Trend Following Manifesto

Michael Covel's daily trend following, economic & political commentary. Trend following trading has produced decades of above average performance. Learn how right here.

Trend Following Systems Available | All Michael Covel Books

6 Responses to “Pattern Hunting”

1. Mohammad Saleh Says:
   October 6th, 2008 at 3:44 am

   I think some of us tend to explain a complex system by wrong sizing the patterns and ignoring the range of sizes and their occurrence probability, in two ways:

   1) Over sizing a small pattern:

   To provide a simple explanation, and since there are a lot of patterns; we tend to focus on a small part of the picture that relates to the large picture while ignoring the rest, therefore there is a symmetrical relationship but a false probability measure

   2) Down sizing a large pattern:

   vise versa we would shrink the importance of a large pattern when we are in denial, for example losing a large sum of money in a single day in because of a mistake, so instead of feeling bad we down size the importance of what happened and believe that there is a high probability that things would come around like they do on the short term but our loss is actually a long term loss and the probability is low, we could increase the probability of recovering by increasing the position size but symmetrically we increase the probability of blowing out, so the mistake has to be sized correctly as a high probability loss that happened and cannot be reversed with a high probability win which includes a danger of another high probability loss

   Seeing UFOs may belong to over sizing small patterns, UFOs are expected to fly and have very high technical components that are even better than aircraft engines and so on, so we see them as complex large patterns although they are mere small light patterns or god knows what it was, maybe a Frisbee thrown really close but we perceive it as far and by perceived ratio it must be really big while it is actually not.

   We may understand that we cannot explain the whole picture, yet we can observe the hierarchy or probabilities that build the big picture and choose which size of patterns to observe and learn this range of patterns that affect us, also learning has a hierarchy, for example we would react to small high frequency events on the short run and react to low frequency events on the long run and in between there is a range of frequencies and their probabilities, and against these probabilities we build matching balanced reactions.

2. Simple Simon Says:
   October 6th, 2008 at 5:31 am

   A very interesting and timely post.

3. Mohammad Saleh Says:
   October 7th, 2008 at 3:08 am

   Instead of thinking how much to bet, why not think:

   How much can a market move in 1day, 2days, 3days….

   How many 10days (for example) the market moves 80% of its total moves

   How many 10days (for example) the market moves 50% of its total moves

   and so on, these are patterns, they repeat in nearly fixed percentages on the long run, fire up excel and create 10000 rows of randomly generated 1,-1 and on the side move a window of 10 rows or any size of rows along the 10000 rows, the % probability of occurrence of a move in one direction is a pattern.

   There is not magic, in science this is called the law of large numbers, and the limitations of the market moves are called the law of iterated logarithms.

   Forget the above scientific naming system which someone else have created, dont even read about it, just fire up excel and research, slowly you will recognize patterns that come in a range of sizes from 1 row patterns (100% moves) to any higher number of rows and the bigger the pattern that you are looking at the more it will contain of range of smaller patterns which build up to create the big one.

   On the long run adding the randomly generated 1,-1 rows together and dividing by the number of rows reaches almost 0 and if you get any historical stocks price sheet you will find the exact same thing that if you add the stock movement and devide by the absolute movement the result will be also nearly 0 on the long run

   Therefore market movement is the same 1,-1, but the only deference is that it is non linear which means, instead of 1,-1 you can say 3 to 5,-3 to -5 dollar as a daily average and keep adjusting this average

   What I am saying is, you don’t know how much the market ‘will’ move, you just need to know how much it ‘can’ move and the set of probabilities, then you will know how much to bet until the probability comes up and you make a profit or it may not come up and you make a loss and adjust your trading size

   Some people call them cycles, some people call them patterns, I really see no reason why I would call them for example moving averages, what are they? If I don’t know the details of how they work and see almost sure fixed statistics coming out of them then i will not use them.

   people build something like an indicator, a religion, a car, a building and give it a name, another one builds on top of it and gives it another name, them you come and learn the name and use the thing. You may know how it was built but you don’t know ‘why’ it was built (against what statistics, what will it affect and get affected by), just because it works doesn’t mean its right but it maybe right, there is always the simple raw price movement and raw cement material and raw math to learn from scratch.

