Help Needed to Debug Existing EA with proven strategy | Page 9 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Help to debug my EA 2 replies

Developer Needed To Debug my EA 2 replies

MT4 EA coder required to fine-tune/debug existing EA 0 replies

Debug code in strategy tester 1 reply

debug message MT4 EA in Strategy Test 7 replies

Attachments: Help Needed to Debug Existing EA with proven strategy
Exit Attachments

Help Needed to Debug Existing EA with proven strategy

  • #161
  • Apr 5, 2018 6:32am Apr 5, 2018 6:32am
Hi Rockit,

Thank you so much for the speedy reply.

Rockit, I will start a new thread in a few days time to discuss on the new developments. I hope you have read what I have written on the new developments. I sincerely hope you will join us in the new thread and I intend to use your EA as the base as it can already trade in both legs concurrently.

For now, I think don't be too bothered about removing the bug - let me finish this thread first.

Once again, thank you very much for your generosity.

Cheers Rockit!
  • #162
  • Apr 5, 2018 8:28am Apr 5, 2018 8:28am
Hi Rockit,

Quote:

"Now, regarding testing: You know there is a tester built into that freaky MT4? Therefore you can test any expert, at any time, on past data, also visually, and see outright if it works alright (and how it performs). There you can also see that the strategy is not exactly profitable - in live you may see the closed-in profit trades, but at the same time losses from open positions are mounting.. Your strategy is basically a grid, and grids (grid strategies) are not known to work."


Rockit, I have to disagree with you here. It is NOT a grid system. I know what is a grid system, I have traded in grid system and in fact had a few ideas of my own which were coded into EA for me. I do agree with your statement that all grids are not known to work. Let me explain to you why this is NOT a grid by using a live example of the on going cycle on the pair gbpjpy which is on going and not closed yet.

Please look at the screen shot here:

Attached Image (click to enlarge)
Click to Enlarge Name: current gbpjpy.jpg Size: 607 KB


Please look at the trade I highlighted and the trade immediately above it. Do you see the gap between the two orders??? 97 pips!!! That is what I have been trying to explain to Yalgaar this morning - namely, NOT all the levels are triggered when the price goes against us but the opposite is NOT true - when the price goes in our favour - EVERY SINGLE LEVEL will be triggered - can you see Rockit??? This is NOT a grid - this is where I believe the 'edge' is - Honestly Rockit - don't you agree with me that in the live example above, if we have the other leg trading concurrently, the whole cycle would have been closed somewhere along that 97 pips gap when ALL the levels in our favour would have been opened (almost 7 orders would have been triggered in that 97 pips gap) and should close the whole cycle in profit!!! Don't you think so Rockit??? This is NOT a grid although I must admit that it looks like a grid.
  • #163
  • Apr 5, 2018 8:43am Apr 5, 2018 8:43am
Rockit,

That's why I said yesterday that if we are to trade both legs together, we can only trade a VOLATILE pair and I had clearly stated my choice would be gbpjpy because in volatile pairs, chances are that less orders would be triggered in the leg against the direction whereas ALL the levels would have been triggered in the other leg. The pair gbpjpy is known to be one-direction minded with little retraces along the way - I am not a newbie, I traded extensively in 2008 to 2012 and I have tried all sorts of systems.

Rockit & other coders, just give it a try, it may or may not be a gold mine - we will see.

AAARRRHHHH...how I wish I could code that bloody thing myself !!!
  • #164
  • Apr 5, 2018 10:00am Apr 5, 2018 10:00am
Hi Philip Lee

It seems like we are going back and forth with this and you seem to be not getting my point and I seem to be not getting your point. :-)

All I have been asking you is to tell where is the edge of this method? I am asking this repeatedly since I do not see the edge. I am not saying there is no edge here...I am saying I just don't see it.

I take full responsibility that I didn't ask you the right question and I didn't present you the exact scenarios since I just could not structure it in the right way and honestly been very busy as well...and just a little bit lazy! :-)

I will now attempt to put this all together in the right way.

1) This strategy can be used to trade when you have a direction decided. Agreed that it will generate handsome profits if your direction is right. But it will generate terrible losses as well when the direction is wrong. I believe you are already convinced about this. Please tell me if not the case and we can talk more about it. So this will prove that it is all about direction and the method is not giving you any edge.

2) You changed your opinion and now believe that you can use this EAs to trade "Buy" and "Sell" at the same time. I can show you why this is also a bad idea and will only generate losses (I know you don't like to read this and I am sorry but that is what I believe right now and I want to be wrong) So before I show you what I think and why I come to such conclusions...I want to ask you how you will run the "Buy" and "Sell" EA at the same time. Lets take an example of "GPBJPY" pair. Lets say this pair moves daily around 150 pips (you can check how I came up with this number) What parameters you want to use for the EAs? Once you tell me the parameters you want to use...like the depth and TP and SL and the rest...I will tell you exactly why this is will not work. I will show you many Market situations(which are very common) where this will only incur losses.

