How to trade and not lose on Forex. Psychology of the Forex market
Having the pleasure to observe and communicate with customers of two brokerage companies for two years, I can state that the behavior of people who want to “play and win on the stock exchange” is typical and well-forecasted, unlike the shares traded by them. Expected return on speculation at the start is usually “not less than 1000% per annum.” After several transactions it is reduced to “at least 100% per annum”. After some time, it falls to “at least return the initial capital”, after which the speculator for an indefinite period of time, until reaching a planned yield of 0% per annum, becomes an investor. In the methods of making trade decisions, a certain evolution is also traced.
At the first stage, a new client reads all the economic publications, listens to the news with such attention, as if he expects to learn them first. Important is the behavior of the Brazilian stock index Boves. Exceptional importance is given to the forecasts of all possible analysts and opinions of colleagues in the dealing room. A typical question that is asked at this stage: “What do you think about the market?” Or “will such an action grow?” (If he already bought it). My typical answer is: “I do not know” or “throw a coin”. The client will not rest until he finds someone who confirms that “such and such an action will grow”. At this stage, the client is looking for someone who would tell him specifically when and what to buy or sell on the market.
At the second stage, the interests of the client, who has already studied Elder and Murphy, are shifted towards descriptive geometry in the framework of the 4th grade of secondary school. The typical question that is asked at this stage is: “Have I correctly drawn a trend line? If so, then you have to buy (or sell), and if so, then do not. ” My standard answer remains the same: “throw a coin.”
At the same stage, as a rule, the Metastock program known in the industry of attracting new money to the stock exchange is being mastered, with the help of which numerous indicators are being minted at a number of prices: differential differentiation and integration. Consequently, there are only two indicators, the rest are their infinite variations. The newest of these indicators declassified in 1978, probably because of their uselessness. A typical question, which is often asked: “What the indicator says.” My sample answer remains the same. At this stage, the client needs to find something that would tell him how to buy or sell.
Convinced that the methods used in the first two stages do not work, speculators who were fortunate enough not to become investors, begin to understand that they need to find some systematic strategy for playing the stock market. At this stage, customers are looking for a “system” that would give out signals. The main problem that arises in this case is that the client, in addition to much labor, must overcome his own prejudices.
I will cite a typical example. One of the long-standing customers, long and hard studying the price charts with indicators, built a system that, according to him, gives out 90% of winning signals, buying on lows and selling at highs. Signals are rare, several times a year. The system consists of a large number of all the same indicators, being built on different time scales (from half an hour to a week). With fairly complex rules, a combination of indicators generates signals to open a position. Signals can be strong or weak, i.e. they are continuous, not binary (or there is a signal, or it is not), as I always thought. Signals, apparently, are ambiguous, and are executed taking into account the subjective factor. All signals are based on convincing theories. What to do with an open position and how to calculate the size of a position is not said. The question is whether the system has a positive mathematical expectation, meets silence.
The system has not yet brought money, because, in the testing process, the author “did not trust her”, and quickly closed positions when they became profitable.
We will analyze this system by points.
High percentage of winnings. You are unlikely to believe that you can get a high percentage of winnings with the help of a random number generator (the same coin). The secret is this. We throw a coin. If the eagle falls, we buy shares. As soon as the price grows, immediately close the position with a profit. If the price falls, we become investors and wait until the transaction becomes profitable. On the profitability of the strategy as a whole, of course, there is no question. But we are striving for a high percentage of winnings! One of our main prejudices is the desire that our current transaction should be a winning one. That’s why we are sitting out losses, waiting for prices to unfold. Losses, as a rule, become even greater. For the same reason, we close the profitable deals ahead of time. But being right and earning on the stock exchange are two different things.
A small number of examples and conservatism. If we drop a coin ten times, and nine of them drop an eagle, it will not follow from here that the probability of an eagle falling is 90%. The law of small numbers asserts that there are not enough examples to make such a conclusion. However, we piously believe a few cases when our system gave good signals. If we have found some combination of indicators, or, in the general case, the method, we believed in it, and on several examples convinced ourselves that it works, we will do everything possible to avoid the obvious that it does not work. In this regard, William Eckhardt said very well: “We do not look at the data neutrally – that is, when the human eye scans the graph, it does not give all points equal weight. Instead, he will focus on a few outstanding cases, and we seek to shape our opinions on the basis of these special cases. In human nature, choose the spectacular successes of the method and not notice the continuous losses that pile you to the bone. Thus, even with a very careful study of graphs, the researcher is inclined to think that the system is much better than it really is. ”
Presentation of data. We believe that the bar on the price chart is a market. In fact, it is not more than a price range for a period, as well as an initial and final price. In addition, we believe that the indicators carry additional information about the market, although this is nothing more than price transformations that carry less information than the prices themselves.
Reliability of data. We believe that data on prices and real prices are one and the same. In fact, extreme transactions often take place with one lot, that is, it was not really possible to make a deal at such prices on a normal volume. Another example: some acquaintances in the summer mistakenly bought shares of Irkutskenergo, when prices were in the area of 2.80, at the price of RAO UES – 3.56. This is a real deal, with a good volume, it will be taken into account in many indicators and trading methods. However, it does not reflect the real market.
