this update is long overdue… the past 2 weeks have been extremely busy.

since finding FX-Bandit, i’ve had the honour of meeting a couple of like-minded traders who were also very interested in what this Expert Advisor could do. it was clear that FX-Bandit had some very good logic built into it. however, after reading the manual and testing it over these 2 weeks, i’ve come to think that where FX-Bandit triumphs, therein also lies it’s weaknesses…

to understand what i mean, one must first understand the “mean reversion” theory. in simple terms, “mean reversion” means buy-and-hold-and-hope-for-the-best. to give it some credit, the “mean reversion” theory, when used with some other form of market analysis, could very well be a profitable strategy. the theory suggests that if one waits long enough, most trades will be able to exit with a profit. there are statistical studies around that justifies this line of thought. however, the MILLION DOLLAR question now is :

• how long should i wait?
• how far should i place my stops to allow this theory to work without risking too much?

at this point, we should now turn our attention to study how FX-Bandit works, as well as how a trader can configure it to make it behave differently…

[updated on 21 nov 2007, 7:30 pm singapore time]

wow! the post is barely half done and it’s already stirring up some interest :o) i had wanted to update sooner… but my girl’s health has been keeping everyone at home rather busy these few days.

ok… back to FX-Bandit. when i first got the demo, i immediately went straight for the manual. to my surprise, for something that’s being sold solely on eBay, i would say that the manual was quite well written. it explained rather clearly some concepts behind the design of the Expert Advisor… although some of them were not in line with my trading psychology.

one particular idea that did not resonate well with me was that the creator advocated that the market is usually volatile enough to close “95% of all trades” at their take-profit level over a course of 3 to 5 days. now.. this leads us to, in my opinion, one of the most controversial parameter in FX-Bandit : Trade_Duration

there was a series of emails between Bill Austin (creator of FX-Bandit) and myself which i initiated to find out how Trade_Duration affects the way the Expert Advisor trades, as well as it’s relation to the various STOP methods that are available in FX-Bandit. this was what i found out…

Trade_Duration is X number of days that FX-Bandit will hang on to a trade before the stops come into effect. in other words, during this X days, the various STOPS are not effective. IN OTHER WORDS, if you set Trade_Duration=4.0, the trades that FX-Bandit opens will NOT have a stop-loss during the first 4.0 days of the trade! sound?…. you decide.

users of this Expert Advisor should take note that FX-Bandit comes with some suggested settings. Trade_Duration in the suggested settings range from 4.0 to 7.0. please note that this can go against you as much as it can help you close a profitable trade. if you put aside the theory about “95% of all trades” will end in profit after X-days, you will realise that the RISK : REWARD ratio is going to lean towards 15 pips : WTF! and no… i’m not refering to The World Taekwondo Federation ;o)

best case scenario = +15 pips. worse case scenario = margin call or whatever the loss is at the end of the X days. if you look at the current FX-Bandit statement which is running with the default setting files, you’ll see that on 12 nov 2007, the USDJPY trade closed with a -384 pips loss, and the EURJPY with -653 pips. they were closed at exactly 4.0 days after the trade was entered.

to give FX-Bandit some credit, the consistency that this Expert Advisor generates winning trades is fantastic. however, the risk / reward aspects need to be studied carefully. good thing about the Expert Advisor is that many of it’s settings are user-configurable. if you are not happy with the 4-day Trade_Duration, you can always make it shorter, or even 0. however, this would mean that you will be using or testing this Expert Advisor with settings that the creator did not recommend. if you are not using the suggested settings from the creator, you are more or less like shooting in the dark… trying to optimise the Expert Advisor by playing around with the parameters yourself.

with the many settings that can be changed, optimising FX-Bandit with different settings will require a systematic approach that may need to cater to, potentially, a huge number of permutations.

with all Expert Advisors (especially the commercial ones), optimisation is the job of the seller/creator. in fact, optimisation should be a continual process where settings need to be adjusted regularly to react to the changing market conditions. it would be ideal if Expert Advisors had the A.I. (artificial intelligence) to adjust or optimise itself automatically… but sadly, this is something that only a rare breed of Expert Advisor coders can do… and if you know of any, please let me know, ok? :O)

so… that’s my take on FX-Bandit. i believe it’s recent performance gives a good indication of it’s strenghts and weaknesses. i would think that one will either love it or hate it. where do i stand? make a guess… :O)

will blog more about Steinitz and DoublePlay in a seperate post…. this one has gone on for too long :O)

ps : appreciate if you could leave a comment if you find this post interesting…. or objectionable :O)

caveat emptor… in a nutshell, it means “Let the buyers beware”

with more people becoming more and more savvy and knowledgeable about forex, automated trading systems and Expert Advisors, i hope one day, we will be able to say :

