PhiBase development and technical support teams have conducted various backtests. These results are presented in strategy test reports page and provide members with almost all the basic details required about the EA's historical performance. Many members have requested detailed analysis to be performed in-house. This will be in addition to studies/reviews conducted by independent third party experts.
The PhiBase team has also been tracking the live trade performance of Ray Scalper since its launch with V1 and after its update to then V1.9. Ray Scalper is being run by many PhiBase members who have provided valuable feedback and suggestions. Reviews of V1.0 by experts like Fernando, Dave and Omar have helped us in improving the strategy to make Ray Scalper to be one of the best expert advisors available commercially.
A detailed statistical analysis presented here is thorough, open and honest. The study also presents test results conducted with various parameter alterations and non official internal lab testing results to enable better understanding of the system. The aim of this review report is to enable all our members to make an educated decision on choosing to use Ray Scalper as part their trading portfolio.
The analysis has been conducted using the 13 Year Back Test with recommended parameters (Default Parameters, 4411 and No Exit Strategy) and Max Allocation: 8 (Medium Risk). Spread of 1.5 pips was used for conducting these tests. The minimum lot size and increment was 0.1 (mini lots) and start capital = US$ 10,000. This is equivalent of conducting the test using micro lots (0.01) on a $1000 start capital which is very practical for most users. The position sizing and money management will also be similar to most standard accounts offering minimum of 0.1 minilot size increments.
You can choose to run Ray Scalper using the recommended (4411) settings when the broker spread is low (anything less than 2 pips is good) and the EA is being run on a reliable internet connection (like on a VPS).
All the EA Parameters are listed below:
Advanced_User_Parameters="Read Manual Before Making Changes"
Restrict_Trade_Entry="Period Day-of-Week and Hour"
Prevent_Trades_Over_WeekEnd="Close Any Open Trade At Day-of-Week and Hour"
EA_COMMENT_PREFIX="Ray Scalper "
MyFxBook strategy tester output is shown below:
The logarithmic chart given below makes both drawdown and growth equally proportionate in all periods.
The following information derived from the test results may be of interest:
|Number of Years||13||Backtest over 10 or 12 year period is considered good|
|Compound Annual Growth Rate (CAGR)||44.85%||CAGR of GOOGLE Stock (2004-2012 Buy & Hold) is 28%|
|Average Annual Return (AAR)||42.69%||AAR above 24% is considered good|
|Maximum Drawdown (MaxDD)||25.5%||Acceptable Range is 10 to 30% depending on Risk Profile|
|ARR/MaxDD||1.67||Value above 1.0 considered good|
|Risk:Reward Ratio||0.47|| Average Loss is twice size of average win. |
System will need 2 gains to cover each loss.
|Success Rate||74%||EA will win atleast 3 out of 4 trades|
|Maximum Consecutive Winning Trades||23||Can increase equity by about 20%|
|Maximum Consecutive Winning Trades||7||Can cause a drawdown of about 12%|
|Average Consecutive Winning Trades||4||Average equity gain of about 4%|
|Average Consecutive Winning Trades||1||Average equity loss of about 2%|
|Average Number of Trades Per Month||23|| Averages about 250 trades per year.|
Averages about 5 trades per week.
The 100 trade moving average (approximately 6 month period) of the equity is plotted in red. It can be see that the equity manages to stay above this line. Dips below this line indicates above average drawdowns. We see above 8 occasions when the equity moves below this average line, but the recover from these dips is fast. The longest drawdown period observed on the gain chart is about 6 months.
Annual equity gains compared with returns of S&P 500 Buy & Hold strategy:
One of the main reasons for user of the system to get disillusioned within a short time is because of wrong expectation. The 5 year or 12 year backtests project the long term profitability of the EA. Although the tests show that the DD and losses are normal part of the strategy, the user in most cases are not able to handle similar drawdowns in live trading.
The figure below illustrates the above issue:
As can be seen it would be more practical to observe the trades on a year by year basis to understand and develop faith in the system. This will enable traders to handle normal drawdowns in live trading without fear and uncertainty.
The chart below show the number of days drawdown was in progress and the number of days required for the EA to recover from the drawdown. The red bars indicate DD period and the green directly above indicates the number of days it took to recover from the lowest Point back to breakeven. The total drawdown period is usually shown as the sum of these two periods. Most traders find the losing period more difficult to handle - So this chart shows the total DD period split in to losing and recovery periods.