   When every cycle finishes we start from scratch, otherwise why aren’t we born 20 years old with information ready in our heads? actually the basic desires are there but so many other things we are not meant to know because every generation has a deferent settings, for example if we are born with the knowledge of making a fire out of rubbing wood on stone then at age 5 the lazy people may still be doing that by age 30 but we are built to learn in a cycle (size of this cycle is from stone and wood friction to an advanced lighter and we choose the advanced lighter most of the time because it is the best range to trade in this cycle which is growing through generations and sometime when are camping we use the other range of making fire from wood and stone, we basically have more appropriate choice – Imagine rubbing stone on wood outside the office to light a cigarette!!!  you would get a arrested) and we die at the end of this cycle, others growing up learn slowly what we left behind, then they choose and adapt, and this is similar to market research, someone builds an indicator and you use it, but why did he build it? Why not break it down and rebuild it so you get a better statistical feel of its outcomes, it’s like driving a car, you learn from scratch… but imagine you get a chance know technical details about the car and past performance statistics, then you may notice something wrong in the car you are driving and either fix it or change it

4. George Says:
   October 17th, 2008 at 9:32 am

   In my opinion, under normal circumstances, it is difficult to understand and interpret reality. If it were not, there would not be so many disciplines that continue to study it such as Anthropology, Sociology, History, Philosophy, Education, Geography and many, many more. Therefore, it is hard to know what you are looking at. One person, under normal circumstances and mood may interpret something one way while another may see something slightly different or even worse something completely different than the first. Of course, under stressful times, as the article points out, the variation in perception is even worse!

   Basically, reality is to a large extent both subjective and objective in nature. We (humankind) can influence many aspects of reality and as the same time are influenced by them. It is often difficult to be able to draw the line between one and the other.

   To complicate the problem further, perception is different not only within the same culture but also between cultures and different types of societies ranging from the most simple to the most complex. It is disturbing to think that there are literally so many ways of seeing something and acting on it.

   Finance is one of many examples that exemplifies the type of ambiguity that interpreting reality can lead to. For example, there are many ways to trade and indeed there are many successful traders who are not trend followers. This is quite amusing due to the fact that we believe that our view of financial reality is the only valid one. For some contrast consider Buffet and others like him. In many ways, his style is completely different from ours yet very successful. How can we explain this in light of the fact that it is our belief that trend following is superior? If both styles are opposites how can they both be valid? I will not debate this question here due to lack of space but will leave it as something for all of you to think about.

   Sadly, it seems that in trading and finance in general what is the most important is not so much the style that one uses to achieve the goal but rather the money management rules that accompany it. For further details into these and other important points of interest, please consult Dr. Van Tharp’s book Trade Your Way to Financial Freedom.

5. Mohammad Saleh Says:
   October 22nd, 2008 at 7:39 am

   Hi George, reality comes in increasing scales, for example cotton comes from a plant and this plant is growing in a field and the field is being maintained and the workers them selves are eating and drinking from the income they get from maintaining the field so they could stay alive and make more money doing so, at the end of this cycle there are clothing and other industries where the cotton will be used to the main income comes from the buyers of cotton clothes for example.

   But you can’t ask people to buy cotton and make clothes; you have to create the product step by step.

   What you do as manufacturer:
   Give workers money to eat and spend on their families, give workers rest space and other needs to maintain them while on the plantation, then when they start working you will buy them seeds to plant and provide them with what it takes to grow cotton, you need all these steps to happen before you grow your finished product and sell it to your buyer for more than what you already spent.

   What buyers do:
   Buyer does what manufacturer did but on a deferent part of the main scale and sells the finished product (a pair of jeans for example), consumers buy the jeans because they need to wear it to go to work and get more money than the price of the jeans and other things they spend on.

   A pair of jeans is an end result of a many cycles of winning positions reaching maturity and paying off so the product is takes to the next step and so on until its last cycle finishes by throwing away the jeans (recycling the pair of jeans is a way to increase the whole cycle another step)

   Depending on your capacity being money or interest, you can choose to involve your self in any sequence of cycles of the main cycle, like they say in fractals of a chaotic systems, there are shapes that if you zoom into them would find self similar shapes in smaller scales building the main shape. For example you could own the farm or just sell the jeans, or take care of laborers, you could also be the labor him self, they all work on the same principle but on deferent scales.

