Follow the BIG Candle

OK. I smoothed the ATR over 7 different periods based on the period used in the setup, 3 quarters below the given period and 3 quarters above, plus the original period as follows:

Suppose the user set ATR Period to 20. The system will generate ATR for the periods 5,10,15,20,25,30,35 and take the average. For ATR Period of 60, periods will be 15,30,45,60,75,90,105. and so on.

Results were better and more consistent than previous results without smoothing.

H1 Time Frame

ATR Period _____ Trades _____ Profit _____ Drawdown%
___________________________________ _________________
10 ____________ 72 _________ 723 _______ 18.35%
20 ____________ 52 _________ 599 _______ 16.64%
30 ____________ 43 _________ 1,745 _____ 15.78%
40 ____________ 53 _________ 3,672 _____ 11.82%
50 ____________ 43 _________ 3,961 _____ 14.71%
60 ____________ 48 _________ 3,711 _____ 13.97%
70 ____________ 47 _________ 2,592 _____ 16.44%
80 ____________ 40 _________ 2,472 _____ 16.74%
90 ____________ 44 _________ 3,586 _____ 16.69%
100 ___________ 39 _________ 3,472 _____ 17.01%


H4 Time Frame

ATR Period _____ Trades _____ Profit _____ Drawdown%
___________________________________ _________________
10 ____________ 14 ___________ 3,979 _____ 14.74%
20 ____________ 10 ___________ 3,371 _____ 15.52%
30 ____________ 10 ___________ 3,346 _____ 15.58%
40 ____________ 13 ___________ 2,188 _____ 15.35%
50 ____________ 9 ____________ 3,100 _____ 18.64%
60 ____________ 9 ____________ 3,050 _____ 18.52%
70 ____________ 10 ___________ 3,375 _____ 16.07%
80 ____________ 10 ___________ 3,397 _____ 15.95%
90 ____________ 8 ____________ 2,306 _____ 15.36%
100 ___________ 10 ___________ 3,403 _____ 16.14%

As before starting balance is 10,000 and period Jan 1,2014 to Mar 1,2014.

I am sorry, for the H1 time frame the Closed_on_Opposite_Signal was set to true, hence the larger number of trades.

I noticed that closing on opposite signal for shorter time frames generates better results, as trend on short time frames don't last long.

Was this a waste of time ?. I don't see any real contributions here.

Maybe it's better to continue this on my own.

In case I don't see you, Good Morning, Good Evening and Good Night !

sorry about lack of contribution have been following the thread passively like many others im sure, the system looks to have a lot of potential please keep us updated

Good work there abokwaik! I like your idea.. Keep us posted! It will be good to see this put into demo testing

Interesting thread, will follow this and try to contribute the best I can.
Looking forward to see some demo/live results!

Hi
Your advisor is very good, thank you good job
have you tried to improve money management, I think it will be better. I can do it and provide you some versions, if you send me the program

OK. Lets give it one more shot.


I have created a Trade Explorer with EA new version (4) and attached it to the thread.

The New version along with parameters used in the explorer are attached.

Pair used will be GBP/USD Time Frame : M30.

Version Updates:

1. Smoother calculation of ATR
2. Position sizing as risk percentage (bug fixed)
3. Trail Stop Jump internally set for both 4 and 5-digit brokers.
4. Using of Static ATR numbers for SL,TP,TS,BE ... etc. Static means the values calculated at the trade open time, stay the same during the trade lifecycle.
5. Mimimizing and Organizing the parametrs

NOTE: I do not have VPS so I am running from my PC.

The reason I used M30 is because I would like to see more trades open (on average 5-10 a month). In order to better evaluate the EA.
However I still think that higher time frames (H4) will generate more reliable signals.

Look you are right, but if you change some features of money management, there will be more interesting results.
if you send me QML file, I can try something and discuss it with you, if you don't mind.

Position Sizing in the new EA is as follows:

If you set the risk_perc to 0.05 and fixed_lots to 0.0.
and suppose that current account balance is $10,000, and the calculated Stop Loss is 135 pips.
The EA will get the pip value (for GBP/USD) = $10.

So our risked amount for this trade is 0.05 x $10,000 = $500.

The posotion size = Risked Amount/(StopLoss x PipValue) = $500/(135*10)=0.37 lots.

EDIT : And of course it will be rounded to the minimum lot size allowed by the broker.

Thanks. did you misspell MQL ? just kidding.

This response is for any one asking for MQL file.

If you are a programmer, I am sure that it will take you max of 1 hour to develop a similar EA (probably better) based on strategy description in this thread. Some coders might already did that. And if you're not a programmer then you don't need the MQL file anyway.

My focus here is to measure the success of this strategy and look for ways to improve it if possible, Or drop the whole idea.

As you can see the strategy is not using any complicated or fancy indicators, only the ATR for the candle size, and even I am thinking of replacing ATR indicator with normal average calculation inside the EA.

Beleive me, coding is the easy part.

So if you are interested to share the "features of money management", please go ahead, so that eveyone benefit.

Today is the NFP day.

Let's see how the EA will act.

As per the current setup of the EA linked to the excplorer, and at the moment, it will be looking for M30 candle > 47 pips.

EDIT:Forgot to say on GBP/USD.

As Big candle Size is changing witch each candle, Now EA is looking for M30 candle > 52 pips on GU.

If NFP numbers are significant I think EA will open its 1st trade.

The EA did not open a trade yet. as it waits for the current candle size to get bigger that 55 pips. around 1.6718

NFP M30 candle was 1.67725 --> 1.67206 candle size was 52 pips. No trade open as we were looking for 55 pips.

Now EA is looking for a candle of size 67 pips.

So what is the win ratio for this EA ?

I used TP=SL and disabled all other parameters, even MA. So either TP or SL is hit.

I ran the test on GBP/USD for the period 1/1/2013 to 28/2/2014, on H1, H4 and Daily timeframes.

Excel Sheet with results attached.

All results except one showed above 50% win ratio.

I am back from weekend, and ran the above test with "Close on Opposite Signal" set to true (or 1).
Again all other parameters are disabled (I mean no trailing stop, no break even stop, no partial profit taking, no moving average).

I ran the test on GBP/USD H1, H4 and Daily. the Excel sheet attached showing Win Ratio, Expectancy, and Profit.

AMAZING Results ! . Specially on larger time frames.

I've also attached the test for the best result on daily time frame.

Enjoy !

For people not familiar with Expectancy.

Expectancy is a simple formula you can use to determine how reliable your Forex trading system is. You should go back and look at all your trades that were profitable versus all your losing trades. Then calculate how profitable your winners were versus how much your losers did cost you."


Here is the expectancy formula:
E= [1+ (W/L)] x P – 1
where:
W means Average Winning Trade
L means Average Losing Trade
P means Percentage Win Ratio
Example:
You have made 10 trades. Six of them were winning trades and 4 were losing trades. That means your percentage win ratio is 6/10 or 60%. If your six trades brought you profit of $2,400, then your average win is $2,400/6 = $400. If your losses were only $1,200, then your average loss is $1,200/4 = $300. Apply these figures to the expectancy formula:
E= [1+ (400/300)] x 0.6 – 1 = 0.40 or 40%.
In this example, expectancy of your trading system is 40%. That means your trading strategy will return 40 cents per each dollar invested (traded) over the long term.

Above text is taken from : forexsocialtraders