Today I saw a comment on a blog that talked about a "great trading system that would come out in December". This new and exciting trading system is called 4xgreed and today will be subjected to my review. Even tough the system has not been released yet, the information provided by the authors is enough for me to analyze and decide whether or not this expert is worth testing or buying. By simply analyzing the trading results shown by the authors I can certainly gauge the likeness of this EA being long term profitable.
First of all, the expert advisors web page is fairly decent with no huge amount of hype or unnecessary marketing. In fact, absolutely no claims are done about the expert's profitability, reason why I imagine the author(s) prefer the clients themselves to decide the profitability in virtue of the live/backtesting the people at 4xgreed show. There is no clear explanation of the nature of the expert's trading but they do say that the expert is able to adjust to market volatility and market conditions via neural networks and genetic algorithms. To me, the mentioning of these factors is just to make people feel the system is "advanced" as there are no proven results in live testing that have shown that a strategy can in fact benefit in the long term from any of these two approaches. In fact, neural networks have been proven to overfit expert advisors to the past, preventing effective adapation to "actual" market conditions.
The most interesting part of this expert comes from the testing section. The people at 4xgreed have done a good job at presenting several brokers, account sizes and backtesting results. This is a sure way to show an expert is both broker consistent and profitable. However, by analyzing the trading statements, several things come to mind.
The first bad characteristic of this EA is the low value of the TP. With a TP in the order of 14 pips (EUR/USD), this expert advisor will sure have problems with one minute fractal interpolation in a very significant manner. Furthermore, the fact that the risk to reward ratio is 65:14 is also not very conforting. This disparity between the stop loss size and the take profit size makes the errors in backtesting favor winning trades more than losing trades because of their much smaller size. This is evidently seen in the ten year backtesting that shows the typical, unrealistic returns when fractal interpolation errors are predominant. An almost 5:1 risk to reward ratio is also not a very favorable characteristic since it means that the expert is relying on the randomness of short term movements to gain profits, rather than reliable trend following or similar types of trades.
Live testing at the moment is rather limited so there is no room for big conclusions regarding them. The only thing we could say is that, as it is inferred from the above discussion, there is disparity between live and backtesting which makes backtesting completely meaningless. I think that systems that cannot be evaluated in the past and therefore cannot be said to be long term profitable with good certainty, are NOT worth buying. Given the nature of the risk to reward and the backtesting results I would be inclined to think that this expert will have similar results as the shark EA. I am however, opened to change my mind after we have two years of live testing data that show that the expert can be profitable, obviously with lower profits than the backtesting shows. Before this, using this expert is as good as gambling as one does not have ANY reliable historical testing that shows if this EA has any "adaptability" at all.
As I have said, for now and at least the first one or two years, I will consider this expert advisor NOT worth buying. If you would like to learn more about my criteria for long term profitability and how you too can trade with long term profitable systems that protect your equity and have controlled risk levels please consider buying my ebook on automated trading or subscribing to my weekly newsletter to receive updates and check the live and demo accounts I am running with several expert advisors. I hope you enjoyed the article !
First of all, the expert advisors web page is fairly decent with no huge amount of hype or unnecessary marketing. In fact, absolutely no claims are done about the expert's profitability, reason why I imagine the author(s) prefer the clients themselves to decide the profitability in virtue of the live/backtesting the people at 4xgreed show. There is no clear explanation of the nature of the expert's trading but they do say that the expert is able to adjust to market volatility and market conditions via neural networks and genetic algorithms. To me, the mentioning of these factors is just to make people feel the system is "advanced" as there are no proven results in live testing that have shown that a strategy can in fact benefit in the long term from any of these two approaches. In fact, neural networks have been proven to overfit expert advisors to the past, preventing effective adapation to "actual" market conditions.
The most interesting part of this expert comes from the testing section. The people at 4xgreed have done a good job at presenting several brokers, account sizes and backtesting results. This is a sure way to show an expert is both broker consistent and profitable. However, by analyzing the trading statements, several things come to mind.
The first bad characteristic of this EA is the low value of the TP. With a TP in the order of 14 pips (EUR/USD), this expert advisor will sure have problems with one minute fractal interpolation in a very significant manner. Furthermore, the fact that the risk to reward ratio is 65:14 is also not very conforting. This disparity between the stop loss size and the take profit size makes the errors in backtesting favor winning trades more than losing trades because of their much smaller size. This is evidently seen in the ten year backtesting that shows the typical, unrealistic returns when fractal interpolation errors are predominant. An almost 5:1 risk to reward ratio is also not a very favorable characteristic since it means that the expert is relying on the randomness of short term movements to gain profits, rather than reliable trend following or similar types of trades.
Live testing at the moment is rather limited so there is no room for big conclusions regarding them. The only thing we could say is that, as it is inferred from the above discussion, there is disparity between live and backtesting which makes backtesting completely meaningless. I think that systems that cannot be evaluated in the past and therefore cannot be said to be long term profitable with good certainty, are NOT worth buying. Given the nature of the risk to reward and the backtesting results I would be inclined to think that this expert will have similar results as the shark EA. I am however, opened to change my mind after we have two years of live testing data that show that the expert can be profitable, obviously with lower profits than the backtesting shows. Before this, using this expert is as good as gambling as one does not have ANY reliable historical testing that shows if this EA has any "adaptability" at all.
As I have said, for now and at least the first one or two years, I will consider this expert advisor NOT worth buying. If you would like to learn more about my criteria for long term profitability and how you too can trade with long term profitable systems that protect your equity and have controlled risk levels please consider buying my ebook on automated trading or subscribing to my weekly newsletter to receive updates and check the live and demo accounts I am running with several expert advisors. I hope you enjoyed the article !
Posts
No comments:
Post a Comment