1 (edited by Popov 2017-12-30 20:41:21)

Topic: How to create such a strategy?

How to create such a strategy? https://www.mql5.com/en/market/product/7205 Help me please.

EDIT:

Superman ZigZag Indicator
Superman ZigZag Indicator is based on ZigZag Indicator and it determines the range (Previous High and Previous Low).

- If price is over the High point or lower the Low Point: Market is Breakout.
- If price is under the High Point and above the Low Point: Market is Sideway


https://c.mql5.com/31/44/zigzag-superman-indicator-screen-1823.jpg

Re: How to create such a strategy?

When you ask questions, please post the info instead of a link only.

ZigZag indicator is dangerous. It's standard implementation repaints the past signals, which may lead to a false backtest. This indicator was removed from FSB 6-7 years ago.

Re: How to create such a strategy?

Hi,
I don't fully understand these white arrows. Why they are both short and long ways in swing h/l areas? Does it depends how last two highs or lows is ordered, and only then deciding to buy or sell?

If you want to enter for a break out of swing highs or lows, you can use Ross Hook indicator. To manipulate sensitivity(period) you can set it to use "Longer time frame".

https://image.prntscr.com/image/0oClA9hGSmyG9nNKJ9Z-Vw.png

Also you can use the same indicator for reversal too. You just can't use both condition in the same strategy.
You can try to find filters with FSB for these different market regimes - "Market is Breakout" or "Market is Sideways".
However if you want to have it in the same strategy with no other filters, probably you need custom indicator, which you can order in paid coding service in the forum.
Hope it helps,
Happy new year

4 (edited by Irmantas 2017-12-30 21:42:41)

Re: How to create such a strategy?

Not so bad for few minutes work smile to go for a break of these recent highs/lows. However JPY pairs was trending for a long time. If market changes to more rangy type, this strategy would fail, like it does with other currencies than JPY. So I guess some filtering is a must. Actually I liked this idea to try to differentiate between trending and ranging condition by looking how two last highs or lows was placed. If market makes higher highs, you have trend and opposite is true.

https://image.prntscr.com/image/vVw6OXxbTP6tiWsJPccTyA.png

Post's attachments

break out UJ d1.xml 9.62 kb, 6 downloads since 2017-12-30 

You don't have the permssions to download the attachments of this post.

Re: How to create such a strategy?

Demo account. Zigzag. https://www.mql5.com/en/signals/130958

Re: How to create such a strategy?

Irmantas wrote:

Not so bad for few minutes work smile to go for a break of these recent highs/lows. However JPY pairs was trending for a long time. If market changes to more rangy type, this strategy would fail, like it does with other currencies than JPY. So I guess some filtering is a must. Actually I liked this idea to try to differentiate between trending and ranging condition by looking how two last highs or lows was placed. If market makes higher highs, you have trend and opposite is true.

https://image.prntscr.com/image/vVw6OXxbTP6tiWsJPccTyA.png

How did you create the strategy?

Re: How to create such a strategy?

wewewe888 wrote:

How did you create the strategy?

Strategy was attached in the last post.

But I can do one more time with a picture smile Used USDJPY daily data.

https://image.prntscr.com/image/LmIwPjy0QbWuzQWv5U2t7g.png

Re: How to create such a strategy?

Thank you

Re: How to create such a strategy?

This is my result.
http://i.piccy.info/i9/4b1e9b99fb8bab7cf7c134cfb15c47d7/1514995784/51387/1209891/2018_01_03_180516_800.jpghttp://i.piccy.info/a3/2018-01-03-16-09/i9-11930189/800x314-r/i.gif
http://i.piccy.info/i9/33babf9d915bc61a9f142926c0876cb7/1514995835/34617/1209891/2018_01_03_180549_800.jpghttp://i.piccy.info/a3/2018-01-03-16-10/i9-11930196/800x288-r/i.gif
http://i.piccy.info/i9/4777ad333d52728801e6392d5b8f94ca/1514995860/27736/1209891/2018_01_03_180634_800.jpghttp://i.piccy.info/a3/2018-01-03-16-11/i9-11930199/800x110-r/i.gif

10 (edited by wewewe888 2018-01-08 02:21:16)

Re: How to create such a strategy?

