Saturday 23 November 2013

Trading Methods – What Works and What Doesn’t

The are many successful traders using a variety of trading methods. However, there are a far greater number of traders who fail to produce consistent profits from the markets. I would estimate the percentage of successful traders about the same as in other professions such as acting, music, sports, etc. Many are called, yet few succeed. Over the years I’ve formed a few opinions on why trading is so difficult for so many people. Most traders I meet are intelligent. They have had success in other careers. They are usually hard workers and devote much time and energy to their trading. Yet most of these traders move from one methodology to another, never finding anything that works for them.
In the area where I live I try to attend as many trading groups and meetings that I can find. Trading can be a lonely business when you trade your own account. It is important to maintain human contact. That is face to face contact and not just communicating on Twitter. One trading group that I’ve belonged to for several years seems to be a laboratory for watching traders who go down dead end roads. I’ve tried to draw some conclusions why these traders consistently go in the wrong direction regarding their methodology.
When a trader first decides to trade for a living, that trader must go through a process of personal discovery to find a methodology that fits their personality. It is quite normal to try many different approaches to find what fits. One could decide that the perfect fit would be very short-term day trading. However, once that activity is put into practice with real money that trader may decide the stress and fatigue of watching the computer screen all day is just not worth the effort, despite how much money that approach might produce. Another trader might decide holding option spreads for a month is the perfect approach, but the lack of activity may cause that trader to be inattentive to his trades and get distracted by other activities and become bored with trading. So finding what fits is best done by trying different approaches. What seems right may not be once that methodology is implemented.
But why are so many traders changing methodologies every few months? Whenever I attend the monthly meeting at my local group, I find that they all have found a new chat room or guru to follow. I hear stories about how that new guru finally has the answer. I hear how that guru called the next days market direction perfectly and has done so every day that month. Really? I heard that the last guru also called the market turns perfectly. If the previous guru was so good, why is everyone now following this new guru, with no longer a mention of the previous guru-of-the-month? There seems to be a need and a real desire to know that there is someone out there with the answers these traders are looking for. They think someone that is successful at trading will be kind enough to give them a profitable methodology, either for free or for a small price. They are led to believe that the guru is actually successful trading his or her own money. But are they? Most of these people are not trading real money while teaching or running a chat room. They will lead you to believe they are successfully trading real money. Many have never been successful. And if these people actually are making the money they claim in their own accounts, why would they be charging a fee to run a chat room and teach their methodology in a trading school with the added work and liability. Wouldn’t their profits from trading far eclipse the relatively small fee they could be receiving from charging for their services? And if their advice or methodology were up to their claims, wouldn’t the word get out and the Internet traffic to those sites become overwhelming? Wouldn’t billion dollar hedge funds want to know those secrets that elude their own research teams? I think these traders looking for an answer are not asking themselves these questions. If they are asking these questions they may be so invested in finding the answer to their trading problems that they don’t want to deal with the hard truth that they may not be on the right path.
Is there a common denominator with many of the approaches being offered that most likely will be a dead end for the trader? Any approach that tries to predict future market direction from non-market generated information is doomed to failure, in my opinion. I’ll explain.
Several meetings ago one trader brought in an approach that tried to extrapolate future market direction from finding similar patterns being developed currently and comparing those to patterns that were developed 70 or so years ago. This trader acted like this was a novel approach. Market letter writers and technical analysts have been doing this type of comparison since the beginning of the markets. It has never worked. You can take the shape of the prices of the current market and try to overlay them on past data and you will find many similar shapes and patterns over the last hundred years or more. But making the assumption that the outcome of that pattern can be determined is just beyond nonsense. What possible relevance would there be in the shape of the price chart currently to that of 70 years ago, or even six months ago. If by chance there were a similar outcome it would be purely random chance. It would be easier and quicker to simply flip a coin.
Another flavor of the month that had everyone excited was using moon and tide cycles to predict where the market should go. I needn’t spend much time discrediting that one. Elliot wave is another approach that in my opinion is a complete waste of time. The theory of Elliot wave is actually correct in explaining and describing the mass psychology of traders. It can explain an impulse move in the direction of the trend, and then explain the logic of the reaction against a trend. On past data most Ellioticians would agree with an analysis, or wave count. However, if you put a hundred Elliot wave “experts” in a room with a chart and asked them where the market is headed, you’d get a hundred different answers and most likely three hundred alternate counts. It is completely useless in trying to forecast the future. The problem as with the moon cycles and overlaying past data, is that the trader is trying to tell the market where it should go rather than listening to the market and hearing where the market wants to go.
Fibonacci analysis is another dead end in my opinion. Fibonacci was quite popular when Elliot Wave was first being reintroduced in the late 1970′s through mid 1980′s. There currently seems to be a renewed interest, along with such techniques as the Gartley Butterfly patterns and a few other rehashes of long forgotten classic techniques. Again, these techniques attempt to tell the market where it should go. A Fibonacci retracement or extension makes the assumption that a market should stop at or go to a certain price because of some natural numerical relationship that defines the spirals of a shell or the relationship of the belly button on the human body. Pure silliness. Sometimes these numbers get hit with precision and turn the market at precise Fibonacci numbers. I’ve dropped a horizontal line on a chart at random and have hit that random number with about the same frequency as that on my Fibonacci retracement tool. Many Fibonacci experts will cluster numerous starting and stopping points on their charts so there will be many lines. Many will also include numbers in between the main Fibonacci numbers. As a result there are so many lines going across the chart that one of them is bound to be hit. The only number that I find useful on a Fibonacci retracement tool found on most charting software packages is the 50% retracement level. And 50% is not really a Fibonacci number. But it is the retracement most often used by most analysts as a guide. It is probably useful because so many watch it and price turns at that point become self-fulfilling.
