Posts made by OntradingX

For example, I want to draw the line LINE1 and when it appears on the graph, I want to define a number in a constant, for example 4, so the EA must draw 4 lines above it in the sequence, LINE2, LINE3, LINE4

@jstap The problem I face is that I want to find a market closing time that always happens every day and this is proving difficult, I need to close orders and stop some rules always before the market closes and this is complex to do if During my backtests I see that I set my orders to close at 8:00 pm but for some reason the broker closed the market at 7:50 pm. every time this happens the ea is lost because if the market is not open the order closing or variable reset commands are not executed.

Hello friends, as you can see in the image the broker informs that the trading hours are from 01:05 until 23:55, but in practice in backtests this does not work well for most of them, because if we return a period longer than let's say 3 years some of them start to change these times to 00:00 until 21:00 or even to 02:00 or 03:00 so when I configure my filter to have fixed times the nightmares happen, for example when finalizing negotiations at end of the day, if I set it to end at 23:55 and it's a day where trading ended at 20:00, this harms the results of the strategies.
I would like to know how you deal with this time filter issue. Can you please explain this so I know if I can use some ideas? My main idea is to identify a common time pattern that occurs over a long period of time for both opening trades and closing trades.

@l-andorrà I understand, the steps I mentioned really helped me to solve slow problems when optimizing robots, before when I didn't create a well-elaborated logic thinking from the beginning about future optimizations, when I was going to perform optimization, the same thing happened to me, to perform the tasks. it was an eternity.
But it is difficult to help you without at least having access to what you are actually doing, perhaps something in the middle of your project is causing this problem, my suggestion is that you review your entire project or create a new project and copy and by pasting parts of the blocks and testing to see when this is happening, perhaps this is a way of locating at which point you can apply a logic change to your EA.

@jstap said in My first MT5 optimization on fxDreema...and the fisrt issue!:

basic rule only have the ea doing something when needed

I like this " basic rule only have the ea doing something when needed " great words

Look at this project I made recently, it has a lot of blocks and is super fast in backtests

@l-andorrà I believe that Fxdreema also contributes to the slowness, but on the other hand I think that the logic involved in the projects also causes the slowness, for a while now I started to optimize my EA's which are huge in number of blocks and the adapt my strategies to open orders when (closing) the candles using the block (once per bar) and I realized that this, in addition to helping with optimization so that it becomes faster, improved my entire logic, currently most of my EA's I focus on Always look at the bar closing to avoid frustration with slowness.

@sajadhmz-vp-phd Adjusted below

https://fxdreema.com/shared/LSAlVglzc <--- click here test and return feedback for me.

@sajadhmz-vp-phd
Use this

create one variable per condição e salve.

@l-andorrà
You can share the Settings tab screen to see more information, for example:

1. What model are you using?
2. Which period is optimizing.
3. How many inputs are optimizing and how many steps are being consumed.

All of this makes a huge difference when optimizing because the longer the time and the greater the amount of data to be optimized, the longer the processing time will be for it to find results.

The MT5 optimizer works very well, but there are many steps to be analyzed internally in MT5 and also in your EA in terms of parameters.

I believe that most people develop EA's and want to test in OHLC, but the correct thing is to test in Each tick based on a real tick, but it is also necessary to analyze in your EA where to place the blocks (once per bar), currently I use it in practically every my EA and all my strategies this block and only in the management part I do not allow the use of it, as it is necessary to check (at each tick) trailings and money information, so what I want to say briefly is that all its logic it has to be well thought out so as not to overload optimization through MT5.

Another very important detail is that it is necessary to clean internal data from the Terminal and Tester folders, internally there is data that is saved from previous optimizations or from symbols that were downloaded during previous backtests and this consumes space on your disk. (Optimization takes up disk space).

Use Folder Size program free to scan

What I noticed is that in the real market, the field, after being changed, assumes what was written and passes it on to the variable, but in the back, the field, after being changed, erases the entered content, returning to the previous content, preventing the variable from receiving the updated content. , it seems that the problem is in this block or mt5 in backtest environment does not allow data to be changed.

@jstap I tried and it doesn't work. I believe it is either a limitation of MT5 or a bug in this block of editable fields.

@jstap I tried to do exactly as you said, it still doesn't work in a backtest environment.
I even generated a new project for testing.
Can you share your project so I can understand what you are doing to make this work?