Sorry, I could not create the strategy.
This is the right strategy.
http://i.piccy.info/i9/963c10a0f5bf9bfdc999fa65fd22cbff/1515371222/51756/1209891/2018_01_08_022001_800.jpghttp://i.piccy.info/a3/2018-01-08-00-27/i9-11941993/800x195-r/i.gif
This is my strategy.
http://i.piccy.info/i9/370afc941ef8cbc02a2609e0784086a4/1515371377/43116/1209891/2018_01_08_022534_800.jpghttp://i.piccy.info/a3/2018-01-08-00-29/i9-11941995/800x186-r/i.gif

Post's attachments

usdjpy-h1.xml 9.89 kb, 2 downloads since 2018-01-08 

You don't have the permssions to download the attachments of this post.

Re: How to create such a strategy?

I painted it.
http://i.piccy.info/i9/0adfad480cd025d697426ffc9b4446c4/1515377321/55160/1209891/zygzag.jpghttp://i.piccy.info/a3/2018-01-08-02-08/i9-11942023/800x400-r/i.gif

Re: How to create such a strategy?

Wewe, is there a special reason to want exactly that strategy?

In most of the case focusing on a single pattern and trying to make it profitable is a waste of time. I suggest you to run the Generator and it may create you several good strategies for the market you want.

Re: How to create such a strategy?

Popov wrote:

Wewe, is there a special reason to want exactly that strategy?

In most of the case focusing on a single pattern and trying to make it profitable is a waste of time. I suggest you to run the Generator and it may create you several good strategies for the market you want.

The reason is profit 7525.42% https://www.mql5.com/en/signals/130958

14 (edited by hannahis 2018-01-09 08:46:21)

Re: How to create such a strategy?

wewewe888 wrote:
Popov wrote:

Wewe, is there a special reason to want exactly that strategy?

In most of the case focusing on a single pattern and trying to make it profitable is a waste of time. I suggest you to run the Generator and it may create you several good strategies for the market you want.

The reason is profit 7525.42% https://www.mql5.com/en/signals/130958


Yes, that's sounds like a very valid reason and a big motivational force, who would want to give such a profit 7525.42% a missed.  I fully understand

Re: How to create such a strategy?

Yes, it is an excellent result. However it is much simpler to subscribe to the signals instead to try to reverse engineer the trading rules.
 
The most experienced users of the forum must know a strategy that we post 10 years ago, which still works even today. It is amazing that this is one of the most simple possible strategies. I found it on the first days I experimented with the Generator prototype.

Here is a very extreme, however 100% correct result. (profit 13 221%)

https://s9.postimg.org/tzclgz4lr/screenshot_98.png

Re: How to create such a strategy?

Nice Popov, though I haven't tried this strategy.


Can you post the link to view the updated results and the download of the EA somewhere in another post for new comers to view

And you can use it as your marketing and promotional tools to let them know that when they buy your software, they inevitable "inherted" such good EA and be impressed by FSB's capabilities.

Re: How to create such a strategy?

This is my result
http://i.piccy.info/i9/c2d376f7e6374b57cf267629e09e7b84/1515517240/47986/1209891/2018_01_09_185900_800.jpghttp://i.piccy.info/a3/2018-01-09-17-00/i9-11946746/800x300-r/i.gif

Re: How to create such a strategy?

Popov wrote:

Yes, it is an excellent result. However it is much simpler to subscribe to the signals instead to try to reverse engineer the trading rules.

I can not subscribe to the signal. This is a demo account.

Re: How to create such a strategy?

Instead of getting too fancy developing strategies, why not work with what is available in the software today.

Believe me, there is no reason to look to add things, we have a lot of tools that can generate some nice results.

