Cutting Through The Noise To Find The Most Profitable Trading Strategies. Forget the Hype - Read It Here First

Tuesday, September 27, 2011

Forex Predict Indicator Review

This is not a traditional Expert Advisors review but what it potentially offers caught the eye. Forex Predict Indicator costs $27 and offers a 60-day guarantee. Thankfully, their promotional material is light on pomp and their screenshots tell the story.

After you purchase the software you are given a download link for a zip file. The Zip file has a readme file to help you install the code. Copy the necessary files into your Indicators folder and reload MetaTrader. The new indicators will appear in your Custom Indicators folder.

In the Readme file the authors recommend using the predictor on a M15 and M30 timeframe. The indicator can be applied to all pairs.

The authors suggesting trading with a Target Profit of between 15 and 25 pips when the indicator predicts either an up (Long) or down (Short) phase. Beyond this it offers no more advice, such as a stop to use. In their examples they don't use a stop at all. 

The Readme file offers additional links for Expert Advisors and Indicators should you be interested in more.

After you apply the indicator to a chart you need to ensure the MetaTrader Shiftbutton is selected as you need to see the prediction on the chart. If you use the default chart view the predictor will lie flush against the price axis, obscuring it. 


The indicator looks neat and there is no ambiguity as to what it's trying to predict.This is the EURUSD M30 chart for September 27th.


I dropped the timeframe to M1 and tracked its behaviour in this short video. On the addition of each new time bar the prediction updates. You can see how the prediction evolved on each new bar, although placing stops in line with the prediction becomes difficult and the chance for whipsaw appears quite high (but again, this is a M1 prediction which suffers more noise than their suggested M15 or M30 timeframe).



However, the indicator is not without (some) merit. I ran the predictor across all timeframes for EURUSD and got a more consistent level of prediction.


So while the M1 predictor was favouring the upside, the M5, M15, M30 and H1 were indicating downward prices. But there are additional quirks. If you look at the early prediction for M30 the suggestion is for higher prices, but the M5 and M15 both clearly direct to lower prices for this period. Likewise, H4 suggests higher prices in the near term but the H1 is pointing lower. The daily prediction is higher but the weekly timeframe is suggesting more range bound trading while the monthly timeframe is pointing sharply lower.

When using this tool, it may be prudent to frame any trade decision based on the next longer timeframe. For example, if trading in M1 then consider the trend direction of M5 and/or M15; maybe use a higher prediction on the M1 to enter counter trades based on the direction of M5/M15; in the above example you would look to short upward predictions in M1 in line with the downward projections derived from the M5 and M15 timeframe.

If using the projection as intended it may be necessary to use an ATR or some other volatility-based stop to protect against missed projections. Unfortunately, the authors don't use stops so testing win % isn't possible without delving into this further (a later article perhaps).



However, if we are to give the Predictor credence then based on the monthly (MN) prediction for EURUSD, the currency should be trading down around $1.17 by June 2012. If this proves to be true then you can say you saw it here first, thanks to Forex Predict Indicator. 

Going back to our earlier M1 chart, one can see how the more dominant long-term trend overwhelmed the shorter M1 prediction. If you haven't watched the video (there is no sound), watch it - then compare it to the chart below.


If you would like to give it a spin, then you can buy it here for $27.

Happy Predicting!

Update:  The M30 chart prediction didn't turn out to bad; chart below:


Our Current Strategy Leader is:
 

Wednesday, September 14, 2011

FreeFXTradingRobot Review

This week's review is for a system which you can obtain for free here. The free version covers EURUSD but they have a full version of the system which covers USDCHF, EURCHF and USDCAD. For many of you, the free version will probably be enough.  The paid version retails for $77 and I suspect when you follow the link there will still be "14 copies" left (if not more!).

The free Robot is very much bare bones - there is no manual or installation instructions. You get an email with the .ex4 file attached and you have to drop the file into your Metatrader experts folder. You may then need to exit and reload Metatrader before the robot is listed.

This strategy advertises itself as a low drawdown system. The Relative Drawdown of their EURUSD example is 15.7%, but in my tests it's actually lower.


One issue of concern is the StopLoss. It's set at a staggering 800 pips, with a TakeProfit of 70. Our tests haven't come across an environment when this was an issue, but I wouldn't want to be holding a position when this proved to be the case.


In their defense, they do default to a cautious 0.01 Lot size and Optimization is selected as default (our tests do not use Optimization).

Comparative Test Period

As before, we use our three test periods to compare performance with other FX Robots. We default to using no Optimization, 0.1 Lots, $10,000 starting capital, trading Long&Short in EURUSD on a 15-minute timeframe.

