A Guide to Trading Systems
|August 5th, 2011, 10:16 AM||#1 (permalink)|
Trading for Fun
Futures Experience: Advanced
Favorite Futures: Stocks
Posts: 127 since Mar 2011
Thanks: 96 given, 85 received
A Complete Guide to Trading <under construction>
I ask you not to comment or read until the label "under construction" disappears. The text is a mess
and contains notes which would make no sense, if you read them now.
As the thing is getting bigger and bigger I will continue in a word processor.....
I am not affiliated with any of the companies I might list. If asked to open a demo account, you can do it with any broker you like.
How these guide differs from other guides:
run away from things like Free Trading Signals. Run as fast as you can.
Your capital is 10000, calculate 2% from capital..
part about extra inet cnn, two monis
flip coin experiment
ideal theoretical entry/exit
harder than normal job, full commitment, 12 hours a day, no vacation. mention things getting harder, need for constant adaptation, not a hobby.
no success in the beginning, the only time is when someone gives you the edge... to find edge you need lots of exp.
show a chart showing points where a certain strategy with 2:1 in opposite direction will or will not work under trend edge.
follow what I say or leave. I only want those who can follow what is said. This is an extra exercise on its own, if you can't do what I tell you, then how are you going to adapt to those new things necessary to succeed?
There is no such thing as a win or a loss in trading, since there is no end in this game, you either continue loosing money and continue winning. Trading is
Why trading is not a get quick rich thing: Trading is not a lottery. It is not possible with a 1000 dollar account to become millionair tomorrow. You will need to enter trades
over and over and even you have no edge, even if you are in big plus, things will go against you and you will lose money.
Imagine you trade with 1:1 RR ratio (RR stands for risk to reward), which means, you win as much as you loose. No commissions, no spread and your strategy generates 50% winners.
You winn $100, loose $100, +100 +100 +100 -100 -100 +100 -100 -100 .... = $0.
You will end up with exactly 0 dollars in the end. Now we add commissions of 2$ per trade.
You win $98, loose $102, +98 +98 +98 -102 -102 +98 -102 -102 .... = -$20 (minus 20 dollars)
Your commissions will eat you to death, if you have no edge. You can include commissions into your profits, but your win ratio will suffer, because there will be a trade that will not reach your take profit of 102 and the price will only touch the 100 and then go back.
Let's talk about what your win ratio should look like. It is easy to calculate the profits:
Number of wins x average win size - number of losses x average loss size
You win 100 times with 150 dollars average win and you loose 100 times with each loss of $100.
Having lost won 15000 dollars and lost 10,000, you gained 5000. Now, this is a very good result. It usually will be worse in real world.
If you divide dollars won over dollars lost, 15000x1000=1,5, you will get a so called profit factor. As long as it is over 1, you are in plus. Long time investors usually have higher profit factor than, say, scalpers, because their commissions are relatively low to the move that they capture and they don't have as many losses. However, they make far less money than daytraders or scalpers.
Imagine a scalper with a profit factor 1.15, slightly above one, For every dollar fifteen he has to lose one dollar. But since he makes several trades a day, and there are 220 trading days in a year, he will end up, trading twice a day, with 440 trades and $66 profit (non compounding). If to assume that his loss of $1 is 2% of his trading capital, then his trading capital is $50. That means, throughout the year, his non-compounding profit is +%132 and that is impressive.
If his capital is $50,000, his profits are $66,000 dollars and all that with only a profit factor of 1.15.
I chose this number, because my own historical profit factor for scalping is 1.1708, which is good for scalping. For my swing trading it is about 1.5.
Many think: all you need is only a 50% win ratio and your wins twice as big. Which is actually a profit factor of 2.
A profit factor of 2 in day trading is very hard, even the best traders I've met, who managed to have a four digit yearly return, had a lower profit factor. You can have a profit factor of 8 during a month or so, but if you trake into account many years of trading, the digit 2 is unlikely.
Don't forget, a profit factor of only slightly over 1 will bring money.
Why you won't overtake Bill Gates: It is possible to have a huge return on small account, however, if you have half of a billion in your tarding account, things are not the same. You will not find a fill for your orders even in the most liquid markets. You can't scalp or daytrade with a billion dollars account.
It is possible to double your account each year daytrading and scalping, but when your account gets larger and larger, you will get slipped on orders and your profit factor might go back to 1 and under that. If your take profit is 10 ticks and you pay 1tick commission and you get slipped in average 1tick per trade, you are giving away 20% of your profits plus you will most likely get slipped on your stop loss too. It might make you not profitable at all.