” Let the sellers beware!!! “

scam-reports are easily found nowadays with the help of the most powerful search engine in the galaxy - BackRub.  never heard of it?  Google it and you’ll know which one i’m talking about :O)

some tip and tricks…

to really get every bit of information anyone can ever get about a certain thing on the internet, you will first need a free BackRub email account.  once you have that, you can then configure a BackRub Alert that works 24×7… relentlessly scouring the internet for every trace of the information that you are looking for.  the best part of it all is that it searches the ENTIRE internet… looking at old and current contents, and (DIG THIS!!) ALL FUTURE contents that have YET to be published on the internet!!  how do they do it?… you’ve got to visit BackRub’s website to find out more.

 so… as an example… if you are interested to find out more about me… you can configure your BackRub Alert to monitor the keyword “ckowyong”.  from that point on, every past, present and future references to “ckowyong” on the internet will be tracked and sent to you in real-time (if you want it that way).  you’ll know my each and every move…. when i make a post on my blog, when i create a reply on a public forum (any forum!!), which online community i belong to… etc.  after a week, you’ll probably know me better than my wife does :O)

so… if you want find out about everything that has been said about an expert advisor, you could set up a BackRub Alert to check for the name of the expert advisor or it’s website URL.  after a couple of days, you will probably have enough information to know if it’s a scam or not.

would appreciate if you can leave a comment if you find this post helpful.  a coffee would be preferred though :O)

ps : i’m sorry that some parts of the blog has been a little out-dated.  i’ll get everything update it in a couple of days once i return from a short trip. 

FX-Bandit v1.31(3rd run) was doing great… until 2007.11.05 22:05 GMT. all 3 open trades closed with a loss. oh well… maybe it was meant to be. but shockingly… at 2007.11.06 12:00 GMT, it happened again! the GBPUSD open trade got closed with a -31 pips loss! both events coincided with the restart of my MT4 platform… so i suspected something was wrong.

i contacted Bill Austin immediately and found out that he was already aware of the situation, and was already knee-deep into investigating the issue. he suspected that it’s got to do with the DST time change. in case you did not know, DST ended on Sunday, 4 Nov.

about 1 hour after reporting the ‘bug’, Bill came back… honestly acknowledging that there was indeed a bug in FX-Bandit that caused the trades to close once the MT4 platform was restarted. it was somehow indirectly linked to the DST change that happened recently. he added that he’s already working on the fix, and have placed this as a top priority so that he can send the fixed version out to his customers and trial users quickly.

a in v1.3… and now a in v1.31… makes one wonder how many more will surface. then again, RobinHood had a bug in v4.8, and DoublePlay was patched to v3.5.2 to correct how risk was being calculated. looking at this holistically… and being a programmer myself, i must say that discovering this now is better than finding out when it’s trading with real money! bugs are very common in the programming world… and the more complex a system is, the more likely that it will contain more bugs. with that knowledge, i probed Bill tactfully over Skype and managed to squeeze out some information about FX-Bandit. i found out that v1.31 is actually quite a simple Expert Advisor that’s made up of less than 200 lines of codes! frankly, i think simple is good. at least you know that there’s not many corners where a bug can hide :O) it’s not about how many lines or how complex an Expert Advisor is. bottom line is… it must perform in forward-testing!

will i trade it LIVE when Bill sends me the fixed version? likely not :O) i may wait until v2.0 or until the forward-tests show good results over a full month. i’m looking forward to receiving the new version so we can reset all this mess and start with a clean slate. despite the bug, i think we must give Bill some credit for being so responsive to the recent events… a quality that i look for in an Expert Advisor creator/seller. :O)

please post a comment if you have found this post useful… tks :O)

nope… this is not another new Expert Advisor that we are going to forward-test here. 

it’s been a full 3 months now since starting this personal quest of finding the best Expert Advisor.  the journey has been very educational and rewarding at the same time.  i’ve learnt so much about forex and expert advisors that i think it would have taken me at least a year to get where i am now.  a big part of this is attributed to the many astute followers of the blog whom i’ve had the honour and pleasure in communicating with.