   One more thing, these cycles can be measured, the more cycles you measure in relevance to your interest the more you will understand and be able to react because like Jesse Livermore and i hate to quote people because no one owns a saying:

   “there is nothing new in Wall Street. There can’t be because speculation is as old as the hills. Whatever happens in the stock market today has happened before and will happen again”

   This is like saying there is nothing new under the sun, because everything no matter how complex can be measured as a set of probabilities that pile up to create something.

   Predicting price movement is impossible but knowing the probabilities of price movement on deferent scales is very possible, for example for 40 years of data of a stock that moves a dollar a day in a window of 10 days; how much percentage% move would the price move a 50% move? 20% move, 10% move? If you calculate the probabilities you will find that they repeat in nearly exact percentages on nearly all stocks on the long run, if you know the probabilities you don’t need to predict.

6. George Says:
   December 12th, 2008 at 11:32 am

   Hi Mohammad,

   Thank you for your input and response to my post. I am sorry for the lateness of my reply.

   From what you have said, it seems as though you are a believer in cycles and perhaps what traditional technical analysts refer to as cyclical analysis. You are correct for saying that cycles do exist and there are many of them that we repeat each and every single day. Examples, include morning turning to night and later turning back to morning the next day. In commodities, there are cycles at work as well known as growing seasons that have made some traders rich.

   However, cyclical analysis does not always work due to the fact that there is more than one cycle occurring at any give time. There are small cycles that last from 1 to 2 months, medium-terms ones that last from 6 months to one year and long-term ones that last 1 year and beyond. the theory is that if all cycles fall in a perfect sequence according to harmonic and synchronicity principles then your have a better probability of successfully timing the market and knowing when the current movement will end. However, it is still highly subjective due to the fact that many cycles that we experience especially those in the markets are just as much effected by fundamental as psychological reasons. Therefore, the valuation of one specific asset according to this theory is still guess work at best. There are some gurus out there that can do it but most investors and traders cannot. Therefore, although you are correct in your assumption that cycles do exist, some, especially those in connection with the markets are highly subjective in terms of when they will begin, how long they will last and when they will end.

   Calculating probabilities for such phenomena is likewise also a subjective exercise. Besides, if you really wanted to do it all you need to do is use the rate of change indicator offered by most trading packages and use the following settings of 9, 12, 18 and 24 months at the same time and when all bottom or top at roughly the same time, that means that a new cycle is about to begin. Then, you can choose to exit either on one of the smaller ones touching the other top or bottom or the largest one touching the other top or bottom. It is up to you. Personally, if I believed in it for successful market timing I would not be frequenting this site.

   In addition, it seems that you are also a believer in what is referred to in sociological theory as structural functionalism. Basically, this theory states that everything in our society serves a function and all parts are needed to make society work properly. For example, a prostitute serves a function by providing a sexual outlet to those that need it, spreading disease keeping medical professional busy, getting arrested which keeps police, judges and lawyers busy, needing shelters keeping social workers as well as governmental officials busy by conduction research, finally, perhaps even construction professional will stay busy by building new shelters to help them to stay off the street. Basically, everything effects everything else and all things are interdependent on each other.

   However, functionalism suffers the shortcoming of being an idea that promotes that the rich and poor likewise serve a function and therefore, there is no need to change or modify any type of inequalities in our society. After all, if stratification on political, economic or social grounds serves a function to the advantage of some groups over others why bother to modify it? That is my point, reality can be interpreted in a multiplicity of ways not just one. Who is correct and who is not is difficult if not impossible to determine.

   In conclusion, reality is a circle not a square. You cannot conveniently draw a reference point that is advantageous to you without considering all the temporal, cultural, natural and historical complexities. I am not saying that you are incorrect only that you should open you mind more. If reality were so easy to interpret, there would not be so many historical conflicts. To end, when Socrates visited the Oracle of Delphi, it told him that he was the wisest man in the world. He disbelieved it and set out to disprove the claim. Ironically, he proved it by setting out to disprove it!

Leave a Reply