Interview with a forex champion

Tony Marso from the USA participates in the Championship under the nickname of 'amelabs'. Before starting to trade forex, he was interested in equity and index options. His experience in the forex market started in 2004. In 2006 Tony learned about the automated trading. For Tony who by that time had been a software developer for over 24 years it was not so hard to start programming his own trading robots. Currently Tony is working on a forex strategy ebook.


You can be called our regular Participant. Last year your result was not a very good one. What lessons have you drawn from the ATC 2007?

I think if there's anything to learn from the previous contest it is that you can't predict what the market will be like when the contest starts. I designed my EA expecting the market to trend a certain way from the very beginning, but instead the trend reversed early on. My EA at the time was not designed to handle consolidation periods and failed terribly. Another thing I learned is that if you think your EA is aggressive and you enter a contest, you'll soon discover there are many people whose EA's are much more aggressive.

Can you call one or two names of last year Participants who had more aggressive EAs in your opinion?

Honestly, I thought I could win the contest with an ending equity over $30,000. So I designed my EA to try and capture steady profits that would cause it to end with a balance of just over $30,000. Clearly, anyone who ended with more equity than that was thinking much bigger than I was. Just about anyone who ended in the top 10 beat my projections for the contest. Most especially was "Better", who had over $100,000 at the end. I couldn't even imagine increasing my account ten-fold or more. In my mind it just wasn't possible without a tremendous amount of luck. By the end of the contest in 2007 I realized my thinking was faulty.

Please tell us a little about yourself, about your acquaintance with the Forex market.

I learned about forex one night while watching TV when I saw a commercial for a product that helps you trade forex. That was around 2004 I think. Before that I was trading equity options and index options. The idea of 24-hour trading was very appealing to me, so I attended some forex seminars, signed up for a demo account on FXCM and immediately started playing around with forex. After a few months I was convinced I could make it work, then proceeded to start trading live and drained my first 4 live accounts, each in a matter of days. However, that was not enough to stop me. I could see the potential and knew it was just a matter of trial-and-error until I could make it work for me.

Along the way, I came across some successful forex traders, but had no idea about automated trading. It wasn't until about 2006 that I learned about MetaTrader. At that point (being a software developer for over 24 years) I knew that I could make this work but it would take time. I tried several people's EAs with no success. I tried to write my own and had some success but they stopped working if the market conditions changed. It wasn't until a few months before the 2008 contest that I finally had an EA that I thought was worthy of trading a live account. After a few months of successful trading, I thought I'd enter the EA into the contest and see how long it would last.

What is your general trading strategy?

I do not like indicators. Most indicators just tell you what already happened. I use two moving averages to determine the overall trend and I use recent price action to determine when to enter and exit a trade. Although I consider myself somewhat of a scientist, I try never to apply a lot of complex math, indicators or oscillators to my strategy. In my mind I should be able to look at a naked chart: the 4 hr and the daily ones, with no indicators, and have an idea of where the market is going. I can look at the last complete bar on both charts and almost be able to decide if it's a good time to trade or not.

You are using the same Expert Advisor on a live account, though with lower risks. What results does it show now?

Actually, I'm not doing so well on my live account recently. Just today I took a big loss, even though the contest EA is doing very well. It will recover so I'm not worried. One thing I need to mention is that my EA (the Cross Trader) is not quite perfected. It cannot work for long term without human intervention. If left unattended for an extended time, it will die a horrible death unfortunately.

However, I did not design it for long term growth. Instead I designed it for what I call "explosive short term growth". If I use it and do not regularly make withdrawals from the account, then I will lose all of my money in a matter of months. But, for regular cashflow it is wonderful. As you can see, at the most aggressive settings it can quadruple your account in a matter of days. However, it can also drain your account just as quickly. My hope for the contest is that the downtrend on GBPJPY persists long enough for the Cross Trader to survive the entire contest.

How much do risks of your EA on a live account differ from those of a competing one?

In the contest I trade approximately 1 mini lot for every $500 equity I have in the account. In my live account I trade 1 mini-lot for every $2000 I have in my account. It is a factor of four increase in the contest.

As for now you haven't had any loss trade. Is result the same on a live account?