the past months have been an accellerated learning process for me… constantly analysing charts, scrutinising forward-test statements, sifting through forums, and reading/replying to many mind-stimulating emails that so many have sent my way.  with so much happening… i think it’s appropriate to take a step back and have a look at where i am with my quest of ‘finding the best expert advisor‘.  consider it a ‘reality check‘ if you will.

my first experience with Expert Advisors started with RobinHood.  after LIVE trading with it for 3 good months, i stopped using it when it registered a few consecutive drawdowns.  it hit me that the scalper’s 6:140 reward:risk ratio was not what i wanted in my trading.  i then came across DoublePlay… an excellent performer that doubled a demo account in forward-testing within a month (see review sections on the right for more details).  i started trading LIVE with DoublePlay on 1 oct 2007.  however, since i started LIVE, DoublePlay has been hit by a string of losses.  so much so, i’ve stopped the Expert Advisor and have been sitting on the sidelines since 17 oct 2007.

throughout these months, i’ve been eating and breathing Forex and Expert Advisors.  i hope you believe me that i’m not exaggerating…  searching, forward-testing, analysing and reviewing Expert Advisors is hard work.  running this blog and replying to the hundreds of emails that i’ve received since the start of the blog has also kept me very busy.

so… with so much said and done, you may ask… ‘have you finally found the BEST Expert Advisor?’   sadly, the answer is NO.  but after all the hard work, i’m happy to be able to say that i have found 2 that i really like : DoublePlay and FX-Bandit.

DoublePlay : this Expert Advisor is still my favourite, despite the fact the i’ve lost money with it when i was trading it LIVE from 1 to 17 oct 2007.  i like it because it’s easy to understand.  it also has a fixed stop-loss and a rather effective trailing stop which worked like a marvel in the past.  however, DoublePlay has been hitting losses after losses recently.  why am i not worried?… because i see this as an OPPORTUNITY!!  an excellent opportunity for Hal Chapman to decipher the data that has been collected while it was making those losses and use the results to evolve DoublePlay into something better.  with DoublePlay v4 already in the pipeline, i am really looking forward to the good stuff that Hal Chapman will be putting into this new version.  if you have read the recent interview with Mr Matthew Brown (currently 4th place in the 2007 Automated Trading Championship), you will find that there’s a lot of wisdom in his words.  i quote :

Matthew thinks that “all EAs, especially the good ones, need to
be optimized and updated from time to time.  If you learn the
code and optimize it yourself, then you can keep it up to date
and continually make money”.

his every word resonates strongly with what i believe in too!… and currently, DoublePlay is going through an update that would make it better than before.

FX-Bandit :  this Expert Advisor shows good potential.  the strategy is sound, and it capitalises on the Europe and London market opening hours to make it’s killing.  it’s typical to see this baby close a few winning trades within a 6-hour window.   the Expert Advisor also has a few neat methods of stop-loss that can be customised by the user.  however, the creator of FX-Bandit designed the Expert Advisor based on the ‘mean reversion‘ theory, in which prices are expected to return eventually to the mean or average price (in this case, the ‘entry’ price).  hence, when using FX-Bandit straight out of the box, the default setting file may be configured such a way that there will not be a stop-loss within X number of days after a new trade is opened.  this is in line with the ‘mean reversion’ theory where it believes that each trade will come back into profit if one waits long enough.  my concern with this is the potential losses that open trades may incur… since they do not have an initial stop-loss at the beginning.  but on a positive note, these stop-loss settings are easily configured and are very flexible.  the trick now is to find an optimised setting for this Expert Advisor that would reduce the risks and improve it’s profitability.

there!  these are 2 of the better Expert Advisors i have seen so far.  you can see a ‘running’ review of them on the right panel.  i’ll try to update them as often as i can (they are a bit out-dated now), and will continue our search for the ‘Best Forex Expert Advisor’.

to our success!! - ck

———————————————————————
…at this point, i would like to urge viewers to join my mailing list.  how would you benefit from it?… you can find out more here >> http://ckowyong.com/blog/?page_id=31

for the record, we have been forward-testing the following Expert Advisors :

1. DoublePlay v3.5.2
2. PipBoxer v2.0.6
3. FX-Bandit v1.31 (time-based trial)
4. Steinitz HAS MTF v2.31
5. PipForia’s Hedge Grid (time-based trial)
6. RobinHood v4.9 (stopped) 
7. EuroX2 (stopped)
8. and a few others which are being privately tested