Let me know your thoughts?

-Y
  • #165
  • Apr 5, 2018 10:08am Apr 5, 2018 10:08am
Quoting Philip Lee
Hi Rockit, Quote: "Now, regarding testing: You know there is a tester built into that freaky MT4? Therefore you can test any expert, at any time, on past data, also visually, and see outright if it works alright (and how it performs). There you can also see that the strategy is not exactly profitable - in live you may see the closed-in profit trades, but at the same time losses from open positions are mounting.. Your strategy is basically a grid, and grids (grid strategies) are not known to work." Rockit, I have to disagree with you here. It is...

I really don't agree with you on this. Once you reply to my previous post with the parameters you want to use, I will prove to you exactly why I don't agree.
  • #166
  • Apr 5, 2018 10:09am Apr 5, 2018 10:09am
........or maybe I will learn something new and that would just be great! I hope this is the case but I highly doubt.
  • #167
  • Apr 5, 2018 10:32am Apr 5, 2018 10:32am
Quoting tictac-nur
{quote} Forex RM Version 2.0 Download Link : {file} Change Log : 1) Display Maximum Draw-down 2) Display Maximum lot size 3) Display Maximum total order 4) Display Server Time 5) Auto Entry base on time (Day and Hour) 6) Added One Cycle Only (After profit target.. no new entry until you turn it back on.. it is like ATM machine.. if you need USD100..just key in USD100 target profit..after target accomplished..withdraw your money like ATM machine.. ) Lucky you.. 7) Trailing Profit (By pip.. can get more money if the market in our side)...
TICTAC NUR
Hi Buddy Thanks for this great EA, ive been useing this EA on Demo and back test With 0.02 lot size ,both buy and sell true, one Ea on 1h chart, profit target = £10 and it reaches tp every time , and in most cases makes £10 buy and £10 sell = £20 like in the graph below, i would like to ask one last favour, could you put an option as to when the take profit reaches its goal it closes the other side (buy or sell side down and deletes all pending) because this is where the DD is somtimes, not all the time but it can be and would be nice to just have this option Cheers
Attached Image (click to enlarge)
Click to Enlarge Name: TesterGraph.gif Size: 9 KB
  • #168
  • Apr 5, 2018 10:53am Apr 5, 2018 10:53am
Quoting montego
{quote} TICTAC NUR Hi Buddy Thanks for this great EA, ive been useing this EA on Demo and back test With 0.02 lot size ,both buy and sell true, one Ea on 1h chart, profit target = £10 and it reaches tp every time , and in most cases makes £10 buy and £10 sell = £20 like in the graph below, i would like to ask one last favour, could you put an option as to when the take profit reaches its goal it closes the other side (buy or sell side down and deletes all pending) because this is where the DD is somtimes, not all the time but it can be and would...
Hi Montego,
can you post your set file ? Ive back tested this a few times and have not had good results using buy and sell. Be interesting to test what you tried.
Thanks.
  • #169
  • Apr 5, 2018 11:07am Apr 5, 2018 11:07am
Quoting nihilist23uk
{quote} Hi Montego, can you post your set file ? Ive back tested this a few times and have not had good results using buy and sell. Be interesting to test what you tried. Thanks.
Hi buddy have just tried but when i try and download it to ff its saying -ERROR FILE TYPE NOT SUPPORTED ???
Any ways it all v2 standard settings apart from
1/lot size 0.02
2/tp £10
3/ buy true/sell true
4/trail false
if you back test look at the results you hardly get any drawdown only on the second trade, and somtimes even that goes through without drawdown.
Edit :important - 1 x ea on gbpjpy 1 h chart
  • #170
  • Apr 5, 2018 11:10am Apr 5, 2018 11:10am
another random date
also forgot to say tested with every tick option- same result.
Attached Image (click to enlarge)
Click to Enlarge Name: TesterGraph.gif Size: 10 KB
  • #171
  • Apr 5, 2018 11:17am Apr 5, 2018 11:17am
Quoting montego
another random date also forgot to say tested with every tick option- same result. {image}
occasionally it loses £30 -£40 but imagine if you did this every day and 80 percent you got £30 and 10-20 percent you lost £40
it would also lose less if Tac Tac is kind enough to code that extra option in for us, Needs more testing for sure, but its looking good so far.
yesterdays back test lowered tp to £7 still got £20
Attached Image (click to enlarge)
Click to Enlarge Name: yesterday.gif Size: 10 KB
  • #172
  • Apr 5, 2018 11:19am Apr 5, 2018 11:19am
Yal, will answer you tomorrow, its 12.18 am now.
  • #173
  • Apr 5, 2018 11:19am Apr 5, 2018 11:19am
Montego, you are doing only 1 cycle at a time?
Building a Freedom All Time Return: 5,178.8%
  • #174
  • Apr 5, 2018 11:23am Apr 5, 2018 11:23am
Quoting J1mm
Montego, you are doing only 1 cycle at a time?
Hi bud no 2, but set ea to close after profit, for example ea does sell side first takes profit and closes and then it after a day or so opens trades on the buy side then hits profit and closes, this is ea set to one cylcle only, but the ea is at the moment doing 1 cycle buy and one cycle sell, hope that makes sence..
  • #175
  • Apr 5, 2018 11:24am Apr 5, 2018 11:24am
Quoting Philip Lee
Yal, will answer you tomorrow, its 12.18 am now.
No Problem. There is no rush. Rome was not build in a day!

