Hello, Alexey. We would like to congratulate you on such a good performance of your Expert Advisor. Please tell us a little about yourself: what is your occupation, how long have you been in trading, how have you got acquainted with Forex?

Thank you for congratulations. I have a technical education. Presently I work as a programmer and trade in Forex. First I heard about Forex about 6 years ago. First my attitude was a little skeptical, but then I started to try myself in this new occupation. From my childhood I am interested in programming, so naturally I started trying to automate the trading process using Expert Advisors (at that time in MQLII). For the last 6 years I created a large number of indicators and Expert Advisors, including those made to order.

And can you tell us at least approximately how many EAs you have developed?

I am afraid, I cannot tell even approximately. I think about 200.

Have you ever tried to develop multi-currency Expert Advisors? If yes, what are your impressions of such EAs?

I have a positive attitude towards multi-currency systems in general, and diversification of risks in particular. But I do not have any separate multicurrency Expert Advisor. In my opinion, it is not so easy to manage such an EA. Besides, there are some difficulties in testing such EAs on history. That is why I prefer using several Expert Advisors trading different currency pairs, which gives the effect of the multi-currency profile trading.

Have you traded on live accounts using an Expert Advisor?

Totally approximately 6 Expert Advisors trade on a live account. I say approximately, because sometimes the number is larger and sometimes lower, depending on market conditions. As for the Expert Advisor presented in the Championship, this is its first true demo trading, because I am not a supporter of long-term demo testing. A tester and at least a micro-real are enough to detect all the details of the EA operation. But this does not concern all Expert Advisors, but only "the selected few". Some EAs do not go further than a strategy tester.

And now let us talk about your Expert Advisor competing in the Championship. The deposit growth started only at the end of the first month of the contest. Are there certain training mechanisms implemented in your Expert Advisor?

True, there are certain mechanisms. But they are quite simple, i.e. they are not connected with neural networks that are so popular today - this is just a simple adjustment according to entries in some time interval. Initially I didn't plan my Expert Advisor for the Championship. And 2 weeks before the start I decided to try myself and see how the EA would behave in new unstudied conditions. I changed one of my Expert Advisors specially for the Championship, mostly simplifying its code (because of the Championship restrictions) and using the simplest variant of Money Management.

Why have you chosen exactly EURGBP?

More correctly is the question why I have chosen exactly this Expert Advisor trading EURGBP, because generally I trade not only this currency pair. The purpose of the Championship is to get the maximal profit, that is why I selected the "scalping" strategy. Well, the attitude towards scalpers is contradictory. They are often blamed, treated as prohibited and it is considered that brokers treat such strategies negatively. Partially this is true if speaking about "error-hunters" or "spammers" (as I call them). Attempts to catch a necessary quote, "bombarding" of a broker by constant requests, numerous order modifications, as well as multiple placing of pending orders (especially without their triggering) - all this overloads a trade server and various measures are taken against such "work".

But there is another type of scalping, I call it a "logical" scalping. It is based on the observation of a certain pair, analysis of its movement and further attempts to find the resemblance of such a behavior in future trading. Its entries are governed by logics. Even an indicator can be created reflecting the entry points, because they are calculated, non-random as contrary to those trying to catch a "splash". That is why generally brokers do not have complains against such "logical" scalpers - the number of trades is not so large, there are no pending orders, no modification of orders. The load on a trade server is minimal. Correspondingly it is harder to influence the operation of such an Expert Advisor.

Judging by trades your Expert Advisor trades using some oscillator and opens trades inward a channel, doesn't it?

The EA's operation is based on the attempts to find peaks and hollows and proceeding from these data to choose a trading direction. True, trading is conducted inward a certain channel, but with such small takes it is unreal to use for this purpose an oscillator because of the lagging of indicators. For entries first of all the price itself is used (as a kind of an indicator) and, secondly, additional filters for "sifting away" doubtful signals (part of them is not used in the Championship version of the EA). A distinctive feature of this approach is the noncriticality to quotes (and therefore to choosing a broker), as well as the spread size is not strictly important - with larger spreads profit can also become larger, because there is a potential of increasing TakeProfit levels.

