Help make the robot profitable - page 4
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
I have several such branches there in English, and many with EAs in the branches.
And I didn't ask the coders to make free EAs for me. They just saw that the system was profitable, they saw trades, etc., and made them.
Why? Because if I or any other user wants to improve these EAs (for example, to make a martingale as false or true, martingale by a fixed lot, breakeven level, trawl by some indicator, etc.) - this is freelancing to the coder, who made this EA.
There are 26 pages in that thread, but that's because after that I did optimization with claud to find settings for pairs and timeframes.
---------------
If I hadn't found it I would have searched (and users would have suggested it) and there would have been 5 more pages for example.
If I hadn't found two versions of Expert Advisors and hadn't posted them in this thread - I would have gone to Freelance (and not only me - a lot of people would have wanted a profitable Expert Advisor based on the public strategy that was developed in this thread).
---------------
Because what a programmer/coder does is work.
Many people think, for example: "It takes the coder 10 minutes to create a simple Expert Advisor or indicator" ... It may take you 1 minute or half a day, but it's work.
And we have to do everything so that the programmer feels comfortable coding, so that he knows that the strategy is profitable (at least, was profitable at the time of testing).
You cannot just say: "I have a good idea - make me an Expert Advisor for free - it's only five minutes.
Because who will be the author of the Expert Advisor?
After all, not the one whose idea, but the coder. And who wants to code a plum strategy for a "public plum strategy"? No one.
So you have to prepare well and develop and test your strategy before you start any coding.
And you have the first post with just a request, and some not quite readable charts on the second sheet, that is all the work you did at home without showing, and here is just a request ...
---------------
Of course, as many people as there are opinions.
I just described here - how I would open such a thread (and how they do it on the leading English forums).
I have several such branches there in English, and many with EAs in the branches.
And I didn't ask the coders to make free EAs for me. They just saw that the system is profitable, saw the trade, etc., and made them.
Why? Because if I or any other user wants to improve these EAs (for example, to make a martingale as false or true, martingale by a fixed lot, breakeven level, trawl by some indicator, etc.) - this is freelancing to the coder, who made this EA.
There are 26 sheets in that thread, but that's because I did optimization with claud to find settings by pairs and timeframes after that.
---------------
Of course, I was lucky and immediately found a profitable indicator set-up. Had I not found it, I would have searched (and users would have suggested it) and there would have been 5 more sheets for example.
If I hadn't got two versions of EAs and posted them in a branch, I would have gone to Freelance (and not only me - a lot of people would have wanted a profitable EA using the public strategy developed in this branch).
---------------
Because what a programmer/coder does is work.
Many people think, for example: "It takes the coder 10 minutes to create a simple Expert Advisor or indicator" ... It may take you 1 minute or half a day, but it's work.
And we have to do everything so that the programmer feels comfortable coding, so that he knows that the strategy is profitable (at least, was profitable at the time of testing).
You cannot just say: "I have a good idea - make me an Expert Advisor for free - it's only five minutes.
Because who will be the author of the Expert Advisor?
After all, not the one whose idea, but the coder. And who wants to code a plum strategy for a "public plum strategy"? No one.
So you have to prepare well and develop and test your strategy before you start any coding.
And you have the first post with just a request, and some not quite readable charts on the second sheet, meaning you did all the work at home without showing, and here is just a request ...
---------------
Of course, as many people as opinions.
I just described here - how I would open a similar thread (and how they do it on the leading English forums).
Sergey, I understand your point of view.
I just think that instead of checking by hand, it's more correct to check your system as a robot.
I think this way is more correct, because manual trading, as desired, does not cover all encountered situations (unlike the robot). I have tried it many times. I have tested strategies manually, but was repeatedly convinced that I made many mistakes, either interpreting incorrectly or omitting situations or taking unnecessary ones.
That's why I decided to use robots to test everything.
You're right, you should find the right indicator set-up. I want to do exactly that and would be grateful for any tips.
