Yesterday I received a message on the website's chat from a visitor requesting the review of a new automated trading system called forex retracer. In order to satisfy this person's request (by the way, thanks a lot for leaving review suggestions ! :o) ) I am going to review the forex retracer expert advisor today. As always, my objective will be to judge the soundness of the expert advisor's trading tactics as well as the validity and reliability of the evidence provided on the website as a backup for the system's overall profitability. After carefully analyzing all this evidence and going through all the material available on teh website I will give you my opinion about this trading the system and whether or nor I consider it worth buying and testing. As always, my review will be as unbiased as it can be as I receive no compensation for giving either negative or positive feedback about this trading system.
I can say that the forex retracer website seems to be taking the totally opposite approach to all the commonly overhyped trading systems being sold. I don't think that expert advisor sellers should take such two extremely polarized points of view. I am all for not having a lot of cheesy marketing and fake stories but I don't think that you should only place back/live testing statements on a website to sell a trading system. It is always good to have some information about the trading system's overall trading tactics, the type of market movements it wants to catch, its expected risk/profit targets expected from the live/back testing evidence, etc. It is not good to have a bunch of statements with no information about the system's overall trading tactic since trading such a system in a successful manner would be terribly difficult since it would be very hard to trust a system you know absolutely nothing about.
Now, for what the people selling forex retracer tell us, it seems that they want us to figure the profit/risk targets for ourselves from the evidence they provide. Looking at the website we see that the evidence provided is nothing but backtesting statements from 2007 to 2009 on 30 minute charts. It becomes evident after looking at these backtests and the profit figures they reach that some serious exploitation of backtesting faults is happening. This can be explained in terms of the TP values used on the AUD/NZD which are prone to a very strong generation of one minute interpolation errors due to their small values (relative to the pair's spread of 8 or more pips). It becomes evident that this backtesting evidence simply cannot be trusted as reliable due to these obvious problems.
Then we see that there is absolutely NO live testing evidence of the forex retracer trading system. The same question always comes up. If the system is so profitable, why don't the creators trust it with their own money ? Why don't we have any live testing results ? To me it seems obvious that backtests are overestimating profitability to a GREAT extent (making them useless) and live testing results are bound to generate VERY different results when compared with simulations.
Now, when we analyze the backtesting statements provided things are even worse. The expert advisor uses a risk escalation technique which is even beyond the exponential growth of a Martingale (this however has been debated by several people who say that the progressive money management only happens once). For example, a loss which happens on 13/09/2007 with a lot size of 1.5 causes a lot size escalation to 5.6 lots in the first trade. More than three times the lot size used before ! Add to this the fact that the expert's risk to reward ratio is about 5:1 and you get that the second trade would cause a loss almost 5 times bigger than the first one.
To sum it up, even though the intentions of the sellers of forex retracer is to provide a no-hype system based on only "reliable evidence" we find that the system does not offer any reliable evidence of profitability. The forex retracer is merely and expert advisor which is being sold on unreliable backtests which are almost with certainty not going to be in line with live testing results. The overall lack of live tests is also not good since it points out that the authors are not willing to put their money on the line to test their own system (never a good signal). (this however has changed recently as the author seems to be setting up a 5000 USD live account to test the system). The trading tactics also reveal very unsound money management with a very unfavorable risk to reward ratio and lot size progression which is bound to make this EA very risky in the long term.
In the light of the unreliable backtests, the complete lack of live tests and the use of unsound trading tactics with a very bad risk to reward ratio, I consider the forex retracer NOT worth buying or testing. The creators would need to remove the progressive money management, increase the size of the TP to allow for accurate simulations and provide at least 6 months of live testing evidence (with investor access verified) in order for me to consider redoing this review. Meanwhile, the forex retracer is what appears to be another extremely risky trading system which is sold on unreliable backtesting evidence obtained in virtue of one-minute interpolation errors or similar tactics.
However, as you may see on the comments, I have been contacted by the creator so that new live testing evidence regarding the profitability of this EA can be provided. As always I am opened to debate and I strongly appreciate the will of the creator to provide additional evidence. This review will be rewritten during the following month once results of a 5000 USD live account funded by the developer start to come out.
If you would like to learn more about my journey in automated trading and how you too can start to learn how to trade and use automated trading systems for long term profitability please consider buying my ebook on automated trading or joining Asirikuy to receive all ebook purchase benefits, weekly updates, check the live accounts I am running with several expert advisors and get in the road towards long term success in the forex market using automated trading systems. I hope you enjoyed the article !
