Best posts made by fxDreema

These functions are also very easy to use - AccountEquity() and AccountBalance() - so you can use them directly in some input field. For example AccountBalance() * 0.01 or AccountBalance() * 1 / 100

@roar This will output OrderProfit() as a string. Better use the other "Text" option:

I'm not sure if the WebRequest() function in MQL supports HTTPS. In the documentation they used the word HTTP only. The problem is that if you make this with HTTP, I can easily make a redirect to my local server and respond to any requests from the EA in any way I want. It's not very secure. With HTTPS not only the messages are encrypted, but also the identity of the destination server is checked.

Otherwise if you don't know what GET is, this is for example when you request the website from the address bar. Key-Value pairs can be send in this format: http://example.com?key1=value1&key2=value2&key3=value3. The whole request is in this address string and there is some limit in the length of the string. POST is another method, it's a little bit different. All web servers can handle GET or POST requests.

No, no, no... don't add one of these "inp..." things in Constants. And don't use == when you compare prices or money. Use anything else, but not ==.

10%, but 10% of what? Balance is the amount of money you had the last time you had 0 opened orders. Close all opened orders and the Balance is equal to Equity now. Do you really want to compare Balance with Equity?

10% sounds to me like such a small value that you should not reach it so quickly. If Equity can be less than 10% of the Balance, this EA is not good from the start.

If you mean 10% from some initial value... for example you can deposit 1000 dollars and you want to close everything when you reach 100 dollars, even if this is 6 months in the future... then the problem is that the EA can't really know what is that initial value that you had in mind half a year ago. There are ways to remember values for a while, but in this case maybe it's better to hard code the value in the EA

Otherwise you can try something like this:

Remember that here Equity and Balance are those global values that are result from all trades.

Or this one:

Here this block calculates the profit from the running orders only, it doesn't care about closed orders. AccountBalance() gives you the balance. This is function from MQL4. There is also AccountEquity() and other functions.
So, let's say the we have initial deposit of 1000 dollars. This is the Balance. Then we start losing and the block calculates that out profit is negative. The block would pass if the profit is < 20, because -1 * 1000 * (0.02) is 20. I made my tests and it was easier for me to close the trades when the loss was below 98% of the balance, that's why I used 98 there.
Note that the values used there have .0 at the end. Apparently, if you do it like this (100 - 98 / 100), the calculation will fail, because in MQL4 these numbers are integers and the final result is also integer. The result can't be 0.02, it will be 0. To get the result as floating point number, I wrote these numbers as floating numbers.

@l-andorrà It works with indicators, but I noticed that for some of them there are certain differences that I don't know how to manage.

iBandsOnArray()... I don't remember why I skipped this function. I see that I never wrote any code for it. From all "...OnArray()" functions I only wrote code for "iMAOnArray()". I remember that this one was probably the hardest function for me to write for MQL5 and I can even see that now I have this comment in it for MODE_LWMA:

// TODO lazy fix - I was too tired and lazy to do this in the best way

So to be honest, I don't feel that I really want to make these OnArray() functions right now Knowing myself, I will choose to do something else that I consider more important or easier for me to do

In general any generated EA will be slower that the same strategy if you are able to carefully code it manually. This is because the generated EA is full with useless code. My EA builder is not so clever to only include the code that is needed and to optimize it in the best way, so you always end up with bloated slow code. But it should not be too slow... like waiting for hours to backtest one day. It is important what blocks are you running on every tick. Try to connect them in such a way, so the slowest block are running more rarely. Of course, you can't always say which blocks are slower than others... even I can't tell that without testing them. But maybe you are using some custom indicator - maybe this indicator is poorly written and it is too slow. Put that indicator in few blocks and the time stops

The Condition block takes 2 values and compares them. If you know what values you want to compare, select them on the left and the right side of the block.

There is one for the profit in "Account", it's called "Profit (Equity - Balance)" and obviously it gives you the difference between equity and balance. Balance is ouf course the last known money you had the last time when you had 0 opened trades.

Otherwise you can check how much profit was made out of the currently opened trades with "Check profit (unrealized)". Also there are those Bucket blocks whete you can do the same in a different way... depends on what value you want to check.