The time frames are:

March 31st 2000 - 2001
June 29th 2003 - 2004
May 17th 2001 - 2002

Comparative Results

The initial results were promising. The Robot enjoyed a very high win percentage which offered a sizable Profit Factor. The Relative Drawdown was also very attractive.


The cash:equity curve show the steady returns


Compared to our current leader, Forex MegaDroid, how did it perform?



Comparisons were very favourable (particularly for a free system). In terms of Net Profit it only offered 28-81% of Forex MegaDroid returns, so the robot didn't bank as much. But the returns offered came with a considerable drop in Relative Drawdown - well over half the drawdown of Forex MegaDroid. The Win Percentage was also very high at over 99% winning trades.

Random Test Period

The robot was also tested against three randomly selected 12-month periods, between 2000 and 2010. These periods were:

January 4th 2005 - 2006
July 12th 2007 - 2008
January 25th 2000 - 2001

Random Test Results


As with Forex MegaDroid, the performance of the robot was not as high in more recent years. This may reflect strategies which were designed optimized for the early part of the decade (or before) and subsequently don't perform as well in the 'post-publish' years (although I have no idea when this strategy or others were first published).  The maxim "past performance may not reflect future performance" rings true for any of these robots.

Despite this, the robot did return a small profit and Relative Drawdown and Loss per Trade was not horrific. The cash:equity curves looked like this:


For the July 12 2007 - 2008 period I ran the strategy again, this time using Optimization. Unfortunately, the end result dropped the profit from $601 to $574 and increased Relative Drawdown by 0.02%.

Conclusion

For a free system, this is not a bad place to start. The robot has a high win percentage because of the weighting towards quick profits and a massive stop loss - although during the backtest it did adjust the stop loss, so getting hit for 800 is unlikely but the fact that it's there means it can happen.

For now, it holds the second best strategy ranking - but it's still early days.

If you are looking to trade other currencies beside EURUSD you will need to shell out $77 for the full version.

The creators offer a number of other products (none of which I have purchased or tested), including "50 Forex Trading Secrets", a Currency "Masters" club, and a Forex Trading Apprentice - which is apparently closed to the Public(!), and a second Forex Trading Apprentice - which is open to the public - but has annoying 'Do you want to leave this page' popups (you have been warned!).

Our Current Strategy Leader is:
 

Thursday, September 8, 2011

Forex MegaDroid Review

The next review is for Forex MegaDroid Robot. Unusual in it's promotion is the pagination on the splash page. Much as I detest the super long marketing pages, this one had the added displeasure of having to click through five pages of content before you got to the final purchase page. I don't know what it is about large fonts, primary colours and award ribbons which make people want to purchase these products (presumably this is why the creators use them???) - it should be results and results only which matter!

For the purpose of this review (and others on this site) I will take the plunge by purchasing their software to see if there is anything of merit to this system. This way, you don't have too. 

On the positive front, this is one of the cheaper systems at $67. So, what do you get for your money?

After purchasing you are taken to a download page. On the download page they make the case for FinFx brokers - should you be looking for a broker, although the MetaTrader installer will recommend a broker for you too. Next on the download page is an additional Robin Hood Mechanical FX System available for free as part of the MegaDroid package (presumably, it's not as good as MegaDroid...). I may test this system in my next post. Below the Robin Hood system are links for the MegaDroid installer and manual. 

Installation was painless with the .exe file. After restarting MetaTrader the MegaDroid system appeared in my list of Expert Advisors.

The manual shows a number of revisions (to the manual) dating back to March 2009 when v1.00 was released. The most recent change was August 2010 on version 1.30, and it looks like MegaDroid is on version 1.35. So it appears there is an active upgrade (improvement?) process for the software which can't be bad. This is a review for v1.35.

The manual has a lengthy Disclaimer and Risk Warnings. In their Forward they talk about measures brokers make to thwart FX robot strategies (in general). I don't know if this is to protect against claims Forex MegaDroid not working or if this genuinely the case - I have no idea.

The next section of the manual outlines the Getting Started process. MegaDroid requires MetaTrader 4 (but presumably will work with MT5 too???). The first part of the manual details installation using FinFX. Next is set up of the Expert Advisors tab which should be set up as follows:


Next the manual outlines a manual install for the Forex MegaDroid files, although the installer handled this process without an issue. However, later updates to the software may require these steps to be completed.

They recommend not using the robot close to major holidays or after the weekend when volatility can be high (good advice).

The robot trades the EURUSD.

When the Expert Advisor is dragged onto the chart the Common tab should be set like this


The manual also goes into detailed explanations of the various Inputs and how to activate them.