I have had losses on my live account. In fact, I drained the account one time right after making a withdrawal. Basically what I do is start out with a small amount of money, then let the EA double my account. Once the account is doubled, I take out my original deposit and now I'm trading with zero risk. I let the EA run until it looks like a reversal in the trend is coming, then I make a big withdrawal and let the EA continue. Recently, when I made the first withdrawal to eliminate my risk, the trend changed and the remainder of my account was drained. I lost nothing in actuality but I still consider it a loss.

You have written that your EA "places trades only when the probability of price reaching a specific target is very high". Could you please specify what targets you are talking about?

Unfortunately, I can't talk about it at this time because I am working on a forex strategy ebook that goes into that in detail. I can tell you, though, that I use a candlestick pattern to decide where to set my stop-loss and take-profit, and my take-profit is hit about 85% of the time under normal conditions. In times of extreme volatility such as this last week, my trades tend to hit the stop-loss with a profit instead of hitting my take-profit.

Why have you chosen trailing stops instead of fixed StopLoss and TakeProfit levels?

In my experience, setting a fixed stop-loss is a recipe for disaster. Instead, what I do is set an initial stop-loss so far out that it's not likely to ever be reached. Then I can withstand some very large drawdowns immediately after the trade is placed. Once the trade goes in my favor and I realize a specific candlestick pattern, I know it's time to start a trailing stop. As far as take-profit goes, I generally set that before stop-loss. The procession is this...

I place the trade at open of a new bar.
At the next open bar I set take-profit if I recognize a certain candlestick pattern.
If my trade is in profit at the 3rd bar but my take-profit isn't hit yet, then I start looking for opportunity to start a trailing stop. Again, this only happens if a specific candlestick pattern is recognized.
When is the "panic mode" activated? What conditions should be fulfilled for this? Has it ever triggered on your live account?

I have 2 panic modes that determine if it's time to close a trade. The first one says "close all trades if a bar closes inside of the two moving averages". The next one is riskier and says "close all trades if a bar is completely encapsulated within the two moving averages". This prevents what I call the suicide trade or trade-of-death from occurring, where I place a trade with a stop-loss of, say, 900 pips away and the price immediately goes against me all the way to hit that stop-loss or drain my account completely. As I mentioned before, if I don't intervene occasionally with my EA, that suicide trade will occur eventually.

Why were all your orders opened with the same initial StopLoss level equal to 197.42?

It is the price of the highest bar within a certain number of bars. I can't remember exactly what number of bars it is, but it is a fairly high number. Possibly over 100. This was just a function of the optimization to get the ideal parameters for trading live. Possibly if I ran the optimizer today it would produce a smaller number.

Why is it based on the Bid price, not Ask? Sell orders should be closed by Ask price, which is higher than Bid by the spread size.

Good question. This might have been an oversight on my part. I thought I was using Ask price for sell orders.

The TakeProfit level was not set at the moment of order opening. But then at a favorable price movement fixing levels were placed and 4 positions were closed by them. It is a rather rare method of order closing. How did it appear in the system? What do such TakeProfit levels mean and how are they calculated?

I believe that most traders know their exit point before they enter a trade. For an autotrader, this is not practical and definitely not necessary. The Cross Trader actually enters a trade at open of a new bar and sleeps for the duration of that bar, just to make sure I'm not setting take-profit prematurely. After at least one bar has completed, I can look at what happened and determine what is the most likely place to set my take-profit. Under ideal conditions, the Cross Trader will enter a trade at the beginning of a bar, set the take-profit at the start of the next bar, and be out of the trade before the third bar is completed. This is under ideal conditions.

Why have you chosen the GBPJPY pair?

I ran the optimizer with about 10 different pairs, mostly the majors and the most volatile. When all the results were available, GBPJPY stood out as the clear choice for the Cross Trader. There were other pairs such as GBPCHF and EURJPY that worked very well, but none came close to the results and consistency I had with GBPJPY.

Have you ever used multicurrency Expert Advisors?

I have not yet done so. However, I have been considering a version of Cross Trader that monitors several pairs at once. I might be working on that after the contest is over.

Thank you, Tony, wish you the best!
----------------------
SOURCE = http://championship.mql4.com/2008/news/401