I tried this one and it works for me: https://fxdreema.com/shared/VMwxP2Odd
But I manually change the symbol by dragging different symbol over the chart.

I found another problem, that if I use "Market Properties -> Name: Symbol", only the initial symbol name is printed. But otherwise the button click is detected.

Do you see some error messages or for some reason the button name is changed?

Hi all. I was also thinking about many possible features through the years, but I guess I really hit some ceilings. The main problem is lack of motivation to do anything. The website generates good profit for my country and even if I want more (as any person in the wold), I don't need more profits. There is a difference between want and need here, and the difference is I guess in the desire to make more

Unfortunately (or not), I don't see competition from other websites... not that I'm looking if there is any. I'm saying this because I was thinking that something like this can motivate me, it's good to have some competition. But if I don't see any, the other trap is that I think that the website is the best it can be and there is not real need to do more. Also, adding more features means more features to be maintained in the future, to which my brain automatically says "no".

So I don't know really. I also prefer to be motivated and do more, but the psychology is stronger than that. Otherwise I'm doing stuff in the background. The database as it is now is not constructed very well and I'm working on a new way to write the data on it. This is what I'm mostly doing - bettering the code of the website itself. But even this I'm doing slowly.

This error happens in part of the code that is responsible for detecting the Trade events. I recently rewrote that code and I'm more confident with the new one. I will update it very soon

From what I understand, once the yellow MA is below the red MA, maximum 3 trades are allowed, no matter for how long the yellow line is below the red line. Then, when the yellow line goes above, everything resets and we are waiting for it to go below the red line again.

Probably the best way to do this is when the yellow line is below, to count for how many bars it was below. While doing that, also count how many trades were closed in that whole period. If they are less than 3 - continue and make a new one. But thе EA Builder is not very good at making such loops. There is a block for making a loop (basically repeating itself multiple times), but there is no block for breaking this loop (to stop it), which is needed. Otherwise there is a block "(loop) break", but it works only for those "For each..." blocks right now.

Here is another idea: https://fxdreema.com/shared/OpEWBBHjc
Here I detect the cross with blocks 1 and 2 and I create an arrow with name SellSellSell (block 3). Additionaly I create one vertical line, only to then see where crosses happened. If there is no trade I search for that object SellSellSell. If there is none, nothing else will happen. But if there is one, I count how many closed trades were closed after the object (blocks 6 and 7). The object's Time property is used for that (in block 6). And if less than 3 trades were closed, I can create a new one (blocks 7, 8 and 9). Block 8 is whatever the extra condition needs to be, I chose something simple for the example. Otherwise I delete the SellSellSell object (the link between 7 and 10). Technically I can Not delete it here, but I guess this could help a little bit with the speed of testing. I also delete the object when the fast MA is > the slow MA (blocks 11 and 10). I can use crossover for this as well, but imagine that the EA is restarted... I just want to be sure.

Well, I found that unfortunately the example above is very slow. I always try to find a solution that can work with some kind of permanent memory - to use the trades themselves, maybe some objects. But the results are often very slow. Here is another idea that uses a single Variable, which is a temporary memory and that EA would reset if restarted, but the backtesting is much faster: https://fxdreema.com/shared/yqieqWzAb

The idea is that the variable keeps the number of opened trades - when a new trade is opened, the variable increments. And the variable is reset back to 0 when the fast MA is above the slow MA.

Even if it's possible to get the values from the horizontal line and the MA, how can you compare these values? I mean, visually they appear to be crossing, but I think this is only the way it looks on that particular subwindow with the particular zoom. If you resize the window or zoom a little bit, I think the MA line will move and the cross would appear somewhere else. So how can you detect such cross when the cross appears only visually. I mean, these are pixels on the chart that MetaTrader draws, but they cross only visually, not mathematically.

