My First EA Build Using FXDreema - Follow Along

Trading Systems Expert Advisors and Automated Trading
1

I’m new here, but thought I’d skip the “Introduce myself” section and jump right in.

First, I’ll give you a bit of background as to how I ended up here:

Two weeks ago I set about comparing the different EA builders on the market. I found 14 of them - from the free to the expensive. After drawing up a table and comparing them side by side as well as watching all their instructional videos I decided that the first one I’d try would be FXDreema.

FXDreema had some great reviews and the EA building interface looked really cool. I also liked the cool visual aids when trying to deal with candles.

At the beginning of this week I decided to read FXDreema’s full set of documentation. In all honesty, it was a bit hard going and by the end I felt I had to dive into the platform for it to really make sense.

On Tuesday I started going through some of the old systems I had on hand that I felt were reputable enough to build into EA’s and easy enough as well.

The first ones I looked at were mainly reliant on support & resistance levels, of which I thought would make it hard going to replicate in an EA. But I’d try anyway.

Then, as luck would have it, an old friend who knew I was about to embark on this journey sent me his own profitable system, one he uses daily that makes about 20 pips and doesn’t have any S&R rules, it’s all based on candle’s and pricing - I figured that it would be pretty easy to code and a great start for me. What a result!

Well, I started using FXDreema on Wednesday (three days ago). I spent a solid 2-3 hours at it and by the end of the night I was addicted, but very frustrated. I couldn’t quite work out how this thing ticked, but had managed to build an EA that worked when I put it in MT4. I was a long way off from completing the system, but it had placed a profitable trade so I was quite happy. After the profitable trade, the EA got confused and did nothing… So I knew I had more work to do.

Here’s where I ended up on day 1:

start.jpg700×118 11.8 KB

Thursday I had a wee breakthrough. I was trying to get the EA to realize the first trade had ended and it was time to put a second one on. About 70 versions later (you have to save it to test it in MT4, it’s the only way to see if it’s working - unless someone here knows of another alternative) I finally stumbled on the solution after jumping on their support forum.

After succeeding, my spirits were high and I was able to start work on other parts of the system. And by the end of the day I had an operational EA that still had a few bugs in it, but I was on the right track and could see the potential. I was also feeling a lot more comfortable with how Dreema worked and was starting to enjoy it!

Here’s where I ended up on day 2:

getting there.jpg700×119 16.3 KB

Friday night I was out at a Christmas party, so no progress on the EA front.

Saturday morning (today), hungover, and feeling like death… I picked myself up wanting to make this thing work and get it profitable.

I was making swift progress and after an hour I felt I was there. I thought I’d added all the components to the system and it should be hitting the 80% win rate my friend had been preaching about. However, after reviewing the trades carefully, I could see that there was still work to be done. After the first trade hits its stop loss it was deleting all the pending orders, which was not part of the system.

Another hour on and I had fixed all the bugs and it had started returning a positive balance… although I was still waiting on some more detail from my friend who would be able to spot any inconsistencies with his strategy.

Here’s my first positive chart on day 3:

firstpositive.jpg700×119 14.1 KB

I then decided to run it through an Optimization test to see if the Inputs needed tweaking, but still making sure I wasn’t curve fitting over such a short timeframe.

And here’s where I landed, after changing the TP’s to 19 pips instead of 20, above:

19 TP.jpg700×117 13.6 KB

So, there we are. My first journey into building an EA has been short (5-6 hours of actual build & test time), relatively painless and successful!

Makes me wonder why I didn’t try this ages ago.

There are a few things still outstanding though:

[ul]
[li]I need more quality back data for further testing (shout out if you can help)
[/li][li]There may be additional feedback from my friend
[/li][/ul]

Leave a reply if you have any questions on FXDreema or the process I went through, happy to help.

1 Like
2

Hi,

Interesting read. If you need help on the tick data, let me know. I can help you with that.

3

Nice to hear about your adventure in writing your first EA. Congratulations!

4

I have used FxDreema for nearly every day, for a year. So ask me if you have any difficulties regarding the logic…
I have built many systems, and most of them fail when testing longer times… So be careful. 23 Successful trades does NEVER prove any ea profitable.
Keep backtesting and optimizing the ea before moving to live test and real money, to avoid the mistakes I made :smiley:

With fxdreema you can also build mt5 eas, and I strongly recommend moving to mt5. Its almost similar to mt4, just faster, more accurate data and better options to fine-tune your ea.

Good luck to you!

1 Like
5

Good for you Cam. Always good to learn new tricks. Will you tell us at some stage what the other EA builders were that you looked at and what made you choose FXDreema?

6

Thanks for alll the support guys!

For some reason my notifications on this post didn’t come through to my email so I missed all the chatter.

5 days on and things have been up and down, to be honest. After chatting to my friend it appears his strategy is to be used on one particular chart at one particular time of the day. So, the testing I did on the charts above proved pointless as it’s not the chart that particular strategy works on.