Sergey, I understand your point of view.
I just think that instead of testing by hand, it would be better to test your system as a robot.
I think this way is more correct, because manual trading, as much as one wishes, does not cover all situations encountered (unlike a robot). I have tried it many times. I have tested strategies manually, but was repeatedly convinced that I made many mistakes, either interpreting incorrectly or omitting situations or taking unnecessary ones.
That's why I decided to use robots to test everything.
You're right, you should find the right indicator set-up. I want to do exactly that and would be grateful for any tips.
Of course you can do it that way too.
I just 'used' your question to answer/describe the situation - what (from my point of view) should the user do, if he's not a coder at all and can not (and will not) make any Expert Advisor himself (and they are in the vast majority in trading/frex - that is: you asked me, and I answered many, that may be 'behind the scenes').
Many programmers here are also the developers of their own trading strategies, so your program version is closer to them.
About the indicator set-up, I think that you should start with it - your set-up, charts, indicators, settings, rules, etc. (i.e. - albeit not manual trading, but not just descriptions in words either).
I just think that instead of testing by hand, it's more correct to test your system as a robot.
I think this way is more correct, because manual trading, as much as one wishes, does not cover all situations encountered (unlike a robot). I have tried it many times. I have tested strategies manually, but was repeatedly convinced that I made many mistakes, either interpreting incorrectly or omitting situations or taking unnecessary ones.
That is why I have decided to use robots to test everything.
It is a well-known fact that manual trading is usually more profitable than robot trading. And it is precisely the robot that does not cover all situations, while manual trading is quite capable, up to and including changing strategies as we go along.
I guess if the hand does not do it, the robot will hardly do it either. I have heard of exceptions, but, imho, that's just talk.
And all this with charts, with the entry and exit points marked, by day and each day, etc. It must have taken a week to do that (just one week).
And then?
Then I did some separate posts on how to trade this system (on the rules of trading this system, timeframes, etc.).
And that's all before coding.
And then I made the first variant of EA by this system(here). How many pages are there in your branch? I will soon have the fourth page ... But the source code of my Expert Advisor has been posted on the sixth page. And its second variant is on the 9th page here with corrected indicators here and here.
Have you tried to test it over the whole quotes history?
And all this with charts, with markings of entry and exit points, by day and every day, etc. It probably took me a week to do it (just one week).
And then?
Then I did some separate posts on how to trade this system (on the rules of trading this system, timeframes, etc.).
And that's all before coding.
And then I made the first variant of EA by this system(here). How many pages are there in your branch? I will soon have the fourth page ... But the source code of my Expert Advisor has been posted on the sixth page. And its second variant is on the 9th page here with corrected indicators here and here.
I looked at the code.
What can I say...?
I finished my work on any kind of indicators just on this variant of construction, because I think there is hardly anything better, because it is the edge.
True, I have managed to implement it on one last bar.
Have you tried doing a test on the whole quote history?
Optimise with claud on several timeframes, then backtest over a year with the settings found. All posted in the thread.
True, it was in 2013 ...
I'm not asking for help. I have several such threads in English. If I find some interesting indicator setups - I will open another branch and will do the same (and all in a branch) - as a result an Expert Advisor will come out at the end of a branch (someone will do it for free, or I will order it through Freelance and then I will post it in a branch). And improved version, if I order through Freelance - I will keep it to myself.
This is the starter's topic asking for help (I never asked for it).
By the way, there was no backtesting with real ticks back then, and in my opinion, optimisation with claud was still on 32-bit systems back then.
MetaTrader 5 build 1045 platform update:
So it's buying something like an Irbis TW57 with Windows 10 Pro 64-bit 128Gb and keeping it for optimization and backtesting. And all the public EAs that are in the branches to improve and then find the settings again and backtest.
Because there is also AscTrend EA in two versions(here and here) with optimization and settings (general summary in first post here), and two more branches with two EAs each.
It's the top starter asking for help (I never did).