Great EA in backtest! - page 92

 
xxDavidxSxx:
not too shabby.

yeah! i think so too

xxDavidxSxx:
I have fxdd real account and have compaired price feeds. They move identical as far as the eye can see. But the auto lot function will change from one to the other cause fxdd real don't do micro lots.01-.09

for now i don't really care about money management/lot sizes. bottom line are the pips gained (or lost).

i have two more demos running but with only 6 pairs. one is on a 1H timeframe, the other on 30M. same settings as the one above. except the trading hours are different. the results are not very good - both are down.

my real account with IBFX trades only the $JPY pair and is up 2 pips since monday

since i don't believe in backtesting too much only time will tell. anyways, i am going to keep those settings for at least another two weeks to get a better picture. i am not so sure about the blocked trading hours given on the CT website though.

AZBOfin

 
AZBOfin:

i am not so sure about the blocked trading hours given on the CT website though.

AZBOfin

in the upper left hand corner of chart is says good trading hour or bad trading hour. If it says good durring an hour you have set not to trade then change your settings untill it says bad trading hour.

 
xxDavidxSxx:
not too shabby. I have fxdd real account and have compaired price feeds. They move identical as far as the eye can see. But the auto lot function will change from one to the other cause fxdd real don't do micro lots.01-.09

hi David,

if you would like to trade micro lots with fxdd then you only need to write them an email and say that you want to have micro lots enabled in your platform.

look at this mail which i got back from fxdd:

"

Dear Sir,

Per your request, your account has been enabled to trade micro lots. Thank you for giving us the opportunity to earn your business.

Regards,

Roodmy Cesar

FXDD Sales/support

75 Park Place, 4th flr

New York, NY 10007

Tel: 212.791.3950

Fax: 212.937.3845

roodmy.cesar@fxdd.com

www.fxdd.com

""

now are micro lots enabled

Files:
fxdd.jpg  54 kb
 
xxDavidxSxx:
in the upper left hand corner of chart is says good trading hour or bad trading hour. If it says good durring an hour you have set not to trade then change your settings untill it says bad trading hour.

now i'm completely lost!

according to the code it can't say "good" trading hour during a blocked hour. it's a simple if/then statement.

(or do i misunderstand your statement?)

what i have meant in my previous post was that i don't know how realiable those blocked hours are given on the CT web site.

AZBOfin

 

Backtests

Aaragorn:
This thing is reallllllly active! It's on it's 16th position of the day. My tests show that reverseindex=8 seems to be best for the usdjpy pair on interbank. I'm starting to develop another theory about this.

I don't know how to explain it but it's sort of how this EA seems to get itself in gear. I see it take it's losses when it's started inside a retracement bounce, on a reversal. once it's gone thru two or three reversals it seems to get a better grip but until then it's basically gambling. It may be gambling the whole time but something improves after it's been running thru about it's third major trend reversal. I'm guessing about a day or two on this pair. Then it just gets better. Takes it oh I don't know maybe 40 positions or so. This is what I keep seeing over and over on the backtests. If I see it on the forward test this week it will be telltale for me.

My analysis says that it's making on average 12 trades a day in the backtester, that's average sometimes much less and sometimes much more....max of 44 and min of 1. That's a lot of activity!

If I can stand to watch it and keep my maxlots down for one more day if my theory is correct..about wednesday afternoon this thing should start to lift off and I could up my maxlots. I think I have to keep it supressed however that long, unless I see that its riding the trends and reversals ok before that.

Im so tired to tell that backtests always give great results- forward test give a half of the back test. When it comes to real account - then it sux. Im running 1.88f (fixed) version and i could double my 10k account in two weeks. On real account im in -$10 for 3 weeks test. The CT skipps half of trades. I have two monitors side by side - one demo, one real. Demo makes trades - Real just sits there. I will try to change settings in MetaTrader Station - may be boost it somehow to refresh more often.

 
Georgiy:
Im so tired to tell that backtests always give great results- forward test give a half of the back test. When it comes to real account - then it sux. Im running 1.88f (fixed) version and i could double my 10k account in two weeks. On real account im in -$10 for 3 weeks test. The CT skipps half of trades. I have two monitors side by side - one demo, one real. Demo makes trades - Real just sits there. I will try to change settings in MetaTrader Station - may be boost it somehow to refresh more often.

Why don't you just manually make the trades on your real account, based on the demo account?? Also maybe you should buy the pro version! That could be your problem...

my .02,

B

 
YupYup:
Why don't you just manually make the trades on your real account, based on the demo account?? Also maybe you should buy the pro version! That could be your problem...

my .02,

B

A person would have to be very glued to the monitor to do that. I have watched this EA trade in live forward account and I have also enabled the 'show' features in the EA and looked at the expert advisor tab to watch it think as it runs. Despite that information which still leaves me a little confused watching it... I can't tell when it's going to exit it's trades. I saw it sit there with profits to take and not take them watched it go against the position and then come back and give it more profits which it THEN took. I don't know how it decides to take profits and when. I just know that it is paying attention to probabilities etc. To actually mirror trades that this is making would take incredible commitment and diligence to monitor and react.