We have mentioned it in the report about the 10th week that your EA does not trade in day hours. Why is it so?

My Expert Advisor is flat-following. It is alive when the market is asleep. That is why it takes a pause when there are strong movements in the market, waits till the trend is over and then continues.

The graph of trade opening frequency shows that the peak of order opening is 3 a.m. and 8 p.m. server time. But the largest number of trades is at 4 p.m., this is a single peak. What is it connected with? Are there any advantages of this time?

Well, I haven't conducted any investigations concerning advantages of certain time periods. It seems that an interesting regularity has been detected that should be further studied.

Are there any filters in your EA for sifting away trading signals?

Of course there are, and they are quite numerous, as I wrote earlier. But in the Championship many of these filters are switched off. Of course this increases the risk, but the number of trades is also increased with the purpose of getting the maximal profit. And it was interesting to check the behavior of my EA in the Championship with such settings that are too risky in live trading.

Judging by the diagram of opening of profitable and losing trades, positions opened at 0:00, 7:00 a.m. and 5:00 p.m. were 100% profitable. What do you think about it? (Profitable trades are shown in green color, losing trades - in red.)

This is interesting, but I think it is only a coincidence. But if reasoning logically, the EA's task being to find a flat period for entering a trade, this time appears to be meeting this criterion. But I will repeat once again - the EA is not designed to enter a trade at a specific time - 7 a.m. or 5 p.m. All hours, except a trend period, are equal for the EA and it searches for optimal entry points within flat periods.

Have you checked the strategy implemented in the contesting EA on other currency pairs?

Yes, it shows quite good results with EURUSD and EURCHF, but the maximal profit is obtained on EURGBP.

The balance of your Championship account was lower than $5 000 on the first days of the Championship. Are you ready to use this Expert Advisor on a live account?

Actually it trades on a live account. The problem with losses is probably connected with the fact that for the Championship I tried to use the algorithm of parameters adjustment, which is not fully tested yet, so it could fail at the very beginning. On live accounts such drawdowns are of course inadmissible.

All positions have fixed StopLoss and TakeProfit values that are placed immediately at the moment of position opening. Were these values found during optimizations?

The TakeProfit level is the minimally accepted value by the Championship rules. The stop levels are also non-optimizable values. The EA was tested with these parameters on the whole available history, that is why there is no adjustment to a certain year.

All trades are closed by SL or TP, and this is normal for a strategy with such short targets. But at the same time we should pay attention to the following fact: during the first two months trades were closed with the profit of 4-5 points. But after trading conditions of EURGBP were made stricter by the Jury decision, your EA started taking larger targets - 10, 11 and more often 12 points. So, we can conclude that the EA correctly takes into account current values of the spread and StopLevel. We can conclude that changing of trading conditions was favorable for your EA, wasn't it?

As I have already said, the EA has the potential of increasing the TakeProfit level. Well, the minimal one is selected, it could be larger. As for various checks, they are the largest part of the Expert Advisor. As the Expert Advisor is designed for the live trading, I tried to include the checking of almost everything possible by programming a reaction on possible problems so that the EA could react on the change of market conditions. No matter what happens, the EA will do its best to make profit.

What would be results of this EA with longer targets? Have you checked it?

For long-term targets I use other Expert Advisors. Do not forget that the Expert Advisor presented in the Championship is a "scalper". And its main target to get the maximal profit in the minimal time period.

Are you satisfied with the EA's behavior in the ATC 2008? Do its results or behavior differ from those obtained during testing?

In general I am satisfied. For the "cut" version competing in the Championship the behavior is quite regular - drawdowns, etc. The EA's behavior in the Championship and that in the tester are similar. Though, during testing I didn't expect that the trading conditions would change.

Alexey, thank you for the interview. We wish you good luck!