@KDASFTG: Thank you so much for sharing your opinion. I respect your post. In the past, I have read a few of your posts and they were all very wise. This one is no exception. Please allow me to answer you, which could also help me identify the error in my ways. I have to absolute agree with you here. If I were to take the first statement a little further, I would add: The convergence of actions resulting from those varied beliefs held by various market participants is most likely the reason for directional movement in the market. However, I also see that you are using the above quote statements as a stepping stone to offer your critique of my belief in general. I appreciate it, and I will attempt to answer them to the best of my ability. Seems to me that there are a few words we might have to define before going further. These definitions will help us understand each other little better. I strictly define Technical Analysis (TA) as analysis of geometric patterns based on chart formations. I define Price Action (PA) as analysis of price movements, and is not related to geometric patterns. When I talk about "pattern" in the context of Complex Systems, I do not restrict it to geometric patterns. A pattern, in this context, could be statistical patterns, PA patterns, TA etc. The way I trade, I use PA patterns. A PA pattern is not necessarily a single horizontal line, but usually a sequence of events that I track. So, yes, they have A->B->C kind of sequence, but the alphabets are events and not geometrical structures. Now, that we have the definitions down, let me try to answer you. The fundamental assumptions you make in the above passage is that one would know what will happen just because one tries to detect a pattern. Your argument proceeds by talking about mental conflicts that could arise. However, the assumption you make is not necessarily correct. Patterns does not equate to consistency. It is true that one could take a leap of faith into believing that a pattern would occur before the pattern is complete, but to do so would be a grave error in perception. In trading, we cannot assume something will happen until it has happened. I, for one, do not make that assumption you have made in your argument when I trade; so, the question of incongruity does not arise. I am not sure if I am understanding you correctly here. So, please allow me to paraphrase you, and then make a comment. If my paraphrasing is incorrect, please feel free to correct me. I understand you as saying that a sequence of events (my definition of a PA pattern) has no correlation to the outcome of an individual trade. Based on this, you conclude that the result of a single trade is random. Comment: If I am able to correctly detect the sequence, I can assure you that the result of the trade is not random. The result can be forecast-ed to a great deal of accuracy. The reason, as I mentioned in my original post, is that it is just not me who is looking at this sequence of occurrences, but a plethora of market participants who all could have different beliefs, and as the price unfolds, those varied beliefs all converge into a cohesive unidirectional action. Some traders call this "getting into the mind of the market"; I prefer PA pattern! May be an example from today could provide a little more clarity. I trade ES. At 10:23:35 CST, a level was formed that interested me (2110.50). Based on my homework on this particular PA pattern in this context, I knew what to expect. I was expecting a "test" of this level. I did not know if that level was going to hold or not hold. Mind you, I was already in a Long trade at that time. I was not ready to close my long trade because of my expectation of a test of 2110.50. I can only close my long when my trading plan tell me to. Also, just because I expect a test of 2110.50 does not mean that the test will happen. But if it happens, I will be prepared for it. The pattern around 2110.50 provided me this: After price touched 2110.25, (seq-a) if a SELL-Bar low moves below 2110.50 before a BUY-Bar moves above 2112.00 then I should expect a failure to hold. If the reverse happens, then I should expect a hold. I have a trading plan for each of these conditions. The 2110.25 and 2112.00 price points are not arbitrary, but are based on what I consider information points, that various market participants monitor. This has been determined based on historical data. Also, you will notice the use of the words BUY-Bar and SELL-Bar, these are my way of classifying data -- I don't trade based on time-bars, volume-bars, etc. For my trading style, I need raw data. Further, I also have criteria to determine when the "test" commences. At exactly 13:40:12 CST, the test for the 2110.50 level commenced. At this time, I had no idea if the level will hold or fail to hold. At 13:42:37 (seq-a) indicated above occurred, and I went short at 2111.50 as per trading plan. The penetration of the level occurred at 13:50:35 CST. I then covered the trade as per trading plan. This is just one example. But, the information that the level was most like to not hold or hold is available if one knows how to find it. This "how" is not straight forward, and changes with context. Now, you could always argue that this trade example is just one of many, and that there certainly must be times when (seq-a) occurs and the level holds. This argument cannot be disproven as we are talking about inductive logic here, and all it takes is one counter example. This is exactly why we use stops when we trade. However, in more than hundred (seq-a) I have analyzed, not one has failed -- the only reason I can think of is this: the positions, once committed by large number of traders, are not easy to overcome. I don't relate hard work to "struggle and fight". There is absolute no "struggle" or "fight" in me. I am at peace with my beliefs and the work I have to do. "Inoperative", and "inappropriate" are words used based on one's belief system. Since we agree that market belief vary amongst market participants, we can only agree that "one man's meat is another man's poison". The true nature of the market is one's belief about the market. What is important is not understanding the true nature of the market or the game one is playing but the consistency of one's belief about the market with their expectations and actions. As long as there is this internal coherency, there is no other understanding (about the markets) that is required. This above statement is inconsistent with the idea of Complex Systems (once again goes back to our core beliefs!). According to Hayek, patterns in a complex system repeat themselves (so we know something about the pattern), but the nature of such pattern cannot be predicted (so, we don't know how a pattern instance will look like -- so it is unique), and the timing of such pattern occurrence cannot be predicted (so, we will not know when such patterns will occur). Thanks for providing an opportunity to clarify my thoughts. Not a problem at all. Please do call me that if you so prefer. All the best. Regards, Monoid.