By keeping it simple, we can use this software to make a nice return, trying to add complex stuff will be a waste of time and end up with frustration.


https://s18.postimg.org/8ymm0l39h/Dave_M-0237.jpg

My 'secret' goal is to push EA Studio until I can net 3000 pips per day....

20 (edited by hannahis 2018-01-12 17:55:25)

Re: How to create such a strategy?

I totally agree.  FSB's optimisation tool has improved tremendously and as long as you have a simple workable trading theory, you can input it in FSB and allow the optmization tool to improve your work further and bring you really good results.

Attached is the results of some of my optimized Strategies.
https://s13.postimg.org/mh9656803/opt_results.png

Here's another portfolio, all optimized Strategies (a collaborative effort with another FSB member, though it's still early but the results look promising for the start).  So from this collaborative team work, I realized that FSB's optmization tool really has made our job in building profitable EA much easier. 
https://s13.postimg.org/agnq4a0jn/opt_portfolio.png

Thus now I can fully agree with Dave (BlaiserBoy), keep it simple.  Let FSB do it's job.

Re: How to create such a strategy?

hannahis wrote:

I totally agree.  FSB's optimisation tool has improved tremendously and as long as you have a simple workable trading theory, you can input it in FSB and allow the optmization tool to improve your work further and bring you really good results.

Attached is the results of some of my optimized Strategies.
https://s13.postimg.org/mh9656803/opt_results.png

Here's another portfolio, all optimized Strategies (a collaborative effort with another FSB member, though it's still early but the results look promising for the start).  So from this collaborative team work, I realized that FSB's optmization tool really has made our job in building profitable EA much easier. 
https://s13.postimg.org/agnq4a0jn/opt_portfolio.png

Thus now I can fully agree with Dave (BlaiserBoy), keep it simple.  Let FSB do it's job.

Hi Hannahis,

It's interesting to hear you say this, I have been reading some of your posts about strategy creation, and you seem to be an advocate of complex strategies, and my intention was to read a lot more of your posts to help me grasp your method..  Is it now that you think keeping it simple is the best way forward? 

Some of you may have seen me show up from time to time over the last couple of years, each time coming back to start anew, but I'm really trying to get somewhere with it this time, regardless of life's obstacles.

I'm trying to work out a repeatable step-by-step process to strategy creation, and how to properly analyse one using FSB Pro's tools - eg; Monte Carlo, and how to properly  go about optimising as well. Would love some pointers smile

I have also just 'Cleaned House" on a VPS that I will use for EA's, which had numerous instances of MT4 with bad file paths, so I made the decision to sacrifice the data they had gathered, (which seemed to have lots of errors, bars with wrong time etc, a notepad++ nightmare..) and create new instances with clear file paths that I will now not mess with. This means my data is very sparse, which led me to reading about your M1 strategy preferences, as I will obviously have more M1 bars in the near future than any other timeframe. Would you say that I am going to need to wait quite a while for my data to come to fruition, or could you suggest an alternative course of action for creating strategies relevant to my broker in the meantime?

FSB Pro was also in a bad way due to bad file paths to data sources, Miroslav recently helped me by advising a total reinstall of the software, it's working fine now, I'm keen to make some good strategies..

Simon.

22 (edited by hannahis 2018-02-16 02:06:17)

Re: How to create such a strategy?

Hi Simon,

There are several ways to make profitable EA via FSB.

1) The data mining cum statistical approach using Generator, MC, etc

2) Input your trading theory approach (using Longer Time Frames, etc)

3) Input your trading theory and use FSB's optimizing tool approach (input your trading theory and then use FSB's optimizing tools to fine tune)

Which of the above mentioned methods would one use depends largely on a person's personal preference.

I personally prefer method 1 & 2

When I said keep it simple, it's with reference to "being stuck" with certain indicators or trading theory that not "found" in FSB and hence, I'm trying to convey the message that there are a lot of "workable trading theory" one can use in FSB to create profitable EA.