Of interest were some of the input features I had not seen offered by the two other robots reviewed here including:

"Stealth": When enabled (default), broker is not informed of true target and stop loss levels. The robot handles them instead.

"Aggressive": When enabled (default), the MegaDroid robot trades two strategies (one conservative and one aggressive) simultaneously.

"NFA": (disabled by default). Trades according to rules imposed by the NFA and should only be enabled on consulatation with your broker.

"S1/S2 Reference": Another work around against brokers targeting robots (these guys seem to take this seriously - Again, I was never aware of this as an issue).

"RemoteSafetyMode" Protects against opening trades during "potentially dangerous periods". Signalling looks to be handled by Forex MegaDroid and apparently "happens very rarely".

"RecoveryMode" Doubles down on losses (enabled by default). This is one I would prefer - and recommend - to keep disabled. If consecutive losses are rare by the system I would prefer to not tempt fate by doubling down if they occurred.

The manual also offers a good explanation on Risk Level and how to set up relevant to the size of your account. It also offers an extensive guide on strategy backtesting.

The only real miss on the Manual was a reminder to input your ClickBank Order number into the Inputs tab to validate your copy!!!

Upon starting the strategy on live this is the information view for MegaDroid with Alpari



Now for the test.

Expert Properties Setup 

Starting account is $10,000 on 0.1 Lots on a 15-minute timeframe. The strategy will trade Long&Short and will not use Optimization in line with other comparisons. Because I am using default settings I have left RecoveryMode enabled - although it's a setting I would not be comfortable with (especially if you were caught on the wrong side of the recent EURCHF kerfuffle peg doubled down, trebled down or worse...).

Comparative Test Period

These were three 12-month periods selected randomly for the period January 1st 2000 to January 1st 2010. The three test periods common across all of my robot tests were:

March 31 2000 - 2001
June 29th 2003 - 2004
May 17th 2001 - 2002

Comparative Results

Although it's early days in the search for the most effective (viable) forex robot strategy the initial prognosis for Forex MegaDroid Robot is quite good. Like many of the commercial robots available it uses a loose stop to boost win percentages, but this can be managed by trading a smaller lot size. Where it proves itself is generating consistent profits.




The Cash-Equity curves for the three, 12-month periods are relatively stable. Only 2000-01 shows a slow start before the strategy really got going.

 

Disabling RecoveryMode (i.e. the double-down on losses feature which I would avoid using where possible) generated the following results.



The percentage values give the relative change in performance between RecoveryMode switched on and off.

The robot still performs well despite the sharp loss in net profit, down over 30%. This is offset by the 29% drop in Drawdown and 26% drop in average loss per trade. For a starting account of $10,000 these are still great results over 12 months.

Disabling RecoveryMode and Aggressive settings gave the following result:


The percentage values in this case are relative to RecoveryMode off. Go figure why a less aggressive strategy should show an increase in Loss per trade for two of the three look back periods!

While unlikely to be statistically valid, there was a 6% increase in Net Profit and a 1% loss in Relative Drawdown, when Aggressive and RecoveryMode were switched off.

Random Test Period

Another three, 12-month periods were selected, using default settings - so Aggressive and RecoveryMode were enabled. The three periods were:

November 13th 2006 - 2007
July 24th 2009 - 2010
February 3rd 2005 - 2006

Random Test Results



The results were a little more mixed. The 2009-2010 period was particularly disappointing given it experienced 5 consecutive losses, with the RecoveryMode doubling down factor considered, increased the loss per trade to $334. Disabling RecoveryMode reduced the loss per trade to $185 and increased the profit factor to 1.8 for a Total Net Profit of $2,083 for this period.  Drawdown also dropped dramatically to a far more tolerable 12.39%.

The cash-equity curves for the robot were as follows:


Conclusion

This is only the third Forex robot I have looked at and the initial prognosis is good. It has a number of benefits compared to its competitors:

[1] It's cheap; $67 for a piece of trading software is below the typical $150 spot of others
[2] It doesn't overtrade. Based on the aforementioned tests it generates 100 - 200 trades on a 15-minute EURUSD time frame over 12 months. On average, you are probably looking at 1-2 trades ever couple of days.
[3] It can be configured to be less risky. Yes, profits are reduced - but so is the risk of ruin. Profits in reduced risk form are still very good. I strongly recommend disabling RecoveryMode.

Maybe the limits of Forex MegaDroid Robot success in real-time are tied to broker "tricks" as the creators allude too??? I haven't tried trading it live so real results may vary. But until I come across something better Forex MegaDroid Robot is a good place to start.