IlIHeroic: Exactly how I trade. Complexity of the pattern depends on the pattern itself. Some are simple, some are complex. However, a complex pattern does not mean it is complicated. For example, the way I read the markets, patterns around cash open in ES are usually complex, but trading it is not complicated. All the best. Regards, Monoid.
Greetings Again Mo, I don't want to be more guilty of hijacking this thread from JS, so I will keep my response brief. This discussion was very thought provoking, and it was a pleasure to exchange ideas and different ways of thinking with you. As you well know, there are many paths one can take to the mountain top in this business. And when speaking with one who has trod a different path, and arrived at the same destination, the pleasure is increased that much more. Your insightful commentary and different point of view was much appreciated. Now to the discussion at hand, you said: “I strictly define Technical Analysis (TA) as analysis of geometric patterns based on chart formations. I define Price Action (PA) as analysis of price movements, and is not related to geometric patterns.”…… Please let me further clarify my statement about TA, and what I was attempting to convey. When I speak of TA and the Discretionary Retail Trader, I think in terms of the entire realm of “Technical Analysis of any kind”, to include charts, patterns, numbers,…etc. So in my book, it doesn’t matter so much what one uses, but what does matter is how it is used, and why it is used. If it is being used to; identify the patterns, define risk, and determine when to take profits, then there is no argument. However, if one believes that his/her TA has the inherent ability to perform "market divining" tasks outside of these boundaries, I believe that they are deluding themselves, and also setting themselves up for endless cycles of pain and frustration. It is my firm belief and conviction that for the Discretionary Retail Trader, any form of TA standalone, without also possessing the correct “thinking strategy” about how it is to be used,and how one is to “think” while using it, is not the road to consistent success in the market. But obviously this is not to say that TA isn’t important and absolutely necessary, because is absolutely necessary and required. What I am however stating is that, bereft of the proper understanding of the nature of the game which one is playing, which in my book is a “Pattern Probability Game”, and the “Mindset” necessary to properly execute this type of game, the “TA Tools and Techniques” standalone, could not possibly bring about the consistency one seeks. “Comment: If I am able to correctly detect the sequence, I can assure you that the result of the trade is not random. The result can be forecast-ed to a great deal of accuracy.” If you are speaking about the probabilistic results of a particular pattern, as aggregated from a statistically significant series of pattern occurrences, then I am in complete agreement with your above statement. Thanks again you for this opportunity to see an interesting and different point of view. KDASFTG
Couldn't agree with you more. You said it better and succinctly than I could ever do. The pleasure has been mine too. Thank you so much for engaging in this discussion. JS, I will follow KDASFTG's lead and stop prolonging this discussion in your thread. Regards, Monoid.
