the FOREX; I never cared that much for stocks.

I can site one example that I followed for a while that used Ellot Waves to call the trades in the lumber markets; drawdown was not factored in. I made money following the trades but that is only because I could withstand the DD problems that a number of the members could not do. {And We paid $300 p/mo too!} Lumber is a thin market and can make some wild swings. The guy that ran the system was very good at what he'd developed, but again, the DD turned into a real commodity trader killer.

What bothers me most are the ones that "HYPE" their "Pure Mechanical Trading System." to the new FX , or commodity traders. They're led to believe that a small account is enough to make their fortune if they sign-up for greatest trading system in the universe; just pay $xxx each month! It wouldn't take long to eat up their account.

That, of course, is just my opinion.

It may sound like I'm on a soap box but I've been in this game for a tad over 25 years and I've seen a lot of things come and go. I guess it just sticks in my craw a bit to see the new traders getting the short end of the stick all the time.

Anyhow, good trading to all.

RT... :clover:

RT,

I do agree with your points above.

The models I develop focus on "risk". I have a predefined set of criteria that my systems' must achieve for me to consider developing them further and then onto trading them. This set consists of risk per trade, MaxDD that I would be happy to trade through, smoothness of Equity Curve, ability to remain stable through specific Mrkt conditions, etc, etc.
If I can develop a system which meets my criteria, then assuming that the Profit Return is reasonable then I would consider continueing with it but the key is always to focus on the risk side first.

Drawdown will always be a problem with small accounts IMO. Either one accumulates a larger capital base to begin with or one uses leverage to increase ones exposure to "more trade possibilities". This does not mean taking more risk!!!
I backtest my systems using an account size of 20k which I believe to be a reasonably small size. I could probably reduce this a little bit more and still achieve a reasonable return but am happy to use this as my starting capital size for each system. Some will say that this figure is a lot of money to begin with, others will say its no way enough. However, for me I believe it is sufficient because "I have developed the system myself and have tested it"... and this is the "key"

Why do I develop me own systems? Well, firstly, I enjoy the challenge and it keeps my brain active!!!.. but more importantly I am able to develop something which MATCHES MY REQUIREMENTS and PERSONALITY.

What annoys me is those which take a lazy approach and look to purchase a "black-box" system without knowing whether a particular system is suited to their own requirements. I've lost count of the number of posts on this site which asks "how good" a particular system is and whether people have made money from it, etc etc? Yet, (as you point out above) there are many different aspects of a system, profit only being one of them......

If these individuals really haven't got the time or inclination to learn how to trade properly themselves, they should place their money in the hands of someone else and let them trade it instead. Whether thats through a Hedge Fund, private Trader, it doesn't matter. However, purchasing an existing blackbox system and hoping that they trade well really is nothing more than a false sense of security...... The exception to this is when someone pruchases a system with the intention of taking it apart and learning from it. However, this is never possible with blackbox systems......

Just my view.....