Otherwise it could be possible to compare two indicators with different values, but if you can put limits on them. For example you have indicator that moves between 0 and 100 and another one that moves between 200 and 500. You can adapt the values, so that 0 relates to 200 and 100 relates to 500, and any value between 0 and 100 relates to some very particular value between 200 and 500. But how relatable CCI and MA are? CCI value 120 relates to what value of MA? If you can solve this problem, then you can compare both indicators. Here is some explanation for how CCI works: https://www.metatrader4.com/en/trading-platform/help/analytics/tech_indicators/commodity_channel_index
Obviously CCI depends on MA values... but MA with other periods... I don't know, I'm very bad in mathematics, for me it looks unsolvable

The names of the blocks is not a problem. In MT4 where you see the word "trade", in MT5 you see "position", but this is because I just left it that way.

Each block has another hidden name, let's call it ID. Almost all blocks from MT4 have their MT5 variant. There are few differences, I thin in the blocks from "on Trade". Also, I think in MT5 there are 1-2 extra indicators. Maybe some parameters are also different here and there, but I forgot what it was.

You can download .mq4 file, rename it to .mq5 and import it. This should create MT5 version of the project, but the block will be named as they were in the MT4 project. And just test how it works.

Of course, there is one huge difference between MT4 and MT5 - if you have Netting account in MT5, it works in a different way, you can't have multiple "trades" in the same symbol at the same time. This changes the whole logic completely. Any equality between MT4 and MT5 is possible when you have Hedging type of account in MT5.

@asclepius67, you are "old" user with some advantage, but there are "older" users

To the really poor people I always say to NOT try their luck in Forex. I'm actually very sceptical about the whole trading thing. What is also very risky, but at least more fun, is crypto. So for anyone that has very little money, I would say - stay away from the world of trading, but if you still want to gamble them, better buy some crypto.

This Skrill always annoys me. I opened their website, quickly clicked on the button to register, registered, and then I found out that this profile is completely NOT business. Not only that, but I can't upgrade it to business. Not only that, but I couldn't delete it (some error or something). Not only that, but I couldn't send a message to the support (again some error). Very disappointing.

Damn, your project looks very professional. This is extremely rare sight. But unfortunately the Group thing is very simple, I only programmed it to accept multiple groups separated with "," (thinking that this was advanced). And 30 groups... I never even imagined something like this. I'm afraid to ask why

By the way, as someone with obvious programming knowledge, maybe you can share what in the EA builder is bad, what must be done, or even what is good.

@uriel-0 said in MACD Backtest:

MACD True.mq4

Put some "Draw Arrow" or similar block below block 1, then 2, then 24 and so on... and observe how rare the arrow starts to become. After the crossover block (20) the signals that I get are already rare, combine this Bulls in a row and maybe the chance to get a signal is almost impossible.

Maybe you don't understand the crossover block. It could only pass in the span of 1 candle where the crossover happens. If you want to check when it passes, try only this block with some "Draw Arrow".

Thank you all for the good works. I don't think I deserve this, but thank you. To get a little bit more hate I had an idea - to replace the Testimonials on the front page with fresh ones, more like reviews, where people can write the positives and also the negatives.

@OntradingX, nice to see that you have benefits. I never learned to manage the risk and I'm sooo bad at this, maybe I should play with this tool more

@AlphaOmega, I never thought about this and I don't remember anyone telling me anything about the non-members. I'm not very convinced that asking someone else to generate the EA is any advantage. Making a whole project without being able to test it multiple times in a minute, I think it's almost impossible. But I kinda like the idea of maybe making some "club" in the forum for members only.

@andrewfrank, yes, these guys are fantastic. I don't even understand what kind of motivation do they have to answer other people's questions (well, except @l-andorrà).

Or, if the EA will only be used to work with manual trades, instead of setting this in every block, just set the magic number to 0 globally from here:

This is actually that strange MagicStart input of the EA.

Put "Draw Arrow" afrer block 2 - run the EA and see where or if there are arrows on the chart. Then disconnect that "Draw Arrow" and connect it after block 3. Again, run the EA and observe. Do this for block 12 and 8 as well. Eventually you will probably find after which block the arrows doesn't appear.

By the way I'm not very sure with Monthly and Weekly timeframes. These timeframes don't exist in the Tester. I hope the blocks are working with them properly, but better check it (in the way I explained above).