The “Ugly Truth” Story of a -14.6 pips Max Loss to 96.7 pips Risk on a Single Trade.
Recently I learned some lessons the hard way. I thought that by sharing them with you, you might have a chance to learn from these lessons without having to make them for yourself. This article is a case-study of these experiences of mine the past 45 days. You will be able to save yourself from some trouble if you will learn from these lessons here that I will describe for you. I hope you will learn from them as I have myself.
I have been working as an MT4 programmer for the last few years. All currency trading tools that I build are about honesty and the idea behind them is to make a Forex trader’s job easier. I hope that, beyond your own education, you will be able to profit from these things I have discovered with my recent purchase as well.
I practice positive mental attitude and do not like to talk bad things about other people or their work, so I will not reveal the actual website where I have purchased this trading robot. I just want to share the things I had learned from this test.
It Starts With The Outcome Of A Competition
I recently saw a sales letter for a product that was very compelling. It was not down at my local convenience store or shopping mall, but rather it was for a new product being launched online. This new product was a robot, but not the kind of robot that could help me with doing the dishes or with building a car. This was the kind of robot that (claimed) could earn money. As a programmer and currency trader I was excited to try it out.
The Forex Trading Robot Competition
It was a Forex trading robot that was making some very fantastic claims about its earning potential. It all started with a currency trading contest that was being promoted online. I did not see the entry notifications or otherwise I would have entered my own currency trading software into the competition.
The premise of this competition was that the winner would receive a large amount prize money and also the title of winner of this highly competitive trading challenge. The challenge, as I understood it to be, had lasted 90 days and there was a number of different trading robots entered. After the 90-day period there emerged a winner as well as the runners-up.
The first and second place winners were both interviewed and their robots were put on sale at the conclusion of the trading championship. It was a package deal for about $1000 USD which included the first and second place winners. The people who were responsible for the competition and the sales promotion had included a 60 day money back guarantee so there seemed to be an acceptable amount of protection to purchase and demo these two winning robots.
What Had Happened During The 90-Day Competition?
This is a screenshot showing the date range and results of the championship for the 1st place winning robot. You can see that the competition ran from the middle of July until the middle of October. This is a good testing time that even includes part of the (sometimes choppy) summer market.
Obviously these results seem to be very excellent and nothing that any trader would ever complain about in their own account. These are the reported earnings and results from the winning robot of the trading challenge. This is the robot that I bought for nearly one thousand US dollars.
What I wanted to do, and the purpose of this case study, is to help you learn from my own lessons. There are some mistakes made that I think you will find very educational yourself as a trader. Let me share this experience of mine with you now. I’ll tell it chronologically in the order it happened…
Step 1 – Research The Trading Robot
I started where I normally start when making a decision to purchase something online—researching and reading reviews. I have my own 5 Step Proven Process to Selecting a Winning Forex Robot. I looked up all of the information I could find online about the opportunity to purchase the robots, but did not find all that much. Going online to one of my favorite sites, ForexPeaceArmy.com, I did not find much information there.
I continued to watch the promotional emails as they came out. These emails announced how things were going with the trading competition (I only started seeing the results at the very end) and which robots were winning. Everything was made very clear and very exciting.
NOTE: The sales material and marketing of this robot was very very good. But do not believe everything you see or read, especially when it comes to Forex marketing promotions. You’ve probably been there before, right, purchasing a $97 Forex robot only to see the money in your account dwindle when you ran it live? We’ve all been there before! Anyway, make sure to always wade very carefully into any sales letter and especially so when it comes down to returns promised or indicated by a Forex trading robot.
Note: It might help me to explain also at this point that I am not a newbie to the idea of testing and working with trading robots. As an MT4 programmer, I have designed several of my own Forex trading robots in the past and have created hundreds of Expert Advisors for my customers based on their private rules. One of my own robots doubled a live account in its first year of live trading. It is the one I used to start currency trading signals service on my own. I mention these things just so that you will know of my first-hand experience with building trading robots. It is a passion of mine and one that I love to share with others. I wanted to stress to you also that it is important to me to test endlessly. It is good to test new ideas for new robots as well as additional settings for existing robots. By endlessly testing, a robot can be built strong, stable and extremely profitable. As with all robots, there must be a good combination of reward potential saddled to strict money management and risk management rules. It doesn’t matter how much a robot can win if the drawdowns (periods of losses) can wipe out the account.