One can develop profitable EA even just using simple MA cross over theory and one of the possible reasons why people don't seem to get profitable EA may due to the following reasons

1) The trading theory isn't "workable" in the 1st place.  Make sure you can see how it works on your MT4 chart.  If you can't even "see" the trading edge on your MT4 charts with your indicators in place, then most likely the trading theory isn't that "sound" in the 1st place.  Over the years, I've developed a workflow that can turn any workable theory into profitable EA.  However, my workflow isn't a miracle worker to turn a terrible trading theory into a profitable one.  Thus the 1st key to developing profitable EA lies in one's trading theory...how "sound" is it.

2) Fail to use the correct combination.  I don't know how not to emphasis this as one of the Key important factor that can determine one's success or failure.  I've demonstrated this essential point before.  Even if you use the "appropriate" parameters, if you use the wrong combinations (smoothing method, base price, signal method), you will ended up with unprofitable EA.  But if you use the correct combinations.  You "hit the jackpot" and you will see what a big difference the correct combinations can do to one's EA.

Hint: Do not use "Close" as your base price.  From my experiment, 99% of EA using "Close" as their base price ended up unprofitable (or have the worst results among other combinations).  Of cos that depends on what's your trading theory.  If the rule is price close above lower band after opening below before it etc, then I understand the need to use "Close" but majority of the time, most of my profitable EA has "Medium" as my base price.   So you may want to try using Medium as the best alternatives.

In the past I use a lot of complex rules to improve my EA and after several attempts (of not able to improve my EA significantly) I decided to search for the best combinations rule by rule.  A very tedious but extremely rewarding task.  To my amazement, even with only 1 trading rules, my results were far better than those with multiple rules (but wrong combinations).  Hence, now I build my EA rules by rules, i.e find the best combinations for the 1st rule and then move on to find the best combinations for the 2nd rules.

I know the thought of having to go through such tedious procedures may put others off.  What's can I say? Is gold found on the surface or does one need to spend efforts to search of it?  The results is far rewarding, that I can conclude.

3) Failure to put much thought to the whole process of EA development.  If one expect to find highly profitable EA via FSB without much thought.  Any how input your Opening and Closing rules and then "Volia" and expect to find profitable EA, then I would say, it's not that simple really (maybe you can find "low quality" EA with low PF value etc).

In conclusion:

1. Keep it simple - yes, you don't need complicated theory to find profitable EA

2. However to build highly profitable EA, one need to build in some "complex" rules to eliminate side way trending for a breakout EA in order to increase it's profitability and reduce Max DD during ranging times.

3. If you have a sound theory, you can use FSB optimizing tools to help you improve your EA tremendously if you know how to use it to your advantage (if you put some thinking into the whole process).

4. My EA are build via trading theory input and I rely on demo testing to confirm my trading results (so even if FSB's backtest tell me that my EA is losing, I still go ahead to do demo test).

5. I never use Monte C. or Multi Market or OSS or walk farward etc in my EA development process (I'm not saying don't use it, what I'm trying to say is, I don't need to use it in order to find profitable EA).  I'm not saying I never tried them before but these features aren't part of my workflow.

6. If anyone can't find profitable EA via FSB, kindly do some homework and learn some trading rules.  FSB is just a trading tool, we still need to use our brains.  I've built countless of profitable EA over the years.  So I'm very sure it's possible to use FSB to build profitable EA.  There is no short cut to success.  However, If anyone wants an easy way out, go and subscribe a trading signal, that's what I would advice.  Building quality EA requires commitment and dedication.  That's where many fail and give up.  Forex trading isn't a get rich quick scheme, likewise, FSB isn't a get profitable EA quick method (without much thots).  But once a person has a successful workflow, yes, it's is very "easy" then to churn out thousands of profitable EA.

Ok Simon, I hope I've cleared up some of the confusions and I wish you all the best in your EA development progress and hope you find success in your journey.

23 (edited by Freesby 2018-02-16 17:30:53)

Re: How to create such a strategy?

I sincerely believe that these statements are completely illogical:

... there are several ways to make an EA profitable with Fsb ....

