Scalp_net - page 19

New comment
 

Hi,

i've changed Expert's code for your needs.

So now Trailing Stop moves with SAR for TrailingStop=0.

Files:
scalp_net_v1_sar.mq4  12 kb
 

scalp_net modification

Dear Igorad,

Many thanks for your help.

I owe you much

God bless and good trade!

 

ScalpNet_v1.5

Edited

-------------------------------------------------------------------------

ND,

Nothing wrong with ScalpNet_v1.5. I did not use 21 and 200 EMA on chart.

Thanks in advance,

TimeFreedom

-------------------------------------------------------------------------

ND,

I am posting the following chart to help with what looks like a bug in ScalpNet_v1.5. All critea was met on EURUSD chart earlier today and EA did not open Buy trade. Please take a look and advise.

Thank you,

TimeFreedom

 
TimeFreedom:
Edited

-------------------------------------------------------------------------

ND,

Nothing wrong with ScalpNet_v1.5. I did not use 21 and 200 EMA on chart.

Thanks in advance,

TimeFreedom

-------------------------------------------------------------------------

ND,

I am posting the following chart to help with what looks like a bug in ScalpNet_v1.5. All critea was met on EURUSD chart earlier today and EA did not open Buy trade. Please take a look and advise.

Thank you,

TimeFreedom

No any bug.

Scalp_net_v1.5 is not original Scal_net.

It was modified.

Look at the image:

21 ema cross 200 ema confirmed by MACD (default settings), I-XO indicator with period 15 and Parabolic SAR (0.005 and 0.05) in 28th of July only.

And this EA took 9 pips.

Why 9 pips only?

Because this EA is trading on the closed bar (look at the second image) and trailing stop is 30.

There are no any wma, 50 ema coded.

21 ema, 200 ema, I-XO (period=15), Parabolic SAR (0.005; 0.05) and MACD (default settings). Besides MACD was coded not in usual way: it was coded signal line only (it was coded some part of MACD not whole the indicator).

No any bug.

Simple this EA was designed for EURUSD only but can work for EURJPY as well.

Files:
scalp.gif  25 kb
scalp2.gif  22 kb
 
newdigital:
I have some vision of the situation. May be I am wrong but according to my understanding there are different forex programming schools:

1. It is allowed in "Russian school" to open one...

....

....

So it was my vision of this situation.

May be I am wrong but i am always looking on the history to understand how it was started from the beginning.

Thank ND! I love the story part of history...

 
newdigital:
It was statement with timefilter https://www.mql5.com/en/forum/173442

I mean the trades were sorted by time (08:00 - 18:00 North Finance server time).

I changed someting inside the code so please find this EA with timefilter (with pre-set file).

HI ND - I think you need to change the "I am testing this version" URL that you post with the statements for scalp_net 1.3 - the URL points to the non-TF version of the EA...

Unless I am misunderstanding, and the statement posts that you are posting here are separate from the Total Leader posts.

Thanks again for all your work,

CS

 

Newdigital,

Did you ever post the code for Scalp_net_v1.5 with timefilter? If so, could you post a link to it. I can't find it, sorry.

Thanks,

 
barry:
Newdigital,

Did you ever post the code for Scalp_net_v1.5 with timefilter? If so, could you post a link to it. I can't find it, sorry.

Thanks,

I posted Scalp_net_v1.5 with timefilter with timefilter now https://www.mql5.com/en/forum/general

Scalp_net1.3 with timefilter should be in files' thread as well https://www.mql5.com/en/forum

I am posting the statements for EAs with/without timefilter all together. You may find the ziped folder namely "total_pips_tf", or "open_pips_tf", or "week_pips_tf".

tf is timefilter.

Please note that Scalp_net v1.5 with timefilter is testing since 04th of July https://www.mql5.com/en/forum/173383/page36

but Scalp_net_1.5 without timefilter is testing since the beginning of June.

Why i am testing EA without tf since June and with tf since July? Because i changed accounts and MetaTraders 3 or 4 times for Scalp_net_1.5 moving it to the other MetaTrader and so on. Besides I did not expect that this my EA will be profitable.

Statements are attached. Besides you may find the statements in 3 threads posting all together in one zip file.

As to Salp_net v1.3 with/without timefilter so please find the statements attached. It was started in the same date so it is ok.

Beside please note that i found the bug in the script to produce the statements with timefilter. I found it one month ago. This script was corrected and it is ok now. I updated all the statements one month ago and i corrected all the figures in weekly performance thread and total leaders thread. It is concerning EAs with timefilter only. It was just small mistake. May be I forgot to correct somewhere so it is the better way to look at the statements.

Anyway we will have yearly leaders soon. I mean EAs which were profitable during the one year forward testing. And after that i will create the threads in Portfolio section about each yearly leaders with statements, excel files, indicators and so on.

 

Newdigital,

Thanks for that information. I can see that you get good results from the Scalp_net series in your weekly testing, so I keep trying to use them, but whenever I backtest v1.3, 1.3tf, 1.5 or 1.5tf I find:

1 They always make a loss over 12 months, even using your presets, and

2 Each 12-month test takes 8-10 hours, so it's impossible to optimise in order to find settings that are profitable.

I wonder what we can do differently. It would help me if the backtesting could be made faster. I think it's always very slow to test an EA that calls an indicator such as I_XO_A_H.mq4, so would it be possible to include that indicator inside the EA? Just a suggestion. I want to see it profitable on backtest before I put it on demo test.

 
barry:
Newdigital,

Thanks for that information. I can see that you get good results from the Scalp_net series in your weekly testing, so I keep trying to use them, but whenever I backtest v1.3, 1.3tf, 1.5 or 1.5tf I find:

1 They always make a loss over 12 months, even using your presets, and

2 Each 12-month test takes 8-10 hours, so it's impossible to optimise in order to find settings that are profitable.

I wonder what we can do differently. It would help me if the backtesting could be made faster. I think it's always very slow to test an EA that calls an indicator such as I_XO_A_H.mq4, so would it be possible to include that indicator inside the EA? Just a suggestion. I want to see it profitable on backtest before I put it on demo test.

I made original Scalp_net long time ago. It is forward testing since January this year and it is profitable. I traded this system manually during the more than 1 month and this system is for EURUSD only. I could not find the right settings for the other pairs.

Then I asked Igorad to modify Scalp_net to have magic number option and to trade with other EAs with the same Metatrader. He helped me and we received V1.2 version.

Then I asked him to improve 1.2 version to have maximum-order-at-the-time option. He did it and we have 1.3 version.

Then I improved 1.3 version to have 1.5 version. Why? Because data feed for M1 timeframe is diferent with different broker. So I wanted for this EA to trade on M5 timeframe. Because M5 timeframe data is much more similar for many brokers. I improved 1.3 version to work on M5 timeframe. It is still for EURUSD.

Backtesting. I backtested this EA. But according to backtesting this EA is profitable for some years (it is not profitable for every year since 2001).

Please find backtesting results for 1.5 version for 1 year backtesting.

And please note that this is risky EA (all EAs based on EMA crossing are risky EAs).

1...121314151617181920212223242526...73
New comment