The Company Offering The Robots After Competition Provided Trading Account Verification As Proof
One of the things I always want to see with any trading system is the proof. I will tend to trust the proof that comes from MyFXBook.com, which is a reputable review and analytics website for currency traders. You’ve likely heard me talk about that website a lot in previous articles—it’s a wonderful site to use to help evaluate trading accounts.
IMPORTANT NOTE: One of the challenges with sales pages can come from pure dishonesty. I am not claiming that this robot’s sales page is/was dishonest, but it is possible. Some marketers will take screenshots of performance figures and then simply alter them using Photoshop. This is a very bad practice that some marketers and advertisers will engage in. You might have had this bad experience before when making a purchase that didn’t turn out like what was originally advertised. Earnings statements can easily be manipulated using graphic design software. Try and read reviews from real users before trusting the images that you see on any sales page. My advise here is don’t trust a screenshot from myfxbook.com or any other site, ask for the link to the actual verification page. It is common sense that if the screenshot is valid it came from the myfxbook page that exists, so why vendor does not share that with others? What is wrong with that page if the vendor does not allow you to see that trading robot validation page?
There Was a MyFxBook.com Direct Link Provided
As you can see in the screenshots above and below that this was a myfxbook.com account. These screenshots I took myself from inside the myfxbook.com verification page. This was an important thing for me to see. The numbers are verified by myfxbook.com which mean that they are real and legitimate. The accounts are linked and verified using a special investor password account access, so this represented a real trading account. I analyzed the trading history on myfxbook.com and I did contacted the robot’s owner asking some questions. I was given temporary investor password access to that MT4 account so this was another thing that made me trust the vendor even more.
Note: It doesn’t matter too much whether it is a real money/live account or a demo account, so long as it is an actual account that is being trading. A live account is always better of course.
You can see from the screenshot here also that the trading style seems to be conservative and the stop loss levels are smaller than were the take profit targets. This to me represents good, solid trading.
I was interested, as almost anyone else would be, in learning more about this robot and trying it out.
Look at the image above. You can see that worst trade of -14.6 pips lost just 2.57% of account balance.
Step 2 – Purchase And Install Trading Robot
There was no trial period available so i just purchased robot and was excited to run on my live account. I purchased and set things up with the robot. There is not much to explain to you here, but it was another step in the process. Something that I would point out is that with just about any trading robot out there, certain steps are required to set up the robot. You want to set it up in the same fashion as what is shown in the examples (unless it is too risky for your own personal account). It is important to set things up exactly the same to try and achieve the same results. It’s obvious that you won’t get the same results as in the past, but what I mean is that you want the robot use the same trading technique and risk management style.
In many cases also, there are multiple different options that can be used based upon risk tolerance, account size and other factors or preferences. But in this case here I simply used the default settings that came with the robot. As it’s instruction manual said, the ultimate risk settings are default settings. Also I have followed one of the main trading robot requirements and that is to use it only on an ECN account with a small spread. My ECN trading account had 0 – 0.2 pips spread.
Actually it may be a huge problem if the vendor does not give you recommended settings to use or trading robot is not working with default settings. This way you do not know what settings to use to get the same trading style the vendor uses. So before you purchase any trading robot, make sure vendor gives you ultimate settings. I know this is not the case for each trading robot. For example if you buy a trading tool, an application that will help you do your work easier by completing some of your tasks for you, you might need to adjust the settings to fit your own needs.
Here is an example. One of my trading tools, the Trendline Trader, is build for currency traders who uses trendlines to enter the market. Now this EA works with default settings, but I never promise any profits. Actually, first you need to draw a trendline manually on the MT4 chart and then set this EA to use it. You will have to read the whole instruction manual to understand each of the EA settings and then set it up to fit your needs. It’s a trading tool to use to avoid sitting in front of your computer all day. If you do not want to spend the whole day or even week waiting for a price to touch your trendline so you could open a long or short position, such trading tool like Trendline Trader may be the best option. Now to make it usable for any currency trader who uses trendlines I have added a lot of options and settings.
With trading robot claiming to make money, you want to get exact the same settings that the vendor or the author use. If you get the robot without settings you just won’t know if it is working like it used to before.
Step 3 – Be Patient, Wait For Results… Then Measure The Results
After installing the software and letting it run for just a few days I started to noticed some problems. Generally you will always want to let a software run for a reasonable amount of time before drawing any conclusions. In this case though, I noticed an issue right away that gave me cause for concern.