- even if Fsb gives a negative result I try the strategy in demo,
- I do not use it montecarlo
- I do not use multimarket
- I do not use OOS
- I must know the rules of trading
- use my brain
- do homework

I think you're trading with coffee funds, head or cross, odd or even, because you do not need Fsb.

Re: How to create such a strategy?

hannahis wrote:

There are several ways to make profitable EA via FSB.

1) The data mining cum statistical approach using Generator, MC, etc

2) Input your trading theory approach (using Longer Time Frames, etc)

3) Input your trading theory and use FSB's optimizing tool approach (input your trading theory and then use FSB's optimizing tools to fine tune)

Which of the above mentioned methods would one use depends largely on a person's personal preference.

I personally prefer method 1 & 2

.

Correction: I prefer method 2 and 3

25 (edited by hannahis 2018-02-16 20:31:06)

Re: How to create such a strategy?

Freesby wrote:

I sincerely believe that these statements are completely illogical:

... there are several ways to make an EA profitable with Fsb ....

- even if Fsb gives a negative result I try the strategy in demo,
- I do not use it montecarlo
- I do not use multimarket
- I do not use OOS
- I must know the rules of trading
- use my brain
- do homework

I think you're trading with coffee funds, head or cross, odd or even, because you do not need Fsb.

Hi Freesby,

I'm not too sure why some of the statements seem illogical to you.

Let me try to see which is illogical...or maybe logical to you.  I'll try to reverse those "illogical statements" I've post and see whether they make more sense to me.

It is illogical that... Let me tackle my "illogical statements" one by one.

1) there are several ways to make an EA profitable with Fsb .... If this statement is illogical...does that mean the logical conclusion should be the statement below instead?

"So it is logical that there is only One method to make profitable EA in FSB?

There is no other way and the quest is therefore to find this "magical method that is so elusive that very few can find it?"


2) even if Fsb gives a negative result I try the strategy in demo,, if this is illogical, idea of logical statement should be....

"I believe wholeheartedly that FSB's backtest is faultless and accurate and that what I see in the backtest will be what I get in real live/demo trading.  So if FSB's backtest shows me profitable EA, that particular EA will therefore definitely show profitable results in both live and demo testing.  Likewise, if FSB backtest show negative results, I should "throw" these EA away and don't bother to do some demo testing to verify whether FSB's backtest is correct or not."

(Comment: if one were to approach this with a "scientific mind, one would at least do some demo testing to verify whether there is congruence between demo results and FSB's backtest results, in order to establish that element of "trust".  Even if you truly want to believe wholeheartedly (by all means) in FSB's backtest results, have you put your "faith" to test? Have you verify it with demo testing?  Or was it blind faith?)

3) I do not use it montecarlo.    If this is illogical, idea of logical statement should be....

"You will only succeed if you use Monte Carlo.  You can't rely on yourself to find the most appropriate parameters.  Forex is so complex that only a computer can compute the most appropriate parameters for you.  Tough luck to discretionary traders cos without any help from Monte Carlo simulation, you can not find any success in your manual trading.  Any successful manual traders out there must be a scammer cos how can one find profitable EA without MC?  So a bit of advice to those who wants to success in trading.  You better use MC"


4) I do not use multimarket  if this is illogical, idea of logical statement should be....

"I believe that all profitable EA should survive and stay profitable in Multi Markets.  If an EA is to be profitable, it must be profitable in all multi markets.  If any EA isn't profitable in 1 market, it probably means it will fail in the rest of the other markets too.  I believe that all markets dynamics are the same.  Hence one size fits them all.  Therefore to find profitable EA, one must run it through all Multi Markets...if that EA pass Multi Markets, it will therefore be profitable.  A jack of all trades, master of none."


5) I do not use OOS  if this is illogical, idea of logical statement should be....

"First of all, I trust FSB's backtesting methodology and hence whatever is the backtest result, it must be the absolute truth.  Hence, if my EA can also pass the OOS test, it means it is profitable."


