In this article I would like to share the experience of creating automated trading systems for Forex and talk about the problems faced by those who want to engage in this business. I hope I will save you a few hundred, maybe thousands, of dollars.
Prehistory
Forex for me is a rogue's fight against rogues. It began in 2003 at the courses of one of the leading companies. The minimum deposit of $ 2,000 for a 19-year-old student was too large, so in order to get access to the micro-lots and learn to press buttons for small money, you had to buy a training course for $ 300 several times, which included a training micro-account with two hundred dollars on account.
At the training, the lecturers sweetly told about the possible millions, and the video with the housewife, who with a towel on her head earns $ 400 for a couple of hours, tickled the toad at heart. It seemed that here it is - the solution to all problems. But in practice, as well as the majority, everything went differently. The first couple of micro accounts went to zero within a month and a half.
In the office of the company was a wardrobe with books. Over the next couple of months, Murphy, Demark and other classics of stock market analysis were read to cover. Everything read was structured and immediately checked at the monitor. The third score didn’t grow much, but it was already above zero. In parallel, the office director six months after my start, despite the fact that I did not have an outstanding result, offered me a place as a teacher. This, most likely, and gave the main impetus to development. When you begin to teach others, finally, you begin to understand yourself.
')
The team of teachers was friendly, we often lingered, sorting out statistics and strategies. The manual miscalculation took dozens of hours, and the strategies themselves were based on simple indicators and their combinations. Then there was a need to process a large amount of data by the technology. The first language was very limited in functions, but it was enough to reduce the analysis time by a hundred times.
Soon the first tester was written, which allowed one to learn how to trade history without losing money in the market. The new director of the company did not like this idea, because the flow of deposits due to this approach began to decline. I had to change jobs. But after working in 3 more companies, the vision of the business fell into place.
“Baby, they are swindlers!”
Working as a teacher and communicating with the back-office made it possible to understand how the company takes money from a trader. In fact, in most cases, do not need to do anything. The client by his actions will merge his deposit. You can draw an analogy with a man caught in a swamp. The more it starts to move - the faster it sucks. There is simple math: the client is chasing in a deal, on average, for 20 points. Spread - from 2 points. It turns out that starting from the moment of entry, in order to get a loss, the price is enough to pass 18 points. And in order to make a profit, the price needs 22. Negative expectation is 10% (we assume a normal distribution of prices and a 50/50 chance of guessing).
Rule # 1: the greater the result in the transaction, followed by the trader, the better the expectation. And vice versa.
For advocates of the statement “Forex is not a casino,” I will immediately give a gag: the basic analysis tools available to most people who go to the market do not shift the expectation by more than 0.5-2%. Therefore, if measured on a larger scale, a roulette with a negative expectation of 2.7% will feed those who want more than forex.
It happens that a random trader falls into a pleasant dispersion and his deposit grows many times in a matter of days. Anyone who has ever seen the managerial part of terminals knows how companies “treat” such situations. In 2003-2007, dealers worked on the correction. Since 2008, automatic systems have spread throughout them — auto dealers, in which all the logic and actions for returning deviations to the norm are spelled out. Most often this is done with the help of two primitive actions: delayed execution and expansion of the spread.
And how, nevertheless, make money on Forex?
If you are not a programmer and are not going to become one, the only way to make money is to expand your trading goals. On large timeframes, market deviations will be used and there will be more chances to beat the company. But at the same time, the high yield will be lost and Forex for you will turn into an ordinary instrument with 20-30% non-guaranteed annual yield. How big goals should be set? - one third of the annual currency volatility. For example, for EURUSD, this value will be about 1000 points. Even if the DC will give you an extension of the spread to 5 puktov with basic 1.5-2-x, the negative expectation that you will need to overcome will be only 1%, which is more than double, better than in a casino. Sports interest will be less, but will stay with the money.
Let's turn to sweets: automated trading. Most companies support Metatrader, which, in turn, makes it possible to write code in the built-in MQL4 language or to build a bridge with the usual software. I’m grieving readers a little: then we’ll consider the methods we used to earn in our team earlier, but now they are only of historical value and can be applied only with additional conditions or in certain situations. There are no eternal engines (or as stockbrokers call them grails). Techniques for extracting quick money from Forex are constantly changing.
Method # 1: Volatility Hunt
The method worked very well in times of fixed spreads. What is remarkable for the market is strong news - they cannot be analyzed and clearly stated where the market will go after them, but one thing is sure: there will be movement. We specialized in labor market data. They are quite strong, from 20 to 80% of daily volatility, fluctuating currency rate. Automation in a cycle 30 seconds before the news release scattered grid orders in both directions at 30-50 positions every 1-2 points. At each position was put the minimum allowable stop loss. In those days, less than 5 points were not. Take profit was 2-3 times more. In practice, they mostly caught the same amount of profitable and unprofitable positions. The feast was arranged when “shots” occurred. The companies guaranteed the execution of all pending orders, so even if there was a gap after the news (a significant difference from the previous quote and the next one), the company did not recalculate the result.
Method # 2: Spike Hunting
This feature of the quotation mechanism of several companies was discovered when OdlSecurities (a British company) exposed us to a loss of a couple thousand dollars for non-market quotes. Emissions prices at the time of fixing the loss were only for this company. Despite the arguments and screenshots from ten other companies that we could provide from accounts scattered around the world, management decided not to pay us damages. We are greedy. We noticed that “feeding” stop loss on one account - a shift in quotations, in order to lick it off at night, occurred on all of our accounts in the company (had 4 accounts). They wrote a simple automatics, which did the following: on one account, it started a stop at a distance of 70 points from the market, and on the other three, when there was a deviation, it poured a full deposit on the return movement. For a couple of nights, the deposit was restored and there was a handful on top. We must pay tribute, we were returned both our money and what we earned from above, but were forever forbidden to open accounts with this company with our names.
Method # 3: Using Market Features
We broke up the deposit from 10 to 30 thousand at the official championship in automatics in 2007. The logic embedded in the robot acted exactly six months, when there was a very high correlation between the pound and the euro. The volatility of the cross-rate of these pairs (EURGBP) was very low. We took the most profitable part of the day, when the deviations were even smaller - the night. As soon as a small spike occurred, the robot instantly reacted and turned on the trade against this movement.
Here we got another rake. Despite the fact that the earnings were based on a temporary market feature, and we did not use the “holes” of a particular broker, several companies refused to return the earned money to us, arguing that the average life of the transaction was less than 10 minutes. At that time, there was no “pipsing” item in the contracts, but there was always something descriptive “at the discretion of the security service” or something like that.
How did it all end?
Forex business, I am grateful for the opportunities and the school, which he forced to pass. I would decide to get involved in it, if I knew what had to go and what situations to go out of? - I do not know. Here is its own irony and its own chance. Now we are not actively trading on Forex, but we go there with the whole team when a group expectation appears. As a rule, this happens as a result of the actions of slow marketers, less often - of market circumstances. We spend most of our time on the American market, there are our own rules of the game, but this is a topic for another article.
Short dictionary:
Spike is a sharp price spike, forex trading is usually not related to price changes in the market. With the help of spikes, brokers can get stop-orders of customers who are close to the current price.
Pipsing is one of the types of trading in which the goal of earning in a transaction is 1-2% of the daily volatility of the instrument. As a rule, the life of such transactions is very short (less than 10 minutes). For forex brokers with weak technical capabilities, it is dangerous, as it can be based on delayed quotations.