Also, there is much effort by many of these gurus to draw conclusions from straight lines drawn from distant points on the chart up to the current prices. First, the markets are fractal by nature. Being restricted to drawing a straight line on a chart is like trying to draw a map of a coastline using a straight line. It cannot be done. There is meaning to the up and down movement of the market. If one understands the concept of price rejection and acceptance around previous pivot, or swing point, the trend is more easily understood. Drawing a straight line back in time and assuming price will react somehow once that line is met is not logical. Sometimes a major trendline will act as support or resistance for a time only because so many people are watching it. But market structure based on what the market is actually communicating is far more important. Sometimes straight lines, such as in triangle, will appear to offer valid signals, but most often on closer inspection the actual swing points are a much more reliable guide. Straight lines cannot connect with all the important swing points. Trying to force a fractal data series to a linear series is bound to miss the point.
Classic price patterns, such as the head-and-shoulders pattern, are another area where the trader is trying to tell the market where it should go, regardless of where the market actual does want to go. Like the Elliot Wave, the head-and-shoulders pattern, along with three-drives-to-a-high, can explain investor psychology very well, but again in hindsight. Many major tops and bottoms occur after these patterns are formed. In fact if you look at enough price charts these patterns seem to jump off the page at major turning points. The problem is that there are far more of these patterns formed during trends that do not turn prices back the other way. When analyzing past data on charts, it is amazing how the human eye will gloss over the many more numerous failed patterns and gravitate to the successful patterns. The trader wants to believe. Reality gets glossed over. Again, like the other methods, patterns are based on an assumption that a particular shape of previous data will have an implication for future price direction. It simply isn’t so. Any predicted outcome is pure chance, or at best self-fulfilling. You can’t tell the market where it should go based on random patterns from the past, no matter how well categorized and documented they are.
Another area that can lead to frustration for the beginning trader is the belief that a mechanical trading system can work. To my knowledge there has yet to be a successful mechanical system developed. If a mechanical system actually worked that system would soon own the entire market and would have to self-destruct at some point. Countless hours of programming on main frame computers using advance methods have failed to turn up a system that stood the test of time. The problem is that most of these system use curve fitting to once again extrapolate patterns from past data, whether using price patterns or indicators, and assuming that will somehow tell the future. The curve fitting may work for a short time, but as markets change, as they always do, the systems will no longer be in synch with the markets. All these systems fail. Trying to find a system that will work is futile, and the trader will waste much time in that search. That time would be better spend learning about how the market works and learning to read what the market is trying to communicate.
To summarize the common denominator that I find that will lead traders down a dead-end road are any of the trading approaches that try to tell the market where it should go. Most of these approaches are based on information that is not directly generated by the actual price action. Of course a trend line or Fibonacci level is influenced by price only in that it is drawn on the price from the same data. But it is backward looking. It is making an assumption that something from the past based on an irrelevant numerical relationship or random pattern will cause or influence buying or selling in the future. It just doesn’t happen with enough reliability to enable consistent profits for the trader. In fact, a coin flip has a better chance of predicting the future than any of the methods mentioned above. The very best mechanical systems have about 30% winning trades, which is below the 50% that a coin flip will produce.
If you’ve read this far you might conclude that it is impossible to trade successfully, and that there is no useful approach to trading. It is true that the vast majority of those who try trading will fail. Most stay on the dead end roads of gurus and approaches that don’t make logical sense.
I would suggest concentrating on two things. First, money management is probably the most important element in a successful trading plan. It isn’t as interesting as learning indicators and pattern analysis. But with proper money management, one could take a far from perfect trading approach, even the coin flip, and have a fair shot at making a profit over time. Even successful gamblers with terrible odds can win if they employ strict money management. It should be first on the list of techniques to master if one wants to have a long career, but this is usually the element of the trading plan that is neglected.
The next thing I would suggest is to learn the principals of market generated information. The Market Profile is a logical place to begin. By learning the Market Profile one will learn the auction process relating to all traded markets in all time frames. It is a study in learning the language of the market. Most traders are too busy trying to interpret Elliot Waves and Fibonacci retracement, and as a result they don’t listen to what the market is trying to communicate. The market is concerned about the here and now as it tries to interpret the future. It doesn’t care about some straight line drawn through price points six months ago. Often the graphic that represents the Market Profile confuses and turns traders away. The graphic is not as important as learning the concept. One can still use bar charts and moving averages and other indicators as a guide. But knowing what the market is trying to accomplish by moving up and down in what appears to be a random fashion can make the difference in how a trader views a chart. It is well worth the time spent learning what this technique is all about, regardless of the type of chart a trader uses. In fact the concept behind the Market Profile was developed to simulate the mental process of a trader in the pit. The buying and selling and seemingly random price moves do have meaning, but most traders are not paying attention.
To trade successfully is very difficult. The learning curve can be very long. The learning curve for the Market Profile approach can be quite long and difficult. Few things in life that are worthwhile come easily. I gave a three-hour lecture on Market Profile a few months ago to the local trading group. To my knowledge not one person in attendance is applying any of the principles discussed. A few people came up to me at subsequent meetings and said the lecture was interesting but it was just too much work to learn that approach. They would prefer to stick to what the group is doing by jumping from one guru to another in search of an easy method that will give precise and winning trades. Good luck to them. Unfortunately they will supply the profits to those who understand the markets.
Doug Tucker
 via - Tucker Report