6) I must know the rules of trading if this is illogical, idea of logical statement should be....

"I don't need to know trading at all, who cares.  I just need to do lots of trial and errors with different indicators and run the FSB's generators for days and months and set very good acceptance criteria and at the end of the generation, voila I get profitable EA, easy peasy lemon squeezy."


7) use my brain   if this is illogical, idea of logical statement should be....

What for?  What's the point of using my brain.  What am I trading, coffee fund? No big deal if I lose.  Trading is so easy and FSB made it even more easier.  Making profitable EA is such a mindless game.

(This statement contradict with one of the statements above that implies Forex is so complex that one needs MC in order to succeed, otherwise, you will fail, if it's so brainless to find profitable EA, why need such complex function like MC to search for you?)


8) do homework  if this is illogical, idea of logical statement should be....

"One can find profitable EA without much efforts.  It is so brainless, you really don't need to know much about it.  So don't bother to do your homework, just go and trade live and you will win."


Ok, Freesby, I try to breakout down which statements is really illogical to you and then try to reverse the "illogical" statement to see whether I can find any logic behind them.  So which of the above counter argument statements (made up by me, I'm not claiming you are saying these statements) seem more logical to you? 

Lastly, it doesn't make any difference to me whether anyone believes my statements or not.  You can go ahead to insult me about my trading funds...it's peanuts anyway, nothing to boast about. 

And at the end of the day, what matters to me is whether I can build successful EA or not.  Having more or less people believing in my statements won't make my EA's Profit Factor any higher or lower. 

I commented here because someone asked me a question and out of respect I try to make an effort to reply.  Otherwise, I'm pretty happy now to say out of the grid (and I would have avoided this "discussion") cos I've learnt, it doesn't pay to give my trading tips away, might as well stay quiet and be contented with my life.  Hence, I have reduce my posting in this forum.


Basically, there are Two main approach in using FSB.

1) You can see FSB as an Execution Tools - translate your trading theory/rules into MT4 EA to execute your trade on your behalf, without the need to learn coding language.

2) You can see FSB as data mining, statistical/analytical tools for you to search for your EA

As for me, I use FSB mainly as my Execution Tools and hence I hardly rely on the rest of the functions.  So Yes, I do need FSB, cos it is one of the best software (so far in the market, though there are still rooms for improvements) that can help me turn my dream into reality.  Just because I don't rely on other FSB's tools doesn't mean I don't need it.  I will be lost without it.


My statements are not just opinions (as everyone is entitled to theirs, even if yours differ from mine).  My statements are based on years of experience of developing thousands of profitable EA.  So between your opinion and my experience, which do you think I'm more inclined to believe in? 

Last of all, I still sincerely wish you all the best.  I can understand my statements may incurred some people "wrath" especially for those who believed so much in their work flow (of using MC, OSS, Walk forward, backward etc) that they think there is no other alternative of developing any profitable EA other than using these tools to aid you. 

And when I said I don't use these work flow processes (such as MC, OSS etc), I'm not saying they aren't useful (that's not my point).  I'm trying to point out alternative methods, that's all.  So that others don't have to think that one MUST use these functions in order to find profitable EA.  One can still build profitable EA without it.  I'm here to introduce or expand one's perspective in their EA development approaches.


In conclusion, I'm just sharing my own experience.  Does that mean anyone who has an opinion to share here, better be a guru, otherwise keep quiet?  And if I share my experience, my ups and downs, what I've discovered working for me and what's not working for me...am I therefore considered a guru?  Isn't what this forum is for, to support one another and help one another find their roads to success?

No, I'm not a guru nor will claim to be.  I'm just an ordinary retail trader...hoping to put my dream into reality by building profitable EA.  So I got nothing to prove and nothing to gain here.  This forum is for people to share freely their opinions (like it, take it, don't like it discard it...nothing personal to warrant any insults or personal attack, even if other people's opinion differ from mine, let's show some mutual respect).

Best of luck and I hope you find success in your method.

Hannah