You may have seen on most of my automated trading system reviews and my system design blog posts that I usually put a lot of emphasis on the risk to reward ratio of trading strategies. The average risk to reward ratio which is simply a number comparing the average loss with the average profitable trade of a trading strategy gives us some very important information about a trading strategy. Perhaps some of the most important aspects of the risk to reward ratio are overlooked by most traders which usually leads to the comfortable use of systems with extremely unfavorable risk to reward ratios. Through this post I will try to explain the many problems of using a system with a very bad risk to reward ratio and why you should get away from this type of trading strategies.
But, what is a bad risk to reward ratio ? I usually define this as a ratio which is very tilted towards the risk side. Trading systems which risk much more in average than what they make per trade are said to have a bad risk to reward ratio. For example, a system which makes 10 USD per trade in average but loses 100 USD on every loss has a very bad risk to reward ratio of 10:1, that is, the average risk is ten times bigger than the average reward.
So what is the problem ? Many people think that a strategy can be successful, even if the risk to reward ratio is terribly bad, if there is simply a very high winning percentage. Of course, a strategy with a risk to reward ratio of 10:1 requires to win at least 95% of trades to be successful, any further loses will cause the strategy to cause overall loses. However it is true that mathematically a strategy could in theory be profitable if it has a bad risk to reward ratio and a very high winning percentage.
However, the problems start to come when we think about the estimation of the winning percentage. How can we reliable calculate the winning percentage of a trading strategy in a market which has no centralized exchange and which is bound to show different trading results on every broker ? Is there any reason to believe that the maximum number of consecutive loses in the past is the worst possible case ?
The fact is that the winning percentage cannot be accurately determined for any given trading system and the fact is that a difference of even +/- 15% may as well be possible not only due to broker differences but because of other aspects of the market like re-quotes and spread widening which are not shown within simulations or paper trading accounts. This is talking about the historically estimated winning percentage. However the future winning percentage may as well be within +/- 20% of the estimated historical value, given the fact that future market conditions are bound to show us a new worst-case scenario, something which usually involves a decrease in the winning percentage.
Now add to this the fact that most trading systems with unfavorable risk to reward ratios use very small take profits and you get an incredible augmentation of live vs simulation and paper trading results which may account for a great overestimation of the winning percentage.
When it comes to live trading, it is obvious that systems with favorable risk to reward ratios hold a much better chance at facing more unfavorable market conditions since an increase in the number of consecutive loses is bound to only cause a modest increase in the maximum draw down. On the contrary, systems with unfavorable risk to reward ratios, will show a very BIG difference in profitability with only a very small increase in their number of consecutive loses (or closely located loses for that matter).
This is the main reason why I do not develop or trade systems with risk to reward ratios higher than 2:1. It is evident to me that these systems have a smaller chance at surviving to future changes in market conditions and a bigger chance of having an overestimated profitability due to the inaccuracy of the determination of the winning percentage and the number of consecutive loses. through paper trading or simulations. The fact that these systems are INCREDIBLY vulnerable to even a small increase in the number of consecutive loses (which may easily be caused by a mere change of broker or the introduction of spread widening and re-quotes) makes the simulations of their performance very lacking at best.
If you would like to learn more about the systems I have developed and how you too can design and trade your own systems with favorable risk to reward ratios 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 !
But, what is a bad risk to reward ratio ? I usually define this as a ratio which is very tilted towards the risk side. Trading systems which risk much more in average than what they make per trade are said to have a bad risk to reward ratio. For example, a system which makes 10 USD per trade in average but loses 100 USD on every loss has a very bad risk to reward ratio of 10:1, that is, the average risk is ten times bigger than the average reward.
So what is the problem ? Many people think that a strategy can be successful, even if the risk to reward ratio is terribly bad, if there is simply a very high winning percentage. Of course, a strategy with a risk to reward ratio of 10:1 requires to win at least 95% of trades to be successful, any further loses will cause the strategy to cause overall loses. However it is true that mathematically a strategy could in theory be profitable if it has a bad risk to reward ratio and a very high winning percentage.
However, the problems start to come when we think about the estimation of the winning percentage. How can we reliable calculate the winning percentage of a trading strategy in a market which has no centralized exchange and which is bound to show different trading results on every broker ? Is there any reason to believe that the maximum number of consecutive loses in the past is the worst possible case ?
The fact is that the winning percentage cannot be accurately determined for any given trading system and the fact is that a difference of even +/- 15% may as well be possible not only due to broker differences but because of other aspects of the market like re-quotes and spread widening which are not shown within simulations or paper trading accounts. This is talking about the historically estimated winning percentage. However the future winning percentage may as well be within +/- 20% of the estimated historical value, given the fact that future market conditions are bound to show us a new worst-case scenario, something which usually involves a decrease in the winning percentage.
Now add to this the fact that most trading systems with unfavorable risk to reward ratios use very small take profits and you get an incredible augmentation of live vs simulation and paper trading results which may account for a great overestimation of the winning percentage.
When it comes to live trading, it is obvious that systems with favorable risk to reward ratios hold a much better chance at facing more unfavorable market conditions since an increase in the number of consecutive loses is bound to only cause a modest increase in the maximum draw down. On the contrary, systems with unfavorable risk to reward ratios, will show a very BIG difference in profitability with only a very small increase in their number of consecutive loses (or closely located loses for that matter).
This is the main reason why I do not develop or trade systems with risk to reward ratios higher than 2:1. It is evident to me that these systems have a smaller chance at surviving to future changes in market conditions and a bigger chance of having an overestimated profitability due to the inaccuracy of the determination of the winning percentage and the number of consecutive loses. through paper trading or simulations. The fact that these systems are INCREDIBLY vulnerable to even a small increase in the number of consecutive loses (which may easily be caused by a mere change of broker or the introduction of spread widening and re-quotes) makes the simulations of their performance very lacking at best.
If you would like to learn more about the systems I have developed and how you too can design and trade your own systems with favorable risk to reward ratios 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
No comments:
Post a Comment