I ended up buying some tick data to help with more testing, but if you have any Lakoenza let me know what you have and I’ll see how it compares.

Anyway, I was able to test my initial strategy against the EURUSD since 2013 and it failed dismally. I’m not even going to post the chart here as it was quite disappointing.

So, onward and upward with my original plan to create my “own” EA.

This one is based on a strategy that looks for 3 bear candles in a row hitting a support level. You then enter the market with a pending buy trade.

Here’s one of my early attempts, showed promise even though it was breakeven:

week2a.jpg700×128 16.8 KB

This was followed up by numerous tweaks to candle sizing, time filters and the SL to see if I could increase the trade frequency and improve profitability:

week2b.jpg700×128 15.2 KB

Nope, that didn’t work.

I then ended up removing the candle sizing altogether as this seemed to severely restrict the number of trades and instead reduced the TP to 10 pips.

All of a sudden, things started heading in the right direction again.

week2c.jpg700×126 14.5 KB

Now, I know what you’re thinking… what’s with the short time frame if you have tick data… Yeah!! Well, I’m thinking the same thing.

I have a problem with my MT4:

  1. It keeps throwing an OrderSend Error 130 at various points in my strategy test.

Anyone have an ideas on how to fix this?

7

Thanks Lakoenza,

If you have some that would be great!

8

Thanks,

Have you had any success stories?

I’ll have to look into MT5 as MT4 is proving to be a bit of a pain already.

Any tips re FXDreema?

9

Will do! I posted it all on a blog so people can check it out for future reference.

10

Tips… Well, if you are going to backtest your ea a lot, make sure the algorithm in fxdreema is as light as possible - the most restricting conditions should be always first.
http://forums.babypips.com/expert-advisors-automated-trading/69261-these-too-good-test-results-credible.html
^^ This ea is relatively succesfull, at least according to backtests. Its biggest broblem is big sl which makes the profits unpredictable. Currently I am working on another project, it might be even better…

11

5astelija, do you know how to add extra pips to a SL based on a candle’s low, i.e. candle low + 5 pips? In Dreema I can see a pip offset option for the Buy/Sell trade, but not for the stop loss.

Any ideas?

BTW, nice work on your first EA. Yeah, big SL is always a concern but in a way you’re stacking the odds in your favor. Looking forward to seeing your next piece of art!

12

I haven’t done that, at least in long time, but I think its possible with the dynamic stop levels. With the adjust box you can do simple mathematical modifications, like add 5 pips to the default value. See if this works out…

13

Thanks, that worked… “+5pips” seemed to do it.

14

5astelija, Any idea how to modify the TP on an existing pending order?

I’m trying to do it after a negative outcome of a “Check last closed trade profit” block.

Using “Modify stops of trades”.

It’s working fine for the positive outcome, but not for the negative.


Any ideas?

15

The logic looks fine… You have the default setting on block 21, profit is greater than 0 ?
Maybe the block 25 modifies only running trades? You could replace that with “delete pending orders” and attach a “buy/sell pending order” after it.

16

Managed to get it working. Needed “No trade is running” then “For each pending order” then “Modify stop”.

Something on the forum led to me to the answer.

Thanks

17

Gday lads. Just dropping by to say hi and have subscribed to the thread. Have been using compilers for a couple of years and fxdreema just kicks ass. Will help out where i can.

Here’s a link to a thread I started that shows how you can get quality data and 99.9% modeling quality for free 301 Moved Permanently

Just for the record the bot referred to in the thread bummed out back in June when the ass-end fell out of the EURUSD lol

18

Hi Guys,

I had my first go at FXdreema this week, I am excited at the potential. I have not gotten anything to work well yet but this business of watching charts all day is not for me. I would far rather design profitable strategies.

19

programmed in such a way that you cannot ever build a profitable robot! then you might choose their paid course. still you will be disappointed.
Just waste of time & money.
CAM the tread beginner is the seller. BEWARE!

20

@dnv999 FYI, I started the thread about a year before I created the course.

I did it because I worked out all the tricks and thought that I could teach anyone (regardless of knowledge) how to use it to build profitable EA’s. It’s actually not that hard.

But the course took me a lot of time to create and there is a community on the back of it which takes time to manage, and I also help all of my students out when they get stuck.

I built this EA last week, in about 90 mins (mainly tweaking the strategy to see if there was an edge). The actual build can be done in under 5 mins: LRSI - Webinar Giveaway EA | Automate My Trading

Here’s the video that shows how I built it: How To Build A Profitable RSI MT4 EA In 5 Minutes

There are also other profitable EA’s I’ve built in under 45 mins that are up on Forex Verified and have been running there for over 18 months and are still profitable.

I would argue that buying commercial EA’s and investing time in testing/trading them, or having an EA coded for you, or learning MQL is where people will actually waste time and money.

2 Likes