Getting Started Trading Systems The 10 Power Principles of Successful Trading Systems

dbphoenix · Apr 29, 2005

Rhody Trader said:
That said, just because someone has an agenda it doesn't mean they can't contribute positively to the knowledge of the trading community.

I find it interesting that novices regularly complain that people who know what they're doing never "share". However, whenever one of those people does share, whatever he or she might share is immediately suspect if he or she has or ever has had any product or service for which he or she has ever received any remuneration. Which is so often why those who do know what they're doing never "share".

jmreeve · Apr 29, 2005

Where did you get that idea? He says they test up to 25 systems a month, and that the point of the article is to know what to look for, and that a system created by oneself can be just as good as one bought.

If they test so many systems they must have found more than one that made money so why not offer a
broad selection of systems with detailed performance reports for each?
Instead they offer a single system designed to trade very frequently.
My suspion is that this company has a profit share deal with whichever broker it recommends
to trade the automated system.

A better website for systems is Chuck LeBeau's which provides a selection of systems
with detailed reports on each and at lower cost.

Trading lots of systems at the same time won't save someone who can't make a consistent profit off of one..

It can help a lot by smoothing out the returns and reducing drawdown.
It is also easier to trade multiple systems because you are not putting all your hopes into one market and set of rules. This makes psychologically easier to pull the trigger or take a loss when necessary.
So trading multiple systems can save a trader who can't make a profit off of one.

dbphoenix · Apr 29, 2005

jmreeve said:
My suspion is that this company has a profit share deal with whichever broker it recommends
to trade the automated system.

It is also easier to trade multiple systems because you are not putting all your hopes into one market and set of rules.

I'm still not clear as to what any of this has to do with the content of the article.

As to hopes, the point is made in the article that one ought to be trading on the basis of tested results, a point with which I agree.

But, again, this is a debate over an extraneous issue. True is true, regardless of where it comes from. A trader can determine the truth of the article without ever spending a dime.

jmreeve · Apr 29, 2005

Why do you disagree with #5?

And for that matter, why should someone trade lots of different systems?

It was understood when this article was being prepared for publication that it wouldn't fit with everyone. The hope was to generate discussion and healthy debate so that traders with different views could share and exchange them for the benefit of everyone.

I disagree with 5 because trading very frequently tends to result in lots of cost and comissions at the expense of profit. Short term systems usually have a poor yield compared to market exposure and so it often makes more sense to trade longer term systems and get the consistency by diversifying. However,
a mix of short and long term systems designed to complement each other is very beneficial.

ZDO · Apr 29, 2005

re: “Trading lots of systems at the same time won't save someone who can't make a consistent profit off of one…” and “If logic applied, then your propositions would be true. But markets aren't governed by logic. Therefore, even automated systems must be adaptable.” - Lacking an adaptive system, one alternative is to trade multiple (2-5) non correlated systems and make 'weighting' adjustments often...

Some comments on the article for newbies:

re

rinciple #1: Few rules - easy to understand
Simple (Principle 1) mechanical systems are just that - they are simple. This means they get slapped around a lot by market moves that turn out to be 'false' to the system. re: "Think about successful floor traders: The only tool they use is a calculator, and they make thousands of dollars every day." Actually they use the tool between their ears and any 'system' calculations they do are very insignificant to producing their results...

re: Principle #2: Trade electronic "When you trade open outcry markets (non-electronic) you might have to wait a few minutes before you receive your fill" - this is subtle guidance into the fastest timeframes. As the many are piling up in the short time frames, consider using longer timeframe systems where it doesn’t matter if you have to wait a few minutes for a fill and your stop isn’t a few ticks away. and liquid markets well – duh !!

re: Principle #3: "Make consistent profits You should always look for a trading system that produces a nice and smooth equity curve" – yet to really ‘expose’ yourself to the possibilities of geometric returns (which is why your money is here and not in a bond or mutual fund, right?) you need to be ready to routinely accept large (like the 20% examples he used) drawdowns …

re: Principle #4: Maintain a healthy balance between risk and reward - ‘Your risk of ruin should be always less than 5%, and your chance of success should be 5-10 times higher, e.g. if your risk of ruin is 4%, then your chance of success should be 40% or higher.’ Nice principle. Thanks.

re: Principle #5: Find a system that produces at least five trades per week ‘The higher the trading frequency the smaller the chance of having a losing month … That's the goal of all traders: Having as many winning months as possible.’ - this is not a principle that can be generalized to months (or whatever time frame). Makes this principle total b*s* and by association calls the others into question. ‘In the meantime, how do you pay your bills?’ Bottom line, like in starting any new business, you should be capitalized well enough to ‘pay the bills’ for 3 years before beginning. In the long run, having staying power (financial and inner) is about 70 times more important than your initial ‘systems’

re: Principle #6: "Start small - grow big..." Like a good proper guidance counselor in school, this part of the article guides you to the great ‘middle’ where hopefully you will be secure and safe from the sharks. In my experience, this is a sure path to being one of the 80 that feeds the 20… ‘That's why martingale systems don't work.’ again a massive generalization that calls the whole article into question because optimal position sizing is a function of the expectancies of each system ie martingale works just fine for certain systems…

re: Principle #7: Automate your trading ‘Emotions and human errors are the most common mistakes traders make. Clearly, they are to be avoided by any means possible. Especially during fast markets, it is crucial that you determine the entry and exit points fast and accurately; otherwise, you might miss a trade or find yourself in a losing position. For that reason you should automate your trading and look for a trading system that either already is or can be automated. Automating your trading makes it free of human emotion. The buy and sell operations are all automatic, hands-free, with no manual interventions and you can be sure that you make profits when you should, according to your plan.’ ie Order one of their systems, sign up for TradeStation today, have fun suppressing your fears … actually you need practice and experience at manually managing yourself and your positions in fast markets regardless of what your 'system' is...

re: Principle #8: Have a high percentage of winning trades ‘Your trading strategy should produce more than 50% winners’ – actually your system should have a solid positive expectation and you need to be internally ready and able to run it across a medium sized population of REAL trades before you even think about tweeking or dumping it – ie ignore ‘confidence’ or lack of it, take the pressure, you will need the practice if you’re going to stay in the game for the long term… ie be ready to “eat bitter” (a Larry Williams quote of some dead asian guy. )

re Principle #9: "Look for a system that is tested on at least 200 trades..." - another OK, but woefully inadequate and misleading, generalization unless you’re shopping to buy a good black box…

Principle #10: "Chose a valid backtesting period..." well duh ! if you are going to test - do it right! back, forward, in and out of sample - maybe just do whatever it takes to see the trees and do anything necessary to keep you from seeing the forest

The unspoken context of this article (and most writing about systems) is it’s all about entry techniques.
Mechanical or discretionary, doesn’t matter - what makes any trading ‘system’ work is what the ‘system’ does AFTER the entry and that simply is not simple… 'even automated systems must be adaptable'

hth

Rhody Trader · Apr 29, 2005

jmreeve said:
I disagree with 5 because trading very frequently tends to result in lots of cost and comissions at the expense of profit. Short term systems usually have a poor yield compared to market exposure and so it often makes more sense to trade longer term systems and get the consistency by diversifying. However,
a mix of short and long term systems designed to complement each other is very beneficial.

OK. I can go along with the cost aspect, though that has certainly become less of an issue the last few years with the expansion of electronic trading.

I can also see where there can be value in running multiple systems, so long as they do not make things overly complex for the trader.

While it is probably also true that shorter-term systems have a lower yield to market exposure, would that not get made up for with higher frequency?

The fact of the matter is there's more profit potential (in absolute terms) trading short-term than trading long-term. Markets move in a jagged fashion, not in straight lines, which is quite easily demonstrated by comparing distributions of returns over various time periods. Just look at the graphs below. The fact that the range of the 5-day distribution is nowhere near five times the width of the 1-day distribution (in terms of max to min) tells you the market almost never moves directly from point A to point B.

dbphoenix · Apr 29, 2005

I don't know that I'd call 5 trades a week "frequent", though it depends on what kind of trader one is. In any case, the profitability is determined by the strategy. When the setup presents itself, one takes it. That may be twice a month or 10 times a day or . . .

JumpOff · Apr 29, 2005

I'm still confused. An automated system that makes consistent profits would have to recognize and adapt to a variety of market situations. If it worked, wouldn't it eventually suck in all the available capital?
JO

dbphoenix · Apr 29, 2005

JumpOff said:
I'm still confused. An automated system that makes consistent profits would have to recognize and adapt to a variety of market situations. If it worked, wouldn't it eventually suck in all the available capital?
JO

You're leaving out the human element again. What works in theory is largely irrelevant in practice, which is why "backtested" systems that seem to be just great often don't pan out because they can't be traded as is, which is probably one of the more important points made in the article.