I can say that the forex retracer website seems to be taking the totally opposite approach to all the commonly overhyped trading systems being sold. I don't think that expert advisor sellers should take such two extremely polarized points of view. I am all for not having a lot of cheesy marketing and fake stories but I don't think that you should only place back/live testing statements on a website to sell a trading system. It is always good to have some information about the trading system's overall trading tactics, the type of market movements it wants to catch, its expected risk/profit targets expected from the live/back testing evidence, etc. It is not good to have a bunch of statements with no information about the system's overall trading tactic since trading such a system in a successful manner would be terribly difficult since it would be very hard to trust a system you know absolutely nothing about.
Now, for what the people selling forex retracer tell us, it seems that they want us to figure the profit/risk targets for ourselves from the evidence they provide. Looking at the website we see that the evidence provided is nothing but backtesting statements from 2007 to 2009 on 30 minute charts. It becomes evident after looking at these backtests and the profit figures they reach that some serious exploitation of backtesting faults is happening. This can be explained in terms of the TP values used on the AUD/NZD which are prone to a very strong generation of one minute interpolation errors due to their small values (relative to the pair's spread of 8 or more pips). It becomes evident that this backtesting evidence simply cannot be trusted as reliable due to these obvious problems.
Then we see that there is absolutely NO live testing evidence of the forex retracer trading system. The same question always comes up. If the system is so profitable, why don't the creators trust it with their own money ? Why don't we have any live testing results ? To me it seems obvious that backtests are overestimating profitability to a GREAT extent (making them useless) and live testing results are bound to generate VERY different results when compared with simulations.
Now, when we analyze the backtesting statements provided things are even worse. The expert advisor uses a risk escalation technique which is even beyond the exponential growth of a Martingale (this however has been debated by several people who say that the progressive money management only happens once). For example, a loss which happens on 13/09/2007 with a lot size of 1.5 causes a lot size escalation to 5.6 lots in the first trade. More than three times the lot size used before ! Add to this the fact that the expert's risk to reward ratio is about 5:1 and you get that the second trade would cause a loss almost 5 times bigger than the first one.
To sum it up, even though the intentions of the sellers of forex retracer is to provide a no-hype system based on only "reliable evidence" we find that the system does not offer any reliable evidence of profitability. The forex retracer is merely and expert advisor which is being sold on unreliable backtests which are almost with certainty not going to be in line with live testing results. The overall lack of live tests is also not good since it points out that the authors are not willing to put their money on the line to test their own system (never a good signal). (this however has changed recently as the author seems to be setting up a 5000 USD live account to test the system). The trading tactics also reveal very unsound money management with a very unfavorable risk to reward ratio and lot size progression which is bound to make this EA very risky in the long term.
In the light of the unreliable backtests, the complete lack of live tests and the use of unsound trading tactics with a very bad risk to reward ratio, I consider the forex retracer NOT worth buying or testing. The creators would need to remove the progressive money management, increase the size of the TP to allow for accurate simulations and provide at least 6 months of live testing evidence (with investor access verified) in order for me to consider redoing this review. Meanwhile, the forex retracer is what appears to be another extremely risky trading system which is sold on unreliable backtesting evidence obtained in virtue of one-minute interpolation errors or similar tactics.
However, as you may see on the comments, I have been contacted by the creator so that new live testing evidence regarding the profitability of this EA can be provided. As always I am opened to debate and I strongly appreciate the will of the creator to provide additional evidence. This review will be rewritten during the following month once results of a 5000 USD live account funded by the developer start to come out.
If you would like to learn more about my journey in automated trading and how you too can start to learn how to trade and use automated trading systems for long term profitability please consider buying my ebook on automated trading or joining Asirikuy to receive all ebook purchase benefits, weekly updates, check the live accounts I am running with several expert advisors and get in the road towards long term success in the forex market using automated trading systems. I hope you enjoyed the article !
Posts
12 comments:
Hey Daniel,
I had a question. I was looking at the site of this Forex Retracer and noted something I see on this EA and a few other EA sites. When they show a statement the trades, I see the trade is made without SL and TP, then immediately after they do a modify to throw out the SL and TP on the trade. What is the benefit of this?
JT
that's funny because http://www.forexpeacearmy.com/public/review/www.ForexRetracer.com
shows all kinds of people who are trading it live and having success. Some of your points about the website are valid, but you would think you would look at other people who have used it and at least test it before you go bad mouthing it.
Hello Everyone,
@JT : Thank you for your comment. The fact that the SL and TP are introduced after the order is a necessary aspect for ECN compatibility. ECN brokers do not allow the placement of orders with a SL and TP but these have to be introduced after the trade is entered.
@zulemara : Indeed, a very valid point and something people have told me several times. However there are several reasons why I don't take people's opinions into account when evaluating systems :
1. Many systems which are very risky and unprofitable in the long term tend to give short term profitable results, something which makes this VERY dangerous. People are bound to get positive results in the short term and to wipe accounts in the long term. It is VITAL to have proof of long term profitability, 10 year backtests with investor access verified live testing performance.