Mo, KD and Hero, I've really enjoyed reading your conversation. It's packed full of information that I'm only beginning to grasp. As a newcomer, my contributions might not add anything. i worked 10 hours today, so I will respond in the morning in more detail. Great conversation. -JS
Mo, To begin, I want to thank you for taking your time to share some of your knowledge in great detail. If my response seems vague and that misunderstood, it's likely because I have. Feel free to clarify. On one side of my brain I hear you talk about a belief in complete information about market behavior or movement. If you're referring to that moment where the collective trading behaviors tip their cards and reveal their true intentions, then I too have seen this. Whether or not one is in the trade is up to the trader and Trading Plan. If this is a correct interpretation of complete information, then I agree that it exists. The other side of my brain disagrees, because I believe that the outcome of every trade is random, and that every moment in the market is truly unique--nothing like past moments, even though the trade may seem like it is. In order for a trade to be regular and ordinary, not only does the pattern have to happen around the same price points, not only do all of those collective traders need to be present, but each trader has to feel and behave in the same manner they did on the last occurrence. Do you believe this? If you do, how do believe this, and also that complete information is available to you? I'm asking out of curiosity. As you can see all this takes a lot of work and effort. I find it interesting when people say trading is easy or simple — it has never been to me. Trading takes a lot of ongoing pre-market analysis to properly understand context, a lot of focus to properly extract information from the market to trade the signals in one's context, and on top of it all, the need to constantly deal with one's urge to make love to that beautiful lady who goes by the name emotion. None of this easy or simple, but does become a routine. It sounds like a lot of work and effort. I'm not against work and effort being put into the Technical Analysis. Emotion is one of a trader's largest hurdles. I believe Emotion to be one the largest common factors in trader failure. Douglas states that a trader with a trader's mindset could execute a "half-way decent" system and turn a profit, compared to a top-notch TA wizard who didn't have the trader's mindset. Do you agree? ================================== What I find interesting, is that not only can I see what you mean when you state "I then removed the randomness assumption and started looking for information within the context which, when properly extracted, should help me align with the majority of market participants" but I can also see how this would help me in my future TA work. Like many have already stated, I look for A, then B, then C, but if D happens around here I need to E. I will more than likely consider this in future work. I've already begun to incorporate it on my own before having read this. The if this/then that routine. EDIT - I want to add I'm not sure I could remove the "randomness assumption" though. ========================= Question --> What/when/why/how, did you begin to believe in "complete information" vs. "partial", and "complex system" vs. "random"? Thanks for sharing, -JS
@JS After I read @KDASFTG's post I realized two things: (1) Not knowing the reader before posting my thoughts can do no good; and, (2) In making an implicit assumption that I thought was everyone's base line, I made a grave mistake. "Knowing the reader" is important 'cos the poster can then communicate to the reader based on a commonly understood framework (terminology, definitions, meanings, etc) without the need to be repetitive, and without the risk of being misunderstood with regard to the common framework. Clearly, here in ET, this common framework is definitely lacking 'cos of members's varied background. I will get to Point (2) above in a bit. Because of the the two points, I realize that I have utterly confused the readers -- in fact, as I understand it now (talk about being a dull-wit!), we are talking about the same things but the terminology is coming in our way. Having opened this can of worms, it is only fair that I try to close it to the best of my abilities. My hope is that by the time I am done, you, the reader, will see how similar we are and what you think might be incompatible differences are really not. Now to Point (2) above: The implicit assumption I made in my original post is this: Any trading plan/system one develops must acknowledge the probabilistic nature of the market. The above statement is agnostic to the type of trader one is -- TA (KD's definition), fundamentalist, etc. Specifically for people who trade using TA, we follow another rule: (a) Our trading plan should be firmly rooted in principles of TA -- which will help us define risk, entries, and exits. Further, as manual traders, we expand our rules to include: (b) We need to have/develop a "trader's mindset" to execute our trading plans. So far, I can only assume that we all are in agreement. Since my original post in this thread was regarding (a), I did not talk about (b) or the implicit assumption discussed above. Moreover, since that post was targeted for providing ideas for research on TA, I just wanted to provide an view of how one's belief system could affect one's research direction. So, there was no need to talk about anything else (including the common framework I assumed away!). On the common framework: Probabilistic thinking (a word used in the implicit assumption) does not mean one has to assume that the market is random. In fact, if one assumes that the market is random, one should not trade, period. Randomness is a special case of a probabilistic system, where there can be no positive or negative skew to trade winning percentage. The long-term outcome of any given trade has to be 50%. Some people argue that even if that is the case, we could have positive expectancy, and hence we can trade. This is an error of reasoning. Although we could know the exact risk on a trade, we do not know what the future profit will be. So, talking about