Granted a truly "automated" system that will not tolerate any interference might "work", but to put such a system in place would require extraordinary capital and gigantic cojones.

frugi · Apr 29, 2005

Principle #7: Automate your trading

Emotions and human errors are the most common mistakes traders make. Clearly, they are to be avoided by any means possible. Especially during fast markets, it is crucial that you determine the entry and exit points fast and accurately; otherwise, you might miss a trade or find yourself in a losing position. For that reason you should automate your trading and look for a trading system that either already is or can be automated. Automating your trading makes it free of human emotion. The buy and sell operations are all automatic, hands-free, with no manual interventions and you can be sure that you make profits when you should, according to your plan.

Automating it also precludes the use of the human brain, which for me is one of the most important aspects of any system if it is going to be adaptable - context-dependent - and thus work long term over different conditions. Computer pattern recognition is simply not as effective, at least not yet. It is still possible to have mechanical rules to determine entry and exit, thus removing a generous slice of emotion, but to allow no discretion at all seems to me to be a shameful waste of one's own resources.

Surely it is only "crucial to determine the entry/exit points fast and accurately" if you are going to take the trade? If a market is too fast to be able to initiate a position manually then I will not take it and thus not suddenly find myself in a losing position. I do not see this as a problem as there is always another opportunity.

As usual this comes down to the old chestnut of mechanical vs discretionary which has been done to death on other threads, so I'll leave it there. Otherwise, imho, a reasonable article that has, at the very least, stimulated an interesting debate.

Mr. Charts · Apr 29, 2005

Automated systems are likely to remain the Holy Grail, imho.
If they are adaptive they may survive longer, but ultimately high automated system efficiencies must lead to market modifications to neutralise those systems. Rather like weapon systems engender defences and those defences result in modifications to the systems and so on. In other words system efficiency is somewhat degraded.
I wonder if there is anyone here who could take this a stage further to look at A.I., Neural Networks, so-called Expert systems and their latest development, NNESs.
I can't take the discussion on myself as my technical understanding of computer generated data and developing automatic "systems" from that data is not deep enough.
My own experience tells me the way to make money from the markets is to recognise patterns and to be able to read sentiment and its manifestation and act accordingly.
Otherwise for many, especially the Holy Grail seekers who want to make money automatically with minimal if any actual effort, it is a matter of being forever travellers.
Richard.

dbphoenix · Apr 29, 2005

Traders are like a body of water picking away nit by nit at a dam. If there is the slightest inefficiency in the construction, the water will find it. Even if the monitors make a satisfactory adjustment, at least some water will have accomplished what it sought to do by finding the inefficiency (if no satisfactory adjustment is made, of course, the dam busts loose).

Or, traders are a tomato. Squeeze it hard enough and it bursts. Or a water balloon. (Choose your metaphor)

Point is, the market is organic, not inorganic. The profit motive will find a way.

commanderco · Apr 29, 2005

There really I not much I can say that has not already been said.

However
"Principle #7: Automate your trading"

Quite frankly, I am smart enough to trade but too dumb to write a code to cover all the conditions that I take into consideration at the entry, the middle and the exit of a trade.

jmreeve · Apr 29, 2005

I think principle 7 is valid but not for the reasons stated in the article.

The only real reason for automating a system is to increase execution speed and reduce slippage.
Automation also allows a single trader to trade lots of systems and gain the diversity benefits
of trading multiple systems.

ZDO - Your post reflects many of my concerns about the points made in the article.

My other concern, which may or may not be founded, is that the article is not wholly independent but is a biased view from a company with a strong interest in generating brokerage revenue.

jmreeve · Apr 29, 2005

While it is probably also true that shorter-term systems have a lower yield to market exposure, would that not get made up for with higher frequency?

The fact of the matter is there's more profit potential (in absolute terms) trading short-term than trading long-term. Markets move in a jagged fashion, not in straight lines, which is quite easily demonstrated by comparing distributions of returns over various time periods. Just look at the graphs below. The fact that the range of the 5-day distribution is nowhere near five times the width of the 1-day distribution (in terms of max to min) tells you the market almost never moves directly from point A to point B.

