Wednesday, June 9, 2010

Playing with Bollinger Bands : A Likely Long Term Profitable Strategy

Last week I wrote a post about the Bollinger Bands indicator and how I thought it could be exploited to achieve good profitability levels in forex trading. After doing a full analysis going through a visual 5-6 year backtest, a mathematical expectancy analysis and the final coding and testing of the strategy I can tell you that I have come up with a likely long term profitable system based exclusively on Bollinger Bands. Asirikuy members would definitely want to checkout last week's video in which I explain the whole development process in detail showing you all the steps necessary to go from an idea to an actual, coded expert with great possibilities of long term success. On today's post I want to give my regular readers a look into this strategy and the results it finally achieves.

My first idea for coding this Bollinger Band system was taken from the thoughts I had a week ago regarding this strategy. I based the making of this strategy in the fact that when a bar closes above a certain number of volatility adjusted steps outside 2 standard deviations an important "signal" is given that predicts long term movements in that direction to a good extent. As I pointed last week, this strategy also signals a retracement in conjunction with the trend following aspect but definitely I wanted to explore the trend following aspects of the strategy first.

The mathematical expectancy analysis of this strategy wasn't disappointing at all with positive results for almost all period perspectives on the one hour charts. However - interestingly enough - long periods of time revealed a great characteristic of this system which is the predictive power over a singificantly long amount of time with almost no moves into unprofitable territory. This means that the strategy would definitely lend itself to the making of a system that could use an extremely favorable risk to reward ratio. Designing my system's money management around the conclusions of the mathematical expectancy analysis results I came up with a strategy that held a risk to reward ratio of exactly 1 : 4 meaning that the system can take four loses for each winning trade. For me this is quite unprecedented sine most strategies I coded have a risk to reward ratio oscillating from 2 : 1 to 1 : 3 but I had never been able to achieve such a high expectancy for winning trades with success. The results I achieved for a 10 year backtest (2000-2010) on the EUR/USD one hour chart are shown below.
-
-
I have to say that I was very impressed to see that such a simple strategy with absolutely no internal closing criteria besides the TP and SL could achieve such an incredible performance. This strategy definitely shows that it is not complexity what determines the success of a trading strategy but true understanding of the underlying price action. Once you have some idea of how the market behaves and how indicators can be used to exploit an inefficiency a likely long term profitable system from a very simple set of logic can arise. It is important here to say that the above shown results are unoptimized and simply the variable settngs are the result of the mathematical expectancy analysis (for those who are wondering, yes, I tried a trailing stop but a solid TP works much better)

Simulations of this system are also bound to be quite accurate since the average value of the TP is well above 200 pips, meaning that this EA has an incredibly large take profit value that could not be faked by any type of simulation interference. Moreover, the EA only enters trades based on last bar's close and therefore explicitely controls bar opening, something that is bound to make trading systems more reliable and back/live testing consistent. Of course, we would need to test this expert to find out but I believe that this system can be improved a lot more before actually releasing it as a Watukushay EA. It is however important to say that GREAT part of this straegy's success is based on its volatility adjusted money management that allows it to adjust to changes in market conditions. Using a fixed SL or TP leads to incredible loses showing how adaptive money management is VITAL to achieve profitability.
-
-
As you see on the above image, the success of the strategy is based in the fact that trades are taken almost always after a successful bollinger band breakout. However the fact that we are not focusing on the actual contraction/expansion of the bands but on the statistical meaning of a highly deviated price result makes us enter only meaningful breakouts while others that might have been entered on a contraction/expansion criteria (which is also hard to define) are avoided. The strategy does take a lot of loses (strategies with high reward to risk ratios are often very hard to trade psychologically because of this) however the great thing is that losing trades become evident after only a small move against us, allowing us to preserve the great 1:4 risk to reward ratio.

Currently I have created a forum post within the asirikuy community forum so that we can start improving this strategy and making it become a solid Asirikuy contribution. Of course, the system lacks any internal closing criteria and coming up with relevant ways to improve the system's closing of positions should be vital if this system is to be released for use in the future. If you would like to learn more about automated trading and how you too can come up with and design systems with long term profitability in mind 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 !

No comments:

Daniel Fernandez's Expert Author Email Alerts
Sign up to receive email alerts of Daniel Fernandez’s latest articles from EzineArticles.com!

Email Address: