Martingale Mania V ~ Currency Pair
Filed Under (EA System, Journal) by Casey on 25-08-2009
Tags: Backtest, Best Forex Sniper, Martingale System
Following the concept of I pondered upon in my Martingale Mania I, I set out to look for the best currency pair. Since Best Forex Sniper uses a cascading/grid order limit of about 30 pips between each grid level. I had to find a currency pair that is not so volatile… IE does not move 330 pips in one shot without retracting back 30 pips to take profit….
So I went about to look for these currency pairs which were not volatile and put it to the tests… How else do we test? Well… by backtesting it of course. I must say backtesting Martingale EA is pretty long and tedious. It practically took 12 hours just to back test one currency pair between Jan 2009 till Aug 2009!!
Man… I had to learn the long way… Not only did I busted 12 hours of my time testing one currency pair… I wasted another 6 hours because I put the wrong settings… Sighhh….. Oh by the way… these backtesting were done on on Best Sniper Forex EA… So the criteria set out by me and the 330 pips movement were base on the parameters from this EA…
Anyway… I had to narrow down or short list the currency pair before doing any backtesting, so I just browse throught the chart of each currency pair and look for sharp trend upwards and downwards. If the sharp trends move more than 300 pips one way without retracting… well I guess I can call it quits on that currency pair.
Ok… not calling it quits but maybe I will test it out after I work on these few currency pairs. After much browsing… I found 5 currency pair.
USDCAD, EURCHF, CHFJPY, AUDUSD, NZDUSD.
Now these are the currency pair I short listed base on very rough calculations and observations. So let see if my selections were accurate.
(Click on the graph to pull out the statement)
——————————
USDCAD

——————————-
EURCHF

——————————-
CHFJPY

——————————-
AUDUSD

——————————-
NZDUSD

——————————-
Not bad… got roughly 4 out of 5 right…
Now what does all this means?
Firstly, I only took data from Jan 2009 till Aug 2009 and it bloody took me 4 days plus to complete backtest all these 5 currency pair. Man was it tedious!! But hey I had nothing much better to do… so I guess it was ok. Anyway… to further put this to the test of consistency… I need to retest these currency pair (except for CHFJPY) with the whole 2008 and see how consistent it is… Another week of testing to do…
Secondly, my idea that for Martingale strategy to work well and consistently seems correct. The price of any currency pair should not sharply move 300 pips without retracting back to at least 30 pips if we want to put Martingale strategy live. To show you what I mean… I also did a back tested EURGBP… and below are the results… Of course this theory is base on what I understand about Sniper EA
——————————
EURGBP

——————————-
Note: Sniper has a build in stop loss… which I think is pretty cool. So I set the stop loss to 330 (according to my Martingale calculations) for all the backtesting above.
I also tested USDCAD without the stop loss, and it was crazy… From the first graph above, during that final sharp downward line, it was on the 10th Martingale level, it just drift right into the price without doing anything until I lost all my margin. (Somehow or rather I could not find that back testing results which showed devastating result without the stop loss)
Gees…. scary… So this little piece of code here seems to be very handy indeed…
Thirdly… I do not think these EA developer put this idea into their thoughts and tests when they develop and recommend which currency pair to trade. They seem to recommend popular currency pair like GBPUSD, USDJPY to trade using Martingale strategy… But if my theory and backtesting have shown… These are probably the currency pair to avoid.
If one were to watch the money that they are risking than they would say trading Martingale on these popular pairs is suicide… I mean if my test of EURGBP was any indication to go by… imagine if we were to trade two or three popular currency together and am in the worst case situation when all two or three currency tank on your account….
Lastly… to sum this all up. If you want to trade Martingale… then I guess one have to be careful about which currency pair you want to trade. Sorry… I should say… if one were to trade Best Forex Sniper, one should be careful on the choice of currency pair they decide to trade.
With all other Martingale strategy EA, I guess the only way to see which currency pair suits the EA, backtesting is required.
Well I guess I have to get back to testing not only the above currency pair with 2008 data but also other non popular currency pairs and see how they fair…








You have to backtest at least on last two years of data.
On this TF the worst trend movements happened ever on most CCYs caused by the crisis.Try it out, especially on EUR/CHF
To be on the safe side u have to run the full 10 year backtest, no matter what.
Euro pairs backtest u can start on 1st of Jan. 2002, the date of introducing the Euro.
Before it was ECU, a basket currency. These converted datas are not trustworthy in my opinion.
I recently came across your blog and have been reading along. I thought I would leave my first comment. I don’t know what to say except that I have enjoyed reading. Nice blog. I will keep visiting this blog very often.
Patricia
http://forextradin-g.net
@onur… yes, i know that this 10 year backtesting needs to be done.. but this is really taking hours and hours and hours just to backtest one currency pair… slow and steady i guess. btw.. what is CCYs? So I guess it is safe to test it out since 2005 to 2009 no?
@Patricia.. thanks for visiting…
I have also built a new EA on the martingale system. However, I have added some new filters and other things of interest to prevent a blown account. if your trading a martingale syste,. email me and lets talk about the best way to manage our EAs. 12-19-11
Garry
New martingale called Zipper53