ex-ante expectancy in a random system (where trade win-rate is 50%) does not make sense. I can only assume that when people here talk about "random", they don't mean what I said above. So, what people really mean by the word "random" is really "non-random but probabilistic". Since, my training includes work on Probability theory, when I hear the word "random" I attach its technical meaning to it. My problem. So to help every one out, lets forget using the word "random", and instead use the word "probabilistic" to mean "non-random but probabilistic". When I do use the word "random" you can assume that I use the word in its technical sense (outlined in the previous paragraph). So, why is a system random? One way to answer the question is to think in terms of Information (Information theory is yet another area I have done some work on, hence this bend). If one has complete information, then the system will be deterministic (baring any positive feedback). In a random system, since the outcome can go either way in the long run, one can only conclude that there is no information available in the system. When the late Fischer Black talked about "noise traders" this is what he meant: traders trading based on no information. So, are there systems that can have partial information availability? This is a very tricky question to answer -- a question that Complex System Theory is trying to tackle along with other issues. One argument put-forth in this area is that there are time when complete information is available, but because of positive feedback (self-organization as an example), the information quickly becomes obsolete. In other words, the system uses that information and generates more information. This new information could be complete or partial -- even bordering noise (no information). The information cascade continues. As you can see, a complex system is necessarily a non-random probabilistic system. Lets put this background information into context as I answer your questions. Complete information means one can make a decision without a need to make assumptions. In a random system, a decision cannot be made without making assumptions 'cos there is no information to start with to make that decision. The ability to make a decision based on complete information does not mean one can predict the outcome of that decision. The predictability of results depends on the system. In a complex system, the result of one's decision is a function of (i.e., depends on) the decisions made by others in the system. In other words, the result of one's decision is probabilistic, not deterministic, and certainly not random. This is precisely how a pattern generating, self-organizing complex system works. So, yes, you could say what you said in the first sentence in the above quote. This is the core of the problem we are having. The context here is missing! . What type of the system are we talking about? In a random system, the above statement is superfluous 'cos the system is itself random. In a system that is not random but probabilistic, the above statement is superfluous 'cos it is a tautology. Our interest lies in a series of trades and not in individual trade. So, why even bother taking about it? May be, I am being too technical here! You are taking about any probabilistic system here. I don't know what you mean by "trade to be regular and ordinary". So, I will skip that part. It seems to me that the meaning of the word pattern I talk about is different from the meaning you attach to the word pattern. I use the word as it is defined in complex systems. Patterns are not static; each instance of a pattern is unique. But the information contained in them is complete and formless. In other words, the information that is contained in a class of pattern is the same; but, no two instances of patterns belonging to the same class are same. Addressed in rule (b) stated earlier. Also, if you know that you have a "half-way decent" system why settle for it? Why not make it better? Shouldn't we as business owners look for continual product and process improvements? If I have failed in my effort to quell the perceived differences where non really exists, then I can only attribute it to my inability to communicate. All the best. Regards, Monoid.
You didn't fail, but I did have to google some definitions of your chosen vocabulary. For me, this assumption is assumed, after I believe that every trade is random, no matter the context. After I believe that every trade is random, then I can believe that an edge/pattern/signal is simply a higher indication of one thing happening over another. Those were just words that are the opposite/antonym of unique. So we agree? -JS
Dear @JS It does not matter if we agree or disagree. What matters most is that you have a plan to trade the market, and you work on your mindset. You seem to know the importance of being non-deterministic. So, I assume that it will flow thru' into your trading plan. From where you sit, everything else is just bickering about details! All the best. Regards, Monoid.
RECAP 26FEB2015: Summary: I spent the past couple weeks attempting to "beef-up" my understanding of Technical Analysis, and to also incorporate a new pattern. Today's events were quite shocking. It was full of "head-on collisions" and doubt. Days like this make me wonder why I chose to incorporate a new pattern, since the past one was performing decently. It wasn't performing to the extent that I want it to perform, but it was profitable. I'm also trying to consider that I've been away from trading for a while, and maybe I'm having to remember some very important lessons that I've learned in the past in regards to a proper trader's mindset. In trading today, I also realized that I don't have a specific plan for playing this pattern--I've heard it before and I'll probably hear it again...why do traders make it so damn difficult? Tell me if I'm wrong. But if your signal/pattern is setting up, this is obvious. Strict rules determine this. Context helps to filter out the ones in the best location. If the pattern set's up, one looks for ISL, Target, then Entry. Upon entry trade can be managed/scratched if need be. It sounds very simple, yet I'm making it difficult. 9 Trades. 3 W & 6L = -$94