 

I don't think that the difference in trading behaviour between demo and real accounts is a CT "problem".

An EA follows mathematical calculations, and that is it. The code is the same for both, the demo and the real account. The only thing that's different is the data fed from the broker to the EA.

look at this... the EA calculates the following formula:

IF A+B=1.0000 THEN trade ELSE dont-trade

Data from a demo account: A=0.5000, B=0.5000

Data from a real account: A=0.5001, B=0.5000

guess which account is trading?

the only way of knowing if an EA is working for a particular broker is by putting it on a live account with real cash. backtests and demo trading can give one only an overview if it could work or if the EA is garbage from the begin with and even then it is not for 100% sure.

i for myself believe that CT has great potential in being a successful trading system. i am using backtests and demo accounts to tweak the settings.

but only my live account results will show if it really works. and here it doesn't matter if one start with a $200 micro or a 50K standard account.

live is live and demo is demo. two completely different worlds. on top of it, if the results are good with IBFX it doesn't mean that FXDD gives the same returns - because the broker's data are different (differently manipulated).

i also measure the success of an EA not in $$$ but rather in pips. -10$ for a standard account is nothing to cry about, but for a micro account it is a lot.

if this great EA can give me a 200 pips gain every single month with minimal drawdowns, i am more than happy. the rest is capitalization, leverage, lot-sizing and risk.

just my 2 pips

AZBOfin

 
Georgiy:
Im so tired to tell that backtests always give great results- forward test give a half of the back test. When it comes to real account - then it sux. Im running 1.88f (fixed) version and i could double my 10k account in two weeks. On real account im in -$10 for 3 weeks test. The CT skipps half of trades. I have two monitors side by side - one demo, one real. Demo makes trades - Real just sits there. I will try to change settings in MetaTrader Station - may be boost it somehow to refresh more often.

this is the version I'm running now.

these settings...

// ---- Global variables

extern bool ExitMarket = false;

extern bool ShowSuitablePeriod = false;

extern bool ShowMarketInfo = false;

extern bool ShowAccountStatus = false;

extern bool ShowStat = false;

extern bool ShowDecision = true;

extern bool ShowDirection = true;

extern bool BlockSell = false;

extern bool BlockBuy = false;

extern bool ShowLots = false;

extern bool BlockStopLoss = false;

extern bool DisableShadowStopLoss = true;

extern bool DisableExitSell = false;

extern bool DisableExitBuy = false;

extern bool EnableMACD = false;

extern bool EnableMA = false;

extern bool EnableFractals = false;

extern bool EnableCCI = false;

extern bool EnableCyberiaLogic = true;

extern bool EnableLogicTrading = true;

extern bool EnableADX = false;

extern bool EnablePivot = false; // Use Pivot_day as filter

extern bool BlockPipsator = true;

extern bool EnableMoneyTrain = false;

extern bool EnableReverseDetector = true;

extern double ReverseIndex = 8;

extern double MoneyTrainLevel = 4;

extern int MACDLevel = 10;

extern bool AutoLots = True;

extern double MAXLots = 0.05; // Max lots size on AutoLots--added by project1972

extern bool AutoDirection = True;

extern double ValuesPeriodCount = 7;

extern double ValuesPeriodCountMax = 7;

extern double SlipPage = 1; // Slippage of the rate

extern double Lots = 0.1; // Quantity of the lots

extern double StopLoss = 0;

extern double TakeProfit = 0;

extern double SymbolsCount = 2;

extern double Risk = 1.0;

extern double StopLossIndex = 2.5;

extern bool AutoStopLossIndex = true;

extern double StaticStopLoss = 15;

extern double StopLevel;

extern bool EnableTrailingStop = false; // Enable Dynamic Trailing Stop

extern double TrailingStopFactor = 1.0;

extern string TimeTradeHoursDisabled=""; // Example "00,01,02,03,04,05" GMT Alpari=10

extern int GMT=-1; // For North Finance GMT = 3, Alpari GMT = 1, IBFX GMT = -1 etc.

extern int MagicNumber=123000; // Magic Number -- change for every pair traded

// ----

int NoTradeHours1=25; // Time not trade

int NoTradeHours2=25; // Time not trade

int NoTradeHours3=25; // Time not trade

int NoTradeHours4=25; // Time not trade

int NoTradeHours5=25; // Time not trade

int NoTradeHours6=25; // Time not trade

bool SavedBlockSell;

bool SavedBlockBuy;

bool DisableSell = false;

bool DisableBuy = false;

bool ExitSell = false;

bool ExitBuy = false;

double Disperce = 0;

double DisperceMax = 0;

bool DisableSellPipsator = false;

bool DisableBuyPipsator = false;
 
Aaragorn:
this is the version I'm running now.

Thank you Aaragorn

I already put that on test today on demo. Risk of 1 is kinda brave move for live account.

Im trying to compare the data feeds for both accounts - demo and live. They are different for now. Ticks move differently and price update is different.

Something happens at 16 GMT for last two days - I got -$70 on my live account and -$1000's on 6 other accounts from different brokers. Were there news out? CT can not handle those for sure - traded opposite in all cases.