2. There is no way to know if people's opinions are real of made up by the seller.
3. The author should provide PROOF about the profitability of a trading system without any need of third party testing. Why would the burden of proof be on the buyers ? The burden of proof is WITH the author of the systems.
I know that my reviews are hard to understand sometimes but I base my opinion ENTIRELY on the evidence provided. I call it like I see it :o) You can learn more about my review here : http://fxreviews.blogspot.com/2010/01/basics-of-my-reviews-understanding-my.html. Thanks a lot for your comments !
Best Regards,
Daniel Fernandez
I'm sorry to say, but your review is dead wrong. you have not done your homework here. First there are a number of results out there, one of which is the demo test done by forex peace army. Second, you make many assumptions. First the recovery trade is done once and thats it. You can turn it off, it still backtests well with it on or off. Finally, I trade this live as do many other people, the creator actually trades it live as well. Will it work forever? I don't know, but if you post a review, please don't just generalize.
Daniel,
I really don't even know where to start here. You post a lengthy, overwhelmingly negative review on Retracer based entirely off the little info you saw from my website? I am appalled to say the least.
I think the least you can do for your followers is actually review MY PRODUCT ITSELF. All you have really done for them is review my website.
I would like to respond to your numbered responses to zulemara though.
1. I have backtested Retracer using the most complete data I could find. I have yet to find a broker that supplies historical data on AUDNZD beyond July 2007. I kindly challenge you to find this data and I will GLADLY backtest it as well. (I have given up on it.)
2. The reviews on FPA are screened. Every single one of those reviews are from real customers. If you want, go ahead and pick a random review and I will try and have that person contact you and share their experiences with Retracer. Or pick a few if you want.
3. This I disagree with completely. How many times have we seen fake or modified results from the developers themselves? Can we TRULY trust the EA developers to give us accurate results? What better way to prove a systems profitability than from ACTUAL customers??? I have a handful of customers that I'm sure would be more than happy to show you a screenshot of their accounts.
Regardless, I will say again that I am disappointed in your unfair review of my expert advisor that took me nearly a year to develop.
Regardless, I look forward to working with you and hopefully getting some of the inaccuracies in your review corrected.
Hello TheTonik,
Thank you very much for your comment :o). I am sorry if you feel that the review I have given your product is unfair but it is my honest opinion about what I have observed on the website. I am sorry if I am wrong about the progressive money management (which you say only happens once) but I was unable to locate further consecutive loses to see if this is indeed the case. I am also aware of the fact that further data on the AUD/NZD is hard to find, reason why I didn't put special emphasis on the limitation of the backtesting results (I think that data from metaquotes servers goes back to 2004 but I have not evaluated the quality of this data from 2004-2007). However your EA is bound to strongly suffer from interpolation errors inherent to the backtester and this cannot be avoided with your current setup, making your profitability most likely greatly over-estimated in simulations.
However, I am all for fairness and being unbiased :o) Don't think that I won't give you positive feedback if you have the evidence needed. I don't want to hand pick any review but I would like you to provide me a link to a live test of at least 1 or 2 months (forward tests are bound to be inaccurate due to spread widening, slippage, requotes, etc, which are bound to cause demo/live differences on your system)with investor access to the account (or verification through myfxbook).
I do think that you have come a long way by posting here and I do appreciate a lot the fact that you care enough to debate my points :o) Again, I am all for being fair and I will listen to any evidence of profitability. Regarding trusting results provided by EA authors, I completely agree and this is the reason why I always emphasize that having INVESTOR access to live accounts is a MUST to believe the live results given by an EA creator.
I would then encourage you to post a link to these live results coupled with a backtest of this same period (to test live/back testing consistency). After you post this I will GLADLY redo the review, again, I am ALL for proven results and if your EA has enough evidence to be considered long term profitable, I will gladly give you a positive review.
Again, thank you for posting here and debating my views :o) I am all for discussion and being fair and I will definitely consider any evidence of profitability you provide. I try to keep myself as unbiased as I can be and I will listen to everything you have to say and show to defend your trading system and show evidence of its profitability :o). Thanks again for posting !
Best Regards,
Daniel Fernandez
Thanks Daniel. I am 100% open to constructive debate as well.
I am switching brokers to TheCollectiveFX, and going live with $5,000 of my own money this coming week (wire is in transit as we speak). I will gladly get you the results after one month from the day I go live.
It's worth noting however, and my current customers are all aware of this due to my close relationship with them via a blog much like yours, that I will be running Retracer on a lower risk 'News Filter' version. This version trades much less often, but cuts stoplosses by 80%. Customers have the option of running the low-risk News Filter, or they can go all-out using the settings I used on my website. The choice is theirs.
Obviously this is up to you, but perhaps you could add a note in your review that the 'final verdict' is pending while you work together with the developer to get more concrete live trading results, or something along those lines.