As can be seen from the chart, over 90% of the drawdown periods on historical tests have been less than 50 days long. We see four occasions when the DD was between 50 to 100 days. The DD periods were above 200 days once in 2007.
The recovery from drawdowns is also seen to be reasonably quick, with equity gaining back lost ground within 50 days most of the time. From this chart, it should be clear that the EA's strategy will involve drawdown periods that may last between 1 to 2 months on an average. In some cases, the drawdowns may extend upto 8 months. DD phases are an unavoidable and an accepted part of any strategy. It is very important to continue running the EA since recovery will follow at any time.
The following chart shows the 13 year backtest returns split into a monthly gain chart.
The EA has more gaining months (109) compared to losing months (50). The average of gaining months is over 7.7% and that of all the losing months is 3.8%. We also see that a losing month is followed by a strong set of winning months. Very few consecutive losing months are seen, but must be expected in future also. The recovery from DD from such periods is seen to be quick. The maximum loss seen in a month is about 11% and the maximum gaining month brought in about 40%. Users will need to set the risk level based their level of comfort - A single losing month of 11% should not force the user to stop trading the EA. If you are not comfortable with such DD level or % loss in equity, the Max Allocation should be set to lower levels (6 or 5).
Profits are seen in almost all kinds of market conditions seen over the years - Ray Scalper V2 can be expected to perform reasonably well in future since it can handle most price actions well.
The following chart shows monthly cumulative gains seen in the 13 yeat backtest.
A Monte Carlo method is a technique that involves using random numbers and probability to solve problems. Monte Carlo simulation is particularly useful when you want to predict the overall outcome of a series of related events when you only know the statistical probability of the outcome of each component event. This study can be used to predict the worst case drawdown that may happen in future if the price action is no longer suitable for the trading strategy.
|MONTE CARLO RESULTS AT 95.00% CONFIDENCE|
|Max_Allocation (Money Management)||8|
|Average Winning Trade||$29.25|
|Average Losing Trade||-$63.42|
|Profit trades (% of total)||74.21%|
|Number of Monte Carlo Samples||100,000|
|Number of Trade Sample||3683|
|Absolute Worst Case DD at 100%||51.43%|
|Average of DD higher than 95% of cases||31.13%|
|Worst Case DD @ 95% confidence||26.95%|
|Average of All DDs||15.72%|
|Monte Carlo Simulated Worst Case Drawdown using Money Management (Allocation 8)||31%|
The Monte Carlo simulated DD for the worst 32% of the 100000 cases is presented in the chart below (Max Allocation : 8):
Monte Carlo Simulations for Worst Case Drawdown using Very Low, Low and Medium Risk Levels.
|MONTE CARLO RESULTS||Allocation = 5||Allocation = 8||Allocation = 10|| Conservative (MA 8) |
2411 + Exit Strategy
|Maximum DD on BackTest with Money Management||14.43%||22.36%||27.29%||18.74%|
|Monte Carlo Absolute Worst Case DD at 100%||30.13%||51.43%||62.37%||42.18%|
|Monte Carlo Average of DD higher than 95% of cases||17.99%||31.13%||38.88%||28.73|
|Monte Carlo Worst Case DD @ 95% confidence||15.47%||26.95%||33.81%||25.03%|
|Monte Carlo Average of All DDs||9.29%||15.72%||19.55%||14.65|
Monte Carlo Simulations give a statistical prediction of what the worst case drawdown one should expect before choosing the risk level with which to trade the EA.
|Max Allocation||Net Gain||Max DD||AAR/MaxDD||CAGR||Average Monthly Gain|
|5 (Low Risk)||4060%||14.43%||2.28||32.25%||2.4%|
|8 (Medium Risk)||35,514%||22.36%||2.49||55.77%||4.2%|
|10 (High Risk)||101,960%||27.29%||2.51||68.68%||5.2%|
| Conservative (MA 8) |
2411 + Exit Strategy
Ray Scalper v2 money management trades with lot size which depend on the amount of free margin available and also the current price of the pair. for a start capital of $1000 trading with micro lots, allocation of 8 will make the EA trade with sizes of 0.03 lots. Ideally this should be the minimum required capital for the EA's money management to be fully effective. However, the EA will still be able to trade within the normal medium risk level for start capital as low as $300.