The solution is to focus on long term movements where your slippage will not even make out a 1% of your profits and long term, it is much harder to do 100% a year, because there are simply not as many trades avaiable to you as in day trading.
Power of compound.
if each of your wins bring you 2% and your net wins (wins minus losses) is 100, then your account will be 200% bigger. No, not exactly. The more money you have, the more money are those 2%.
The right way is to take the number one point zero two, 1.02 and puit it into power of wins. So, if you win twice, it is 1.02x1.02 (it is the same as 1.02^2), which is 1.0404. So you actually made 4.04% instead of just 4.
After 100 trades, that number is crazier: 1.02^100 (1.02x1.02.x1.02....hundred times) is 7.24 or 724% gain, which means your account grows 7.24 times and not 2 times (200%).
The more money you have, the more money you make. Many don't even want a 100% gain on their money. Too risky. Imagine what a 10% gain a year on a 100 million account is. That's 10 million!
In the end it only matters how much money you make. Success is measure in money, not in % gain, not in ticks, not in pips, not in profit factors. A +562% return on 13,000 is nothing like +23% on 28,000,000 account.
Why ticks, % and the rest is a wrong indicator of performance:
Do you know, that if last year you lost 50% of your account, and this year you gained 100%, your performance is not 50% but 0!
Here the math: your account is 1000 dollas, you loose half, you have 500 left. Then you gain 100%, you have 1000 dollars again.
Average yearly return etc are wrong factors to look at. In the end, money is your score. Don't think of money as of money that can buy things, think of it as of
tetris scores. The more you have, the better you are.
You need percentages, ticks, profit factors etc to build a strategy, to control your trades, to manage risk, but that's the money that tells you, how much you make.
If you read in forum he or she made 1000 pips this week, ask yourself, what does it mean? You can be in huge minus in money, but be in plus in pips. She probably lost half of her account with one trade that was only a 50 pips loss and then played with tiny risk and made couple of wins and had 1000 pips in plus, but her account is still down there.
Ignore things like profit in pips, don't even read those people's posts, you don't have time for that, because even the most valuable information takes time to find and to digest.
Trade as much as you can, at least to learn the movements of those bars, the way they develop in real time, at least to minimize the number of errors you make (wrong position size). You will soon find patterns that repeat over and over and then you will trade them. I strongly discurrage you from using any type of indicators. I can't forbid you anything, but there is all the information in the price.
I mentioned higher relative commissions. By that I meant that if your take profit is 10 ticks (10 pips) and you pay 1 tick commissions (I don't know spread on futures, in forex it is about one pip), then you pay 10% of your profits to cover commissions. That's huge. If you take 100 ticks profits and of course pay the same commission, you only give away 1% of your profits to pay commissions. Scapling is hard, because you will have to beat your commissions. So, even if you take more profits in ticks than losses, you might still be in minus after commissions are deducted.
net total for position, closed balance scalpers...
lift-off. Many are afraid that they will suffer a draw down from the beginning. Do avoid the effect of many of mistakes, including the one of not backtesting properly, start with a reduced risk and trade so until you are in a good plus with a room for a drawdown for the full risk trading, Never risk more than 2%. Risking 3% you have a 100% chance to lose all of your money within 10 years. 34 2% losses in a row will give you a 50% draw down. The same number of trades with 3% risk per trade will give you a 65% draw down and believe me, you will suffer some draw down of similar magnitude, if you trade with 2% risk per trade with more than 100 trades a year.
How much risk: Except you are a position trade, you use your closed equity to calculate risk of money per trade. As a position trader, someone who is riding those long trends in oil, indexes, gold, many other commodities, they must use total equity to calculate position size and risk. Total equity would be closed equity plus floating Profit or Loss.
bars vs line vs candles vs rest
it is not enough to follow rules. To have succes one has to do two things: 1. Follow rules 2. Have a profitable system with a set of rules.
The first one is obvious. If you created a set of rules, that show profitability during backtesting, otherwise you are not trading that system and whatever you trade is not that system.
Following rules is like a natural selection, those who don't, are out. I would not recommend to read any of the psychology books on trading, because all you need is to follow the rules. It is very tempting to close your position earlier or move stop loss just one tick further. You can bite your fingernails off, roll on floor crying, beating your dog to death, but as long as rules are followed, you are there, congratulations. Keep what's going on in your head and the actual system apart. You follow the rules and the system will never "know" what was going on in your head. Now, step two: a profitable system. This is the hardest part. You have heard about the turtle experiment: if you are give a set of rules (entry rules, risk etc, basically a system is a set of rules), so, if you are given a system that works and you follow these rules, you are profitable. However. If you are NOT given these rules, they are very hard to find. Very, very, very hard to find. You need a lot of screen time and experience to grasp those market movements that occur over and over. The reason why nobody is profitable from beginning on is exactly this one. You can't be profitable from beginning on. You can make money from beginning on, but you can't be profitable from beginning on. This is very important to distinguish. You can make money and be not profitable. That means, you are having luck, a Gauss distribuiton victim at making money. The heads of a coin are dominating the tails and you are making money. And you call your friends telling them how easy that is, how good you are, 5 minutes a day and soon you are a millionair, you put lamborghini in an online shopping car and a house in Malibu. Soon this will be over and you will collapse to the floor crying that things don't go your way. This is because you don't have the most important part: as system that works.
I also have to tell you this: it is getting harder each year and each decade.
always test more than 500 trades and always test different periouds of time. It is not enough to test 2011 until now (August). Go back a few years and test your strategy there. If everything works, only then test many years of data (to save time).
I must disappoint most off you. You will HAVE TO use bars. I know you don't see a thing there and blah blah blah. I don't allow you to use candles. Use bars. Candles put too much weight on opening and closing prices and they are the least important. The most important ones are high and low. You can see many charts on the web with trend lines drawn through the wicks of the candles, completely ignoring them. This is WRONG. Never do it, please. Never. exercise draw line.
Now make your bars of one color. No difference between an up or down bar. I have my reasons behind it and you will know them, once you are experienced enough. You need to learn to read price.
Things I hope you will never use are:
CCI, RSI, Stoch, MACD, Trix... any of the overbought indicators.
The worst you can use is a traffic light indicator, with two colors, one for sell, one for buy.
I strongly discourage from using Bollinger bands.
Exercise: make a most probable line chart out of bars.
Myths about myths.
There is a myth: A perfect indicator exists. And there is a "wisdom" that states the opposite: indicators that will make you rich do not exist.
I am not a fan of that wisdom, because I know, there are little things, for example and indicator, that exists and that has an edge and that is profitable and that is making you money. Holy Grail is a word often used to describe a system that is fail proof and always makes money. I not only believe it exists, I know it exists. Here an example: Imagine, you are a first person to invent moving averages and all you need is to buy/sell on moving average crosses. Do you actually know, that it would bring you money until the moment where too many people know what it is and try to take advantage of it, which will turn into disadvantage and eventually the death of a simple moving averages cross?
Holy Grails exist, but the reason you don't have it, is because once you have someone else's Holy Grail it stops being a Holy Grail. These are important words: Holy Grails do exist, but you will never find that they do until you actually create one. That sentence must be your guide in trading: you need
Psychology during placing a trade:
Do you feel anxious, are you scared, you don't know will it rise or fall and the long you are about to enter, you are desperate, now, maybe sell better? Moving averages are up, but RSI is falling down, news show short, but MACD is a strong long, it's a daily pivot, it must bounce and fall, but fibonnachi must retrace and go long....STOP!
Don't place a trade if you feel any of the symptoms above. I can't stress it enough: find ONE thing that works and make a rule of entering according to that one thing and always take that signal regardless of what news or fibonacchi or your friend tells. Delete fibonnachi and pivots and everything from your chart and leave only that which you made rules for.
Discretionary trading: bars don't tell the whole story. Within one bar price can move in a differt number of ways, yet this information is lost inside the bar. Discretionary trading cannot be taught. From experience you gathered over long time you will see things happenning over and over and you will start wanting to take advantage of them. You will start noticing things after some month and maybe six months of trading and that may be equivalent to a thousand hours of screen time if you trade a lot you might find those repeating patterns. This is not a part of my discussions.
Picture of ideal entry.
draw downs (importance of it)
PART ONE - Getting Started
PART TWO - Practice
PART THREE - Psychology
PART FOUR - Trading Systems
PART FIVE - Final Exam & Preparation
My recommendation is that you also go through other guides and learn from other sources. Always read about the same subject in three or more different places. And always do the practical work. You read about MACD, you go to your charts and you put it on there and look how it looks.
PART ONE - Getting started.
Before I say a word, we need a demo account. I strongly recommend opening Oanda Fx Practice. Market index RBS uses the same platform as Oanda, you can go with that one too. You can use any of the demo accounts - I do not care, but the fxTrade platform let's you instantly see how much money you will lose or gain at certain stop loss sizes. It let's you move SL and TP right on chart (pretty good for a free platform). It's more user-friendly. If you don't know what TP and SL are, don't worry, I will explain in detail.
I chose a forex broker, because trading currencies is hassle-free (no problems short selling, you don't have to wait for market to open...) and of the easiness of opening a demo account. (if you don't know what short selling is, don't worry). But lo and behold, it is better to practice with real money. Many brokers let you open an account with $1. It will not be different from trading a larger account, but it feels a lot more real than demo, even if small money is involved.