On the sales letter for the robot, and in the myfxbook.com account they used for verification, the maximum loss listed was just -14.6 pips. This was encouraging because the robot appeared to be scalping with reasonable take profit and stop loss levels. It was exciting to see as you can imagine.
But then, guess what the first three trades looked like…
The first two look like normal trades, even though the 1st trade already lost -31.6 pips, which is more than twice bigger compared to the max losing trade reported on the website.
Now the 3rd trade was the one that made me nervous and after that I completely lost trust in this trading robot. To be exact I am talking here about a trade #563014. It’s a positive trade of just a couple of pips and then a stop loss level of just under a hundred, to be exact it was 96.7 pips. So this didn’t seem to line up at all. From 14.6 pips max loss to 96.7 pips stop loss after I start testing the robot. Is this a coincidence?
Now the 2nd trade #561881 which shows a small stop loss value which was set at 1.33688 when the open price was 1.33658 (3.0 pips), but that’s a stop loss at the time when the trade was closed. The original stop loss is not visible in the “Account History” or myfxbook.com. It is only visible in the log file which are usually located at the \MT4-folder\experts\logs\ and \MT4-folder\logs\
Here is the trade entry details from the log file.
15:00:01 EAName EURUSDc,M15: open #561881 buy 0.15 EURUSDc at 1.33658 sl: 1.33458 tp: 1.34058 ok
As you can see the original stop loss was placed at 1.33458 at first (20.0 pips), but later it was modified because EA use trailing stop. Now there is nothing wrong about that, I just want to show you that stop loss from the “Account History” or myfxbook.com may not be the original stop loss used at trade entry. This makes us not know what the risk was at the entry time. Now in this case initial stop loss was at of 20 pips which is not huge, but still more than the max loss of 14.6 pips reported on the sales page.
There was the potential that the robot placed this large stop loss of 96.7 pips on the charts but then would keep an internal stop loss level which was hiding from the broker. At least that is what I would want to believe, however in the chart image above it is clearly seen how the 3rd trade was at a drawdown of more than 50%. So if EA use a hidden stop loss it is still too big comparing to the max loss reported. I would guess that this trade was closed manually when it reached few pips of profit, because there was no trailing stop used on it. If this trade would have hit the stop loss level, it would have taken out about 15% of the account… this would be an unacceptable risk and definitely not what has been promised. This robot was trading with huge stop losses that were many times the size of the take profit levels. Previously you saw an image where you can see that worst trade of -14.6 pips lost just 2.57% of account balance during the first 3 months before robot was released for sale. So it’s almost 15% risk now when there was max risk of 2.57% only before.
It could have been possible that the trader (during the competition) was closing trades out manually when they appeared to go in the wrong direction. Because of this possibility it no longer becomes a trading robot, but merely a gamble for those traders who do not have the vast experience of the trader who was running the software.
But as you can imagine, this was quite disheartening to see as the first three trades. The stop losses were so much wider than the profit potential which means that the robot must be right nearly all of the time. This is a tall order for any trading system.
Now I am used to big stop loss levels and that is something I personally use in my own trading signals service, but I clearly explain that on my website so that there were no misunderstandings later.
The Initial Two Weeks Testing Period
Over the first two weeks there was nothing very spectacular to speak of. The robot appears to be trading far less than what was seen during the competition times. The promises of doubling money every couple of months seems to be untrue, unless possibly trading at ultra-high risk settings (which is not recommended within the instructions.)
The best thing to do at this point in the exercise is to continue testing. Be patient and allow the test to continue on.
NOTE: Because this was a highly promoted product with a high sales price and offering a risk free guarantee, it should be entirely possible to ask for a refund should things not go well during the initial testing. Some companies can be unscrupulous when it comes to actually honoring a refund policy but hopefully this company is not one of them (should the need arise).
As soon as the contest was closed, the MyFXBook.com link was no longer updated so it seemed very suspicious. However, at the time of purchase, it wasn’t a big deal because the trading history was just a few days old.
The account continues to be updated on myfxbook, but it is not traded anymore so there are no new trades to display. We cannot see if it works fine on our accounts as we cannot compare them with any other account being traded by the company. It is always a nice thing if a company will keep an account running so that the customers can verify that their trades are the same as the robot’s trades run by the company. In this case we are not so lucky.
