How long would a gambling casino be able to stay in business if most of the customers won instead of lost? The markets are no different. In order to continue to exist, the markets must operate in a way that causes most participants to lose. There would not be enough money available to pay the winners if the majority were consistently taking profits out of the markets.
Gambling casinos have an advantage over the markets in that they are able to set the rules of the game to insure that the house has a mathematical edge. The markets cannot directly control how the individual participants will play. Most commodity traders are intelligent, competitive individuals. There are seemingly unlimited sources of information about how to trade. There are powerful computers within everyone's reach to help conquer the markets. Why is it then that such a high percentage of traders still end up losing?
A well-known answer is that traders cannot overcome their emotions well enough to succeed. That is certainly true. Another less well-understood reason is that the markets constantly send out "disinformation." I could not find disinformation in my dictionary, but it is a term coined in the intelligence community and now in broad use. In intelligence parlance it means false information designed to mislead and confuse the adversary.
I am not suggesting that the markets have any volition or that there is a conspiracy among insiders (like the "evil" floor traders) to fool the rest of us. This is something that just happens because of the nature of markets.
William Eckhardt is one of "wizards" Jack Schwager interviewed for his book, The New Market Wizards. Eckhardt was the partner of mega-millionaire Richard Dennis. It was his bet with Dennis about whether trading skill could be taught that led to the formation of the famous "turtles." Eckhardt put it this way. "The market behaves much like an opponent who is trying to teach you to trade poorly."
A common formulation of this phenomenon is the concept of random reinforcement. Traders are not rewarded with a profitable trade every time they do something right, nor are they penalized with a loss every time they do something wrong. This makes it exceptionally difficult to figure out what is right and what is wrong. Compare this to an electric fence. Every time you walk by and don't touch it, you feel fine. Every time you touch it, you receive a painful shock. It doesn't take a man or animal long to learn how to relate to an electric fence.
Think how much easier learning to trade would be if you automatically took a loss every time you failed to follow correct decision-making procedures. At the same time, what if you were always rewarded with a profit when you traded correctly? You would be able to learn the correct trading rules much more easily.
As your opponent who is trying to trick you into trading incorrectly, the market is constantly sending you
disinformation. One important piece of disinformation it sends is that the market is constantly changing its behavior so the successful trader must be vigilant to change his approach to keep pace.
Have you noticed what a constant refrain this is from various trading experts? It is a common piece of conventional wisdom that no mechanical approach can be successful very long because the markets change. You are advised, therefore, to change your system to keep in tune with recent market behavior.
It is in the expert's self-interest to preach this gospel. Anyone who tells you that the markets are always changing no doubt has found a "solution" to how to keep his trading method up-to-date. He probably wants to sell it to you in one form or another. If he is not selling his system, he at least can appear incredibly wise and resourceful to his audience. It is a sure thing his audience hasn't found such an elegant solution to beating the markets or they would be rich and would not have to listen to any experts.
Another reason experts invariably claim the markets are forever changing is that it is a convenient excuse for poor performance. Every successful trader has numerous periods when his system or method doesn't seem to work. It is more palatable to say the markets have changed than my system isn't working right now. If your system isn't working, it implies you have failed. On the other hand, if the markets have changed, that is beyond your control. You can just "fix" your system.
This is the origin of the pernicious practice of re-optimization. Let's say you have a great trading system using three moving averages. During the last five years, it has made beaucoup bucks in hypothetical historical testing. Naturally, you use a different set of moving averages for each market because each market "has its own personality." You are in tune with the markets and ready to trade.
But wait! You know that the markets are constantly changing their personalities to keep traders off balance. It is only reasonable to expect that the correct moving average values will change over time in the future. So as part of your approach, you set a schedule for yourself to re-optimize the moving average values every so often.
How often you decide to do this will depend on how industrious you are and how often you think the markets change their spots. Do the markets remain stable for six months, three months, one month, two weeks, two days? Whatever your answer, you will dutifully re-optimize your moving average values to keep up with those elusive markets.
Let me tell you a little secret. Your system will always trade the past perfectly, but it will probably be lousy in real trading no matter how often you re-optimize it. Oh yes, you will have some extended periods of good profits. This will encourage you. The market is just sending you some of that good disinformation, encouraging you keep it up. In the long-run, absent some extraordinary luck, you will surely lose.
I am not saying the market don't change. They clearly do. One of my favorite sayings about the markets is, "The future will be just like the past, only different." What I am saying is that any attempt to modify your approach to "keep up" with this change is like a dog chasing his tail.
When you re-optimize your system every two weeks, you are making the assumption that during the next two weeks the markets will behave like the last two weeks. This is a comforting idea, but where is the evidence to support it? Research has shown just the opposite. The best system parameters for the next period are most likely not the best parameters from the previous period, regardless of the period length you choose. You are as likely to be successful choosing your parameters at random from previously profitable parameters as you are choosing the most profitable parameters.
When people ask me how a system has performed in the last twelve months to get an indication of how it will perform in the next twelve months, I ask them a question. What makes you think 1997 markets will be more like 1996 markets than like 1987 markets or 1993 markets? In fact, 1997 markets are more likely to be similar to some previous year other than 1996.
The purpose of this market disinformation is to cause people to be suspicious of successful trading methods and abandon them too soon. One of the most reliable traits of professional traders is the ability to stick with their system much longer than the typical loser. We all go through losing periods, no matter what type of approach we choose. We cannot increase our chances of success by constantly changing our approach. Since there are many more losing approaches than winning ones, we actually decrease our chances of success by frequently changing our system.
The correct solution is to find a non-optimized approach that works over a long period of history in a wide variety of markets. To avoid over-curve-fitting, use the same rules for all markets. If you can trade it for an extended period in the future in a wide variety of markets, you are likely to be successful, although success is never guaranteed.
The reason this works is that although the markets change their short-term patterns, they are still trending overall. That is your edge. If you play the trends, you will succeed in the long run. Attempting constantly to modify your system to mimic the changing patterns of the recent past will not improve your chances of success. It will more likely insure failure.
Human nature is such that we are always trying to improve. I am not suggesting that you might not be able to create a better system in the future. Just don't fool yourself into thinking that it is better because it is somehow adapting to ever-changing markets. It is better because it is more profitable over a long period of time or because it trades more markets profitably.
[ Table of Contents ] [ Previous Section ] [ Next Section ]