You have a good night!
  • #176
  • Edited 1:00pm Apr 5, 2018 12:00pm | Edited 1:00pm
Quoting montego
{quote} Hi buddy have just tried but when i try and download it to ff its saying -ERROR FILE TYPE NOT SUPPORTED ??? Any ways it all v2 standard settings apart from 1/lot size 0.02 2/tp £10 3/ buy true/sell true 4/trail false if you back test look at the results you hardly get any drawdown only on the second trade, and somtimes even that goes through without drawdown. Edit :important - 1 x ea on gbpjpy 1 h chart
how are you setting a tp of £10 for back testing. ea only has pip setting?

edit,
ok,Ive got you,so each random back testing day.....so we just let run each day for one cycle,reset and re-run ?

if you set ea to 1 cycle and use buy and sell (true) as long as you set t/p correctly as a universal t/p then ea will only trade once and close all trades and stop. (if im understanding correctly)
  • #177
  • Apr 5, 2018 1:09pm Apr 5, 2018 1:09pm
Quoting nihilist23uk
{quote} how are you setting a tp of £10 for back testing. ea only has pip setting? edit, ok,Ive got you,so each random back testing day.....so we just let run each day for one cycle,reset and re-run ? if you set ea to 1 cycle and use buy and sell (true) as long as you set t/p correctly as a universal t/p then ea will only trade once and close all trades and stop. (if im understanding correctly)
HI Bud, no it doesnt , thats what i want TicTac to code, at present with those settings it trades the side that gets triggered first, but when it reaches Tp the other side can still get triggered, which works out fine sometimes, but other times its on a retrace and there seems more dd on a second trade.
  • #178
  • Apr 5, 2018 1:15pm Apr 5, 2018 1:15pm
Quoting montego
{quote} HI Bud, no it doesnt , thats what i want TicTac to code, at present with those settings it trades the side that gets triggered first, but when it reaches Tp the other side can still get triggered, which works out fine sometimes, but other times its on a retrace and there seems more dd on a second trade.
got you ! thanks for that. it looks promising though from all the day tests ive done. ive done some with no stop loss and so far it hasnt missed a day.trouble is, i cant see any draw down like that. what stop loss did you have set for your tests ?
  • #179
  • Apr 5, 2018 1:28pm Apr 5, 2018 1:28pm
Quoting nihilist23uk
{quote} got you ! thanks for that. it looks promising though from all the day tests ive done. ive done some with no stop loss and so far it hasnt missed a day.trouble is, i cant see any draw down like that. what stop loss did you have set for your tests ?
Just the pre-set ones i basically i didnt touch them, set your backtest to finish on todays date and start off backtests in Jan and work forward, the backtest will only last a couple of days maybe 3 and then of coarse it wont trade as its reached both take profits, leave in graph mode when you see its stopped just stop the back test, you will see the potential..
  • #180
  • Apr 5, 2018 1:53pm Apr 5, 2018 1:53pm
Hi All, The Idea seems to be very similar to 7bit Anti-grid snowball EA which when first placed on chart put pending BUY STOP orders above a central point and SELL STOP orders below a central point. The central point is the price at the moment you were placing the EA on chart. If price goes up, EA keep placing buy stops after certain fix distance and vice versa if price goes down below the central point. All orders carries SL and All orders have a universal TP which is calculated based on SL hits and regularly updated after SL hit.

I have experimented with this idea and programmed many many versions of EA based on alterations to the core idea of 7bit Anti-griding. Thoroughly back-tested on Gold data from 2005 to 2017. The good point is that It don't blow a 1K account from 2005 to 2017 which is amazing as compared to griding EAs . I love it because it is tension free but this trading strategy has profitability flaws. The profit is not that much which can satisfy you mentally. Next it gives better results when you decide when and where to start the EA.

Here is the link to 7bit thread.
https://www.forexfactory.com/showthread.php?t=239717
Forex; the game of movements not the game of candles.
0 traders viewing now