When the start capital is less then $500, the money management is not effective and all trades will be traded with minimum allowed lot size (usually 0.01 micro lots). But if your broker's minimum allowed lot size is 0.05 lots, then the EA will be trading the account at a very very high level by using 0.05 lots instead of 0.01, which could blow out the account in case of a normal drawdown. At 0.05 lots the equity will move up/down by $5. A single stoploss about 80 pips will make the account lose $40 which make the equity fall by 8%. If there are consecutive losses of 10 trades the account will lose 80% of the equity or be close to blowing out. On a normally funded account the same loss will only drop the equity by about 2.5% - Even the consecutive losses of 10 trades will only make the equity drop down to its Max DD of 25% seen in the 12 year backtests.
If the EA is to be traded on accounts size lower than $3000 it is recommended to trade only on brokers which allow minimum lot size of 0.01 (micro lots). If the account size is less than $500, cent or nano account will be required to allow for proper money management by the EA.
This section presents the summary of the study conducted to compare performance potential of Version 2 vs version 1.9/earlier version. The complete implementation and improvement has been documented at http://rayscalper.phibase.com/V2_0vsV1_9.html
Strategy test result comparison between v2.0 and v1.9
(All tests conducted with spread 1.5 pips - capital $10000 & lot increment 0.1).
|Ray Scalper Version||Parameters||Net Gain||Max DD||Success %||Average: Gain/Loss||CAGR||Average Monthly Gain|
|V1.0 Pre Launch||MA=8 4411||8718%||33%||79%||0.34||41%||3.4%|
|V2.0 Conservative|| MA=8 2411 |
Ray Scalper V2.0 shows improvement when compared to the version 1.9. We can see improvement on almost all aspects. The latest version almost has better net gains with same maximum drawdown. The improvement in gains obviously reflects well on the CAGR and average monthly gains as well. Ray Scalper will work similar to V1.9 in the default-recommended mode (4411 and no exit strategy). Major difference would be seen when using the conservative settings which enables the exit strategy. Version 2.0 has many user control parameters which can be used to customize the EA's trading to suit your trading style.
The major changes to strategy are listed below:
Version 2.0 will interpret M15 and H1 channels (V1.9 had only H1 channels)
With Conservative settings, StopLoss will be moved to first level Support/resistance
Custom parameters to control EA activation/deactivation at specified time.
Parameter to force EA to close trades at specified time (avoids open trades over weekend)
Parameter to specify percentage of capital to be used by EA
Parameter to enable hidden TP, SL and Trailing SL
Visual display of trailing SL trigger Level
Display of EA's trade entry and exit levels from account history
improved licensing module
The equity curve of any trading strategy will never be smooth for the simple reason that no strategy can guarantee 100% success rate. Although the equity curve moves up and down, a good system profits by its ability to have an edge in having comparatively larger upward movements. Ray Scalper has a definite edge to being profitable this way.
Someone may start trading during one of the upmove's and record gains right from the start. Some others may have started when the equity curve was just about to or when it is moving down. In this case the user may go into drawdown initially. It would not be possible for judging the EA based on these initial set of gaining or losing trades.
Ray Scalper has a very good distribution of gaining phases. The losing phases are smaller than the gaining phases. It can be see that irrespective of the period of starting the EA, the account will not be in DD for long on a monthly basis. There is higher probability of getting started on a gaining phase than a losing streak due to the higher success rate of the EA.
If you find our refund policy to be deterrent in making you purchase decision, we recommend you to study the statistical analysis presented on this page. Some traders claim refund on seeing some losses in the account, while this would just be a normal drawdown and part-parcel of the strategy. Such refund policies actually cause more harm to traders by making them opt for a refund instead of continuing to run the EA. Purchase decisions are made without fully understanding the strategy and with expectation from the system. In most cases the trader loses money because they stop the EA the while in DD and miss the recovery.
We want to see our products work successfully on our member accounts. This is only possible when the user has good understanding and confidence in the product. We aim to increase your confidence in our products by providing technical reports, detailed analysis, factual information, transparency and good support. The refund policy is mostly a marketing gimmick to enable the prospective buyer to jump in without any delay. PhiBase will not follow such methods to sell our products - We will instead trust our development efforts, information and performance to convince you to take a well educated decision.