The point you are making is valid in that if you can capture all the small moves accurately you should be able to generate more profit. However, in general I have found that short term systems tend to have poorer yields than longer term systems. This may be down to the simplicity of these systems and the increased difficulty of timing small moves or it may just be down to increased comission costs.

Whatever the reason, lower yields relative to market exposure means you need to take on more leverage
to get the desired returns and this increases your risk of ruin.

peterpr · Apr 29, 2005

Mr. Charts said:
Automated systems are likely to remain the Holy Grail, imho. <Snip>.

Seconded. But no doubt there will always be those who make sterling and ever more complex attempts at building/finding it. I doubt they will be allowed to corner the market in anything though. The regulators are there to see to that among other things (The Hunt Brothers and Silver are an interesting case in point.)

Here's my two pennies worth:
I've had minor professional involvement with AI over the years. There is some pretty clever stuff out there and its power is becoming increasingly accessible to PC based systems. The basic approach is really to 'curve fit' historical data then apply various adaptive/feedback algorithms to adjust signals as new real-time data is processed. I guess the illustration par-excellance of what can be profitably achieved on a large scale without the monopoly effects (hence market destruction) posed by an earlier post is that of LTCM. At its heart were very clever automated systems running mainly arbitrage, spread and carry trades on wafer thin margins in world-wide markets and on pretty well every conceivable instrument. But, in spite of being overseen by the Nobel prize winning discoverer/inventor of the US options pricing formula and the best computerised AI skills money could buy, the markets got their measure after a few short years and - wham! melt down.

IMHO there aint no Holy Grail out there so far as trading systems are concerned. There are just tools. Some superb and very clever tools too, that can do for your trading what JCB did for digging holes in the road. But - you've got to know where to dig to avoid those high voltage cables, water pipes and drains - you've also got to know which lever does what - which isn't a five minute leaning exercise.

I still say there's no substitute for knowing your instruments/markets; knowing who the participants are and what motivates them (who's going to take the other side of your trade in other words); understanding the interplay between options, futures and the underlyings - sentiment, news etc and watching price/volume action like a hawk using the best tools available. The tools simply provide additional ways of seeing and judging price/volume. But it's judgement and the ability to act dispassionately on it that produces success - maybe a bit of luck along to way too.

Trader333 · Apr 29, 2005

The way in which I trade means that it is just not possible to backtest or fully automate my trading approach and there are reasons for this. No instrument repeats exactly the same conditions from one trade to the next as there are always differences and this would be extremely difficult to bactest or automate in my view.

However, there are principles and guidelines as to how I am likely to respond but there are not absolute rules for how to trade in every circumstance. It is precisely for this reason that I think computer based trading approaches ultimately fail. They will work within a set window of parameters but when other factors start influencing the overall picture they almost always fail. I say this as someone who has done a very large amount of system bactesting and have developed fully automated approaches but the only way that has consistently proved profitable for me is not system based.

Paul

Roberto · Apr 29, 2005

peterpr said:
IMHO there aint no Holy Grail out there so far as trading systems are concerned. There are just tools. Some superb and very clever tools too, that can do for your trading what JCB did for digging holes in the road. But - you've got to know where to dig to avoid those high voltage cables, water pipes and drains - you've also got to know which lever does what - which isn't a five minute leaning exercise. I still say there's no substitute for knowing your instruments/markets; knowing who the participants are and what motivates them (who's going to take the other side of your trade in other words); understanding the interplay between options, futures and the underlyings - sentiment, news etc and watching price/volume action like a hawk using the best tools available. The tools simply provide additional ways of seeing and judging price/volume. But it's judgement and the ability to act dispassionately on it that produces success - maybe a bit of luck along to way too.

_Extremely_ well said, Peter. Profound stuff for a Friday evening.

shazbots · Apr 30, 2005

Systems work people dont

Simple as that if you ask me.

roberk · Apr 30, 2005

jmreeve said:
I
My other concern, which may or may not be founded, is that the article is not wholly independent but is a biased view from a company with a strong interest in generating brokerage revenue.

I thought the purpose of the article was to sell their $600 pdf file? Where did you see them promoting a brokerage?

Getting Started Trading Systems The 10 Power Principles of Successful Trading Systems

Guest Author

Well-known member

Guest Author

Well-known member

Active member

Senior member

Guest Author

1

Guest Author

1

Legendary member

Guest Author

Well-known member

Well-known member

Well-known member

Established member

Experienced member

Junior member

Active member

Similar threads