Degrees of freedom. This is a characteristic that describes the number of rules and parameters of the system. We aspire to have as many degrees of freedom as possible, because we want to have a system that ideally predicts the market. The more rules, indicators and parameters we add to the system, the better the results will be on the historical data. Unfortunately, the less likely it is to show profits in the future.
Simple and complex. We prefer complicated simple. Professor Alex Bavelas conducted an impressive experiment. Two subjects, A and B, isolated from each other, were placed in front of the screens.
They were told that the purpose of the experiment was to learn how to recognize sick and healthy cells, by trial and error. Before each were two buttons: “healthy” and “sick”, and two light bulbs: “right” and “wrong.” Each time the image of the cell was projected on the screen, they guessed, a healthy cell or patient, pressing the corresponding buttons, after which the corresponding light came on. If A guesses correctly, it was lit “correctly”; if A was wrong, it was lit up “wrong”. Soon, A learned to recognize diseased cells in about 80% of cases. In the same he received not the true results of his answers, but based on the answers A: If A was right, then B was lit “correctly”; If A was wrong, then B was lit “wrong”, regardless of the real result. Of course, I did not know this. He was looking for order where he was not. Then A and B were asked, by what rules they distinguish diseased cells from healthy ones. And he proposed simple specific rules. In used the rules are complex and sophisticated. It’s amazing that A did not think that explanations В are absurd or unreasonably complex. He was impressed with the “brilliant” method B and was ashamed of the simplicity of his rules. The more complex the explanations were, the more convincing they were for A. Before the next test with new examples, the subjects were asked whose method is better. Both, especially A, were convinced that method B. In the second series of tests, B showed no improvement. And guessing is much worse!
Lottery. We believe that if we can manipulate numbers, then our chances of winning increase. Hence the popularity of lotteries, although the chances of winning in them are negligible, and all sorts of predictions. In order for us to open a position (bought or sold shares), the market must perform a certain action, according to the rules established by us (what is usually called the “system”). Thus, we have a sense of market control. Because when the position is open, the market lives its own life, no matter what we think about it, we focus on the rules of entering the position. The facts are that this is the least important part of the trading strategy. I experimented with a system with random inputs, and she, in the vast majority of cases (each launch generates a new sequence of random numbers), was profitable. The idea is borrowed from Van Tharp’a, who “the system made money for 80% of the runs, when it traded one contract in each futures market. She made money 100% of the time when a simple capital management system was added – to risk 1% of the capital.
Determinism and randomness. We inadequately perceive chance. Nassim Taleb described the amusing experiment: “The researchers give the birds (some pigeons) in the cage a random way, and whatever the birds do during feeding, they start to do it more to get additional food. This leads to the development in birds of a certain ritual, such as dance. Thus, there is something in the bird’s brains that identifies causality. From the bird’s brains to ours, therefore, a small step. People always think that there is a reason for events. We were unable to accept the accident. “I showed customers the price charts, and they found trends on them, support and resistance,” heads and shoulders, “divergences, in short, the whole set. The most interesting thing was that these graphs were generated by a random number generator, were nothing more than a random walk. On the other hand, we like to catch minimums and maximum prices, believing that they can turn around at any time. It is understood that price changes are distributed randomly, and reversals can be predicted. In fact, there are very large emissions in price changes that can not be predicted from normally distributed random price changes. As a result, we greatly underestimate the risk. If we bought shares, and the price went even lower, we buy more, reducing the average purchase price (averaging). Unofficial statistics read: “averaging in a falling market ruined more Jews than Hitler.”
Risk. We are conservative in relation to profit and are prone to risk in relation to losses. Kahneman and Tversky asked people to choose between an 80% chance of winning $ 4000 and a 20% chance of not winning anything, and a 100% chance of getting $ 3000. 80% of the respondents chose $ 3000 for sure. Then they offered a choice between risk of 80% chance of losing $ 4,000 and 20% chance of not losing anything, and 100% chance of losing $ 3,000. 92% of respondents preferred to take risks. The strategy that the majority chooses (for sure to take $ 3000, and 80% chance of losing $ 4000), has a mathematical expectation of $ 200, i.е. on average we will lose $ 200 each time. As we do on the stock exchange.
Player error (“Gambler’s Fallacy”). We are convinced that if a trend is established in a random sequence (or market), then it can turn around at any time. We are sure that if we have several times in a row losing trades, the next one will necessarily be a winning one. Ralf Vince conducted an experiment with 40 candidates of science, but not professional players, and not statisticians. They were offered to play a simple computer game in which they would win 60% of the time. Each was given $ 1,000 and was asked to put as much as they want, in every attempt. After 100 attempts, only 2 out of 40 (5%) increased their $ 1000.
One familiar trader, who now manages an investment fund in Washington, gave a play to his friend, who works as a leading analyst of Lehman Brothers (or similar, I do not remember), but more complex and close to real trading. He wrote that “she can not go to the next level of the game for the second day”.
Thus, our enemies are within us. Knowing them and learning how to deal with them, we have chances to find a winning strategy, with which we will be dry and comfortable. How to do it – a separate conversation.