-explanation of setting up a platform with pictures.
PART THREE - Psychology
1.I am not gonna talk a lot here, because there is one sentence that will improve your trading considerably: Always blame yourself.
This is a very very important sentence. The minute you start thinking someone is manipulating the price or a broker hunting your stops or just hell with the markets, this is the minute you are blaming someone else for your OWN mistakes. You lost a lot of money, because your position was too large? You did not follow rules? This is the most imporant sentence that you will ever find in any other activity where you need to succeed and be among the best. You did not follow your rules, blame yourself, you did not do your research well, blame yourself. When you do blame yourself, you are more likely to improve. It is like being your own coach. Many say you need to be disciplined, but what does it mean to be disciplined? You can't be disciplined until you start taking responsibility for your own actions and that is through accepting your own mistakes, accepting that you are the one doing them. You can read me between the lines: If you took someone's tip and lost, blame yourself, not him! Do you see, how this sentence actually prevents you from taking any tips? This is a one in all rule which if followed will trim your behaviour in a short time and you will make a huge leap forward in your education.
2.Assignement: Read The chapters one and two in Way of the Turtle by Curtis Faith.
3. Place .... trades ....
4. Read chapters 3-6
This is a full guide on Trading. For some of you this might be the most important thing you've ever read. This will be (I am still writing this guide) the most complete creation I ever wrote about trading. Here, I want to point out the importance of testing and changing systems, the reasons for this and so on. Unlike our Mike, who thinks it is important to stay with one system, I am not of this opinion. You will see why.
Let me go slightly off topic and talk about success rate, luck and profit/loss distribution. You have heard about 5% making money, 95% loosing it and numbers close it. While in Stock Market the number is higher, in forex lower, it does not matter.
Let me prove this statistics wrong: Go to forex factory (it's huge), and look at all those struggling people. Do you really think, that every 20th person makes money trading? I've learned trading to be freaking hard (I will tell you why later).
Let me make a claim: Less than 1% of traders make money trading. Tom Baldwin (professional trader), whom we all know, would back me up on that.
Yes, I know, your grandma bought Amazon shares, because she thought it was actual Amazon river she got to own a part of... And then she sold it and made money! And your uncle...
Let's talk about distribution. Imagine, we all were stupid and could only trade by coin flip. Head is a +1% win minus commisions on your account, loss is -1% minus commisions on your account. You can actually do that experiment at home, it's fun. And it works. So we have 100 people and we all flip the coins. Because of commissions, everyone is loosing money on average. Always and everyone. Yet after the experiment you see, that some people have extremly poor performance, while others have extremely good performance. The majority is in minus though (because of commissions).Those with good performacne had many heads, those with bad many tails. This is what I mean by normal distribution of profits. If someone of you makes money, this is most likely luck. I don't say it is 100% luck, I say, most likely.
The reason 5% make money, is luck. An outcome of a random process. Let me tell you this: if you don't know why you make money, if you can't say me in words what is your edge, you have no freaking idea how to trade profitably and you simply are experiencing a distribution of profits in your favor, which will, when enough trades are done, move against you and you will lose money just like everyone else. Ask yourself: what's your edge? Do you have one? Do you even know, what an "edge" is?
For me this is a joke: There are people, who have been profitable for the most part of their life, without any freaking idea how to trade. It is more usualy for people who make very few trades every year, because there is not enough data to blance their profits back to zero. But there are also people who win lotteries. They are all in the same pool.
Yes, 5% are making money, but they don't actually know poo about trading.
And there is this less than 1% crowd, which knows how to be profitable no matter what. They are not subject to random distribution. They always make money. You wanna be them. It requires a lot work (unless you have Bill Eckhardt telling you what to do) to find that which works.
I've been through a lot. I've learned trading to be a neverending journey, because things stop working.
Go to forex factory and look at all 1785 of systems. You will not be able to test all of them in a life time, but let me tell you this: none of them work. That's a strong thing to claim and needs a thorough proof. Many of you will not agree. If you don't, read the luck part again and look at it as the reason why a system works and why it does not. According to the second law of thermodynamics, there is a tiny chance the pot will freeze if you put it on a hot stove. There is always a chance a crappy system would have its own bullmarket of a life time.
By no mean am I saying you can't make money trading. This is not what these words say. My words say, it is hard and for all of us nearly impossible to find something that works, because something that does is usually something very subtle, yet simple and beautiful.
The Death of a System.
The reason something works, is because it has an edge over the rest of traders. An edge is a knowledge of something. Something you can actually say, show, explain and then transfer into a system. There is a reason behind an edge. Low commission/spread is one type of an edge. Over time, it earns you millions over those who have higher commissions. Insider information is a type of edge. Trend is an edge. It is an advantage over others. Something you know or have others don't. It is nothing magical. Now, once everyone knows and has it, it is no longer an edge. If everyone has low commissions, nobody has an advantage over the other in that field (commissions). If everyone gets the same information, that, say, for example a company is twice as worth as it is now, at the same time, then the price soars to where it belongs. Those with execution edge will get in faster, but the actual insider information is no insider information any more and therefore, not an edge, not a advantage over the others.
This is very important to understand. Forget RSI and CCI and MACD and other cow poo and throw them out of your head. You need to understand it: If you don't have an edge, you are a subject to a random profit distribution. You may lose money or you may gain some with lower chance. But the ability to make money is not there. Why would you do it to yourself?
Now about the death of a system. A death of a system simply means, that the edge of that the system is known by many and no longer works.
A good example is a dual moving average system. It worked great for the last hundreds of years and it stopped working around 2003, because of the overuse by the majority. I attached a picture of trades for the last 15 years for all liquid futures. Pictures 1 and 2. Picture Copyright Trading Blox.As the use of the system increased, its performance worsened. Really, if you were the guy who first discovered moving average cross, you would have an edge. But today, it is gone. And if you think it exists, then only because some are experiencing luck with the system. In the lower part you can see drawdowns of a system and that they increase ass years pass. The logaritmic scale gives you a better picture of consistency.
The only rule choosing a system: It must work in the recent market and it should have always worked in the past. It means, if you are to trade potatoes in 19 century, you must be profitable and if you are to test last year, it must have worked there too.
If it did work in the past but stopped recently: it means too many are using it and the main edge is gone.
If it did not work in the past, but does work recently, it is a crappy system that is experiencing abnormal number of heads in the row. Soon the tails will kick in and you gonna lose money.
While there are many edges: execution edge, commission edge, trend edge... There is this system edge that is the main edge. The edge that in the end brings you profit.
How to find main edge (or the system edge):
Aks yourself, what is one (or several) certain thing that has always worked? For example, insider information works. If you get it, and it is not known by others, you make money. Clean and simple. And illegal. That too, hehe. But for demonstration purposes, works.
All this is the reason for a need to be looking for a new system all the time. I've tested I don't know how many systems. Sometimes I go puking from sitting forty hours all in one weekend at the computer testing them. Open your eyes. Splash some cold water in your face. Noone will give you an edge, unless your daddy is Richard Dennis. You will not find an edge from ANY of the systems on the forex factory or the whole world wide web, and how big that is. Theoretical chance exists, especially on smaller quality forums like this one, that something still works. It's your task to search. All these systems are dead. Do you understand me? They are dead. Most of them used to work, when very few people knew them. But they stopped, because it's there lying on the table and everyone using it.
Put yourself 100 years back and imagine you invent some indicator and it works and you make money. Today everyone knows it and it does not work anymore. What do you have to do today? Same as 100 years back. Invent something that works! Things that are known now cannot work. They can't!!! You need to find something that does. Something your own. Computers for example are an edge, because they can execute and close positions faster than humans, especially if the trade is to be held for a second. That's an edge over the others. Learn to program strategies, it will make testing thousands times faster.
The ability to find an edge is proportional to:
-knowledge (fundamental analysis)
Experience is very important. After some time you will start seeing patterns that occur over and over. I am not talking about Japanese candles patterns. I would not even recommend using candles. Use bars, if you are a novice. You will see much more. I am talking about overall partterns, the way the graph looks. It is always the same. Same things happen. Not often, but they do. You need a lot of screen time and experience to actually see it and take advantage of it.
I believe, it is very unlikely that one will be profitable from start on. It is even very unlikely that one will become profitable at all. Even Tom Baldwin paper traded for six month, before he made his first trade. The exception would be if someone gives you an edge. Turtle Trader would be an example here. You are given an edge, use it, enjoy. But to find it yourself, you need to work your ass off. Lots of work, dedication, disciplin. Even then, there is a chance you will never find it. The chance you don't is about 99%.
My recommendation is to become a investor rather than a trader. You see iPhones everywhere, you buy Apple, you hold, you have profit, you close. Or take advantage of super low prices after financial crashes. Or just catch trends in Gold, oil, etc.. They are the easiest money. But you you want to actually trade and not invest, good luck.
Good Luck. I hope though, it won't be just luck, but some real edge going on there
Last edited by ReaM; September 4th, 2011 at 08:42 AM.