So basically if the EA fails, we do not even know if it is the same robot that was used during the competition that corresponded with the MyFXBook account that was shared on the sales pages. This is a little suspicious to me and caused for a ‘raised eyebrow.’
Make Sure You Study The Trading Results Available With A ‘Fine-Toothed Comb’
If there is a MyFXBook.com account available then make sure to study it extremely carefully. If there are drawdowns over the period of the test time available, imagine what it would be like on your account and how much you would lose if such a drawdown occurred. It is very important to map such things out. Prepare for the worst and expect that drawdowns are going to happen. Are you trading with lot sizes that will allow you to stay in the game should the worst happen? If you are trading too aggressively then you might get wiped out before the account has a real chance to grow. Read more about risk management here in this article.
Very Important Note About Demo VS Live Testing
So what is this huge (and very classic) mistake that I made?
I foolishly loaded the robot onto a live account without ever first testing it myself on a demo account.
BUT, you say, CAN’T YOU JUST TRUST THE RESULTS DESCRIBED ON THE SALES PAGE?!
Well, you can do what you want with your own money, but I am telling you about a universal lesson that I have seen come true many times. It is most important—always when testing a robot—to use a demo account first. It doesn’t matter how good the verification stats are, or how good the sales copy is, you must/should always first test on a DEMO account.
You need to first get a feel for the way a robot performs. You need to see for yourself how a robot will perform under ‘real’ trading conditions before you ever consider putting it on a live account. You would be surprised how different things can be when moving from a back-testing (or even live) account shown in a sales page to a live account of your own. You might find out that the marketing team behind the product is dishonest and has artificially manipulated the results. (You would be surprised how easy it is for marketers to use Photoshop to alter images and results!)
Good thing that it ended not so bad for me. Even in the worst scenario the loss would not have been catastrophic, but we should measure by percentage, not pips or dollars.
Different Brokers Can And Often Will Perform Differently From Each Other
…And Thus The End Results Can Look Quite Different.
I’m going to shift gears here a little bit and share with you another important lesson that is related indirectly to the experience of the case study… Robots that are based upon scalping strategies are especially vulnerable to this situation. If you have just a 1 pip difference in spread between two different brokers and the robot you are trading is looking for 4-7 pips on each trade, your results could be thrown off by as much as 30% simply based on different spreads between brokers.
This is a common problem that you can find with the ‘social trading’ networks that are out there like zulutrade.com, etoro.com and currensee.com. When a provider trader uses one broker and you use a different broker from them there can exist a discrepancy in both the spreads and the times when the trades are executed.
These discrepancies will manipulate and distort the results from one trader’s account to another. It is a beautiful principle overall, this idea of social sharing trading, but it has some issues. Ideally you would have one trader who is very good and consistently profitable. That trader decides to share his or her trades with others and so enters one of these communities. Other traders follow the trades as published (usually in an automated account-to-account fashion) and everything seems fine and good—at first… within a few days the other traders start to notice that while the provider posts a 20 pip gain on a certain trade, their accounts show only a 16.3 pip gain.
How can this be? A frustrating experience, isn’t it!? This is a very common problem created by the discrepancies between brokers. Time delays, order execution delays, spread differences, slippage and even different price quotes between brokers are all issues that create these problems for traders in a forex social sharing environment. It is a universal fact that there are hundreds of different Forex brokers and they all do things a little bit differently. These differences can cause performance issues when trying to copy from one account to another, and thereby alter the overall results considerably.
In similar fashion to the social sharing currency trading platforms’ issues, we notice similar things in the world of robot trading. Sometimes a spread from a testing account owned by the makers of a software will encourage different performance than what is seen on other traders’ accounts when they run the same robot.
Now Back To Our Robot Testing Case Study (Again)
I wanted to summarize by letting you know that this testing will continue on and I hope to update you with a second part when it makes sense in the future. Since I have only been testing this robot for about 2-3 weeks it is way too short of a time to draw any solid conclusions.
With any Forex test, whether manual system or robotic, you should test for at least 3-6 months before beginning to draw any conclusions. I consider 6 months a good amount of testing time. Longer is always better. A year is a great amount of time for testing. But bear in mind also that market conditions can indeed change with time.
I look forward to seeing you back with the next update of this case study in a few weeks time! Go ahead and ask the questions you have below so that I can fill them in with the next installment. Good trading to you this next week!
So, have you ever made a mistake by putting your new trading robot on a live account instead of testing it on a demo account first?