New project --> https://fxdreema.com/shared/KPW4JrvPd

@jstap I didn't understand what you meant.
I tested it by just putting it (on tick) and it doesn't work in the backtest environment, nor in a real environment, and I know it works perfectly, and I just want it to happen in the (backtest) environment, maybe by chance it moves and leaves it just in (ontick ). Did it work in a backtest environment?

Hello, I created this project below to try to change data in an editable field in a backtest environment but I was unsuccessful. Can anyone help me solve it?
Note: In the real market it works perfectly.

Project --> https://fxdreema.com/shared/RabJuHL0e

@jstap
I managed to do it using the trading bucket. Thanks for your help friend

As shown in the image below, I would like to know if anyone has any idea how I can easily locate the opening price of the order that is highest and the order that is lowest. Can anyone give me an example of how to do this, I want to save these 2 prices in a variable but I can't find the best way to locate them

Hello friends, I come here to share a little of my experience on Good practices for carrying out EA Optimizations made in Fxdreema.

1. Before you want to believe that your strategy is winning and develop an EA that allows you to actually test it by strictly following all the entry, exit and management rules in (real account).

2. Test the effectiveness of your strategy in trending, lateral and high volatility markets especially during news of 2 and 3 bulls.
   Always test your EA's in an environment that is as close to reality, there is no point in fooling yourself with OHLC tests to accelerate the results, (with each real tick) it is most of the time what comes closest to the real market, so if To optimize your EA, use this mode even if it is slow and time-consuming, because if you find good results there is a huge chance of replicating the same result in a real account in the future.
   Choose Mixed periods (Covid, Wars, Flash Crash) to carry out backtests of your EA's, because if you seek to profit in the long term, your strategy must be able to withstand these periods.
   
   

3. Establish some gain and loss management standards, many people choose to think that a strategy with TP 2x greater than SL fits most markets, but in reality I realize that this can all be a mistake, in good times leverage and In bad times, close negotiations as quickly as possible, preserving capital means staying alive in the financial market.
   Dilute your Trading Capital into % that will make it last longer instead of risking all your capital in a few trades, be organized in establishing precise numbers of daily or even monthly trades including balanced profit and loss limits.

4. Be simple and objective in your ideas, too many analyzes (filters) can make you miss countless opportunities.

5. Don't sabotage yourself, many EA's that I see in MQL are clearly showing perfection in the backtest curvature due to the famous (Overfit) or skipping specific dates (days) where the downgrade happened, I suggest not getting excited about seeing EA backtest ´s from third parties because only they know what they did within the coding.

6. Many indicators (if not most of them) are delayed and some were even coded on purpose to appear profitable, so know how to analyze the exact signal trigger moment given by them so that you are not fooled by the (visual) but by the possible result it can give you. Why not develop your own calculation or market vision and replicate it directly in your EA? I suggest not getting attached to any miraculous indicator without first having proof of real gain from at least the developer, otherwise they are just indicators like many others out there.

Just like you, I've been in Fighting for a few years and increasingly maturing in this market.
If you disagree or agree, I would like to know what your opinion is on this text.
Below I share a video talking more about the items I mentioned above.

Youtube Vídeo --> https://youtu.be/tkOw7TPOehk?si=msVejlRra7YCHTXx

Thanks All.

@DragonZueloTrends
My biggest suggestion is that you stop looking (ready-made things) and do your own study, or else... That you at least test the signals provided by these indicators on a real account and be able to check if it really is profitable (operating in real account) because backtests using mt5, mt4 or the human eye are very misleading as they do not have the same conditions as brokers. Most indicators are lagging and give false signals or repaint the signals to make it look like they are profitable, pay attention to these small details that save you time and money analyzing them.

@DragonZueloTrends Hello friend, I'm going to give you a golden tip.
It will be very difficult to do something (like) what someone did without having access to the (source code).
Try to find a free SMC indicator or at least one that can have the source code, and by reading the source code it is possible to check what type of calculations, periods the indicator is based on to provide the bos and choch notes. The .ex5 file or just images will not help you achieve your goal.
As our friend Roar mentioned, it is subjective, each person understands what a BOS is and what a CHOCH is in a different way, so perhaps creating your own study to define where such detail will appear would be the best thing ever. taking something ready-made that will probably repaint or otherwise be detrimental to your trading plan.