What do you say?
-Kyle
Hello Kyle,
Thank you for your comment :o) I am glad that you will be running a 5000 USD live account with your EA, it shows that you are willing to risk your own money, meaning that you trust your own creation. I would encourage you to publish your results via myfxbook.com, this way investor access can be verified. I would also encourage you to create more accounts on additional brokers so that broker dependency can be evaluated.
I have also placed a note on the review saying that a live test is under way and that the review will be rewritten once this is done :o).
I will be checking out your website looking for the live statement results during this month. At the end of the month please provide a back test of this same live period so that we can compare live/back testing consistency :o).
Again Kyle, I would like to thank you very much for your willingness to debate my views and fund a live account, it shows that you are truly interested in giving reliable evidence of your system's profitability. However I am still worried about the over-estimation of profitability given by simulations and the risk escalation seen on the backtests (which increases risk almost 5x from the first losing trade). At the end of the month, live/back testing consistency analysis will be able to answer the first question, while the inherent market exposure gained by the second may only become apparent with time.
Thanks again for your posts and participation within this review's discussion, I will be looking forward to your live results :o)
Best Regards,
Daniel Fernandez
Concerning risk escalation..
This is what I call Recovery Mode. Because Retracer has such a high win percentage, the chance of hitting two stoplosses in a row is very small, so it's a calculated risk.
By default, Retracer opens a single trade at 4x the lot size of the original losing trade (1x4). I caught a lot of flak for this, so I reduced the default to a single trade at 3x (1x3). If customers feel this is still too risky, they can modify recovery mode however they see fit. They can do 2x2, 1x2, or disable it entirely. Recovery mode is NOT mandatory, and Retracer still excels wonderfully with it off.
I am now working on Retracer v2, which will incorporate a slightly different strategy. This strategy will not need the recovery mode, so it will be disabled by default (but still there for anyone who chooses to use it, as our win rate will be even higher than it is now).
Hello Kyle,
Thank you for your post :o) I am glad that this is the case. However my concern about risk escalation comes from the over-estimation of profitability on backtests, you may as well be underestimating the occurence of consecutive loses, which is something common with this type of strategy and the heavy presence of interpolation errors.
However if the software excels without the recovery mode it would be great to have live results without it also included (even a 100 USD mini account on IBFX would be great). On your second EA I would also encourage the use of a TP of at least 10 times the spread on whatever currency pair you intend to use to avoid one minute interpolation errors in backtesting, this will allow for more reliable simulations and the accurate evaluation of long term profitability. I think that you are currenly under-estimating the magnitude of this problem which can be in fact VERY big.
Again, thanks a lot for posting your views, it is very interesting to have an EA creator with such a cooperative attitude :o).
Best Regards,
Daniel
Daniel I believe you are over-estimating this whole 1 minute error thing you keep bringing up.
If backtesting is so inaccurate, then why is it that after every trade I have I go back and perform a backtest and it matches up nearly to the tenth of a pip? Sure seems accurate to me.
Now if Retracer was a scalper you would have a very valid point. But I believe you are simply wrong about a system with a 15 pip TP not being accurate on backtesting. My forward trading and backtesting match up pip-for-pip, or come very close. So I believe your interpolation error theory is flawed.
Retracer will stay at a 15 pip TP. Anything more than that will completely invalidate the strategy it uses.
My only concern, Daniel, is will you admit that you were wrong to your followers about Retracer once I prove its profitability?
Hello Kyle,
Thanks again for your comment :o) Well, the extent to which interpolation affects a certain strategy depends on the currency pair and the exact exit strategy used. However my experience in the past with systems similar to yours (similar TP Vs spread relationships and currency pairs) has warned me about these errors and the high over-estimation of profitability they cause. With a system trading at barely 2 times the average spread these errors are bound to be significant. Definitely I can be over-estimating the importance of one minute interpolation within your strategy but again, the evidence has the last word about this.
I believe that you have to backup anything you wish to say about your system matching or not matching backtesting results, therefore if you have a one month live testing record we can verify through investor access and compare with backtesting I'll be glad to evaluate it.
Regarding saying I was wrong, definitely :o) ! If you prove that your system has back-live testing consistency through a one month backtest of your system coupled with a live testing statement of this same period (verifiable through investor access) I will be VERY happy to redo my analysis and call your system worth buying and testing. I am ALL for authors who can backup their claims and I congratulate you for wanting to make the effort :o)
Again, I am all for being fair and as transparent as possible. I'll be waiting for you to post the evidence then as it arises and I'll be happy to redo my analysis once you present the requirements necessary to truly validate your simulations through live testing. Thanks a lot again for your commenting and contributions :o)
Best Regards,
Daniel Fernandez
Post a Comment