Can't Find Any Profitable EA on MQL5 in past 5 years. - page 6

 
Trader con experiencia:
there is no profitable EA in the market, I have bought the ones that are at the top of the market and all have produced losses instead of profits, I have come to the conclusion that it is better not to buy Eas here in the market, the more expensive it is the EA, the EA generates more losses, currently I only try those that are free, in which you can test both demo and real

Disclaimer: This all hypothetical.

Let’s say as a developer, I create a data fitted EA that shows stable performance over the last 10 years. I price this EA fairly let’s say $100.
Since I am in the business of selling software, not making money through trading, I start with an investment capital of $1000. I then ask 10 of my friends to buy my EA on the market and post stellar reviews on the product and I reimburse them $100 for their “marketing efforts”.


Now as a novice trader, one will be impressed by the reviews and probably buy my EA for $100. If I can make 100 sales, then I will have made $10000 less 20% MQL fee and $1000 (initial capital).
Remember, I am into selling software, not trading!


My advice is learning to tell the difference between coders who trade and traders who sell code. Sorry to say but the market place is filled with the latter.


You don’t have to be a developer but if you can learn to trade manually and come up with a trading strategy that is consistent, then get a developer to automate it for you. Personally, it took me 6 years and I think close to 200 blown demos and 2 live accounts to start seeing consistency and I still see red months!


As for the market place, it’s a mess and it takes months to sort through all the EAs for a stable one. That said, it can be much easier if you disregard all the backtests and always look for an EA attached to a live account. Next, watch the performance and versioning of the EA during its live run. If you keep seeing updated versions every other month, it could signify “trend fitting”. Lastly try diversifying the risk amongst different trading systems and manual trading if possible.


Just my two cents

 
Catalin Zachiu:
What do you mean by 50% accuracy , of course there are EA's that can make money on forex , but it's all about the risk , the more you risk , the greater the chances to bust your account .
Dont look for EA that seem to have extra high returns. Look for EA with long time span and very low drawdown.
 

I add my modest opinion.

I am not a trader but a developer for hobby here , looking for a good profitable strategy.

I developed EA, some good, some less good, but at the end all are going to fail and I think it is normal.

Holy grail is my target but I dunno if it exist.

of course there are good EA but in some case you must know how it works and eventually activate and deactivate it in some circumstances.

In my case I prefer selling indicators and on my own I try my EA.

 

There are different types of users.

Whether an EA can be made profitable usually boils down to the user that is using it.

If it's profitable, the user usually or most likely knows why it's profitable, what it does, when it does that, why it does that, and what it shouldn't do and when it should not do this or that,

And all these elements are then cooked into one concept, that consists of combined tuned functions with their exact parameters.

Think of a user that does not know anything about the EA he is using, what are the odds he will get it to work properly ?

And on the other hand look at a user that knows exactly what it should do and when and why and for how long and when to stop.

I tend to draw a correlation with a car.

Everybody can buy a car, but to drive it successfully, you need to know how to make it operate successfully.

This is done by practice.

Knowledge is not given or bought, it is acquired.

It's naive to say that nothing works.

Especially if you don't know how to use it. 

And it's also naive to say that nothing works, simply because when something works, it will leave a decent trail throughout history.

And there are many of these trails so it shows that the statement that nothing works, is false.


I can only think of one thing here,

Someone bought some cars over the past 5 years, but never got his drivers licence.

 
Marco Montemari:

I add my modest opinion.

I am not a trader but a developer for hobby here , looking for a good profitable strategy.

I developed EA, some good, some less good, but at the end all are going to fail and I think it is normal.

Holy grail is my target but I dunno if it exist.

of course there are good EA but in some case you must know how it works and eventually activate and deactivate it in some circumstances.

In my case I prefer selling indicators and on my own I try my EA.

are you good in coding?

 
Andreas Bauer:

Actually I have the idea, that mql5 should open an under-category of signals; signals, that are solely based on the EAs here in mql5 marketsplace.

This way sellers would have a platform to demonstrate their EAs and let it compete with other EAs, and we - as buyers - would have that great overview.

that 's true ! 

 
james-smith:
I have never found a EA which can work with 50% accuracy. What's your experience please share with us thanks

reality is crual. 

It's possible to make good profits, if Volume and ticks volume are provided by brokers. 

Even MQL5 got some gunctions to get the correct volume, NEVER, the brokers will let you access you to this kind of information

Also, MQL5 is here to help brokers to make money, not traders

 
I disagree with the spots where all the EAs place the orders. It makes me uncomfortable to be in the market when nothing is really happening.  
 
quargil34:

reality is crual. 

It's possible to make good profits, if Volume and ticks volume are provided by brokers. 

Even MQL5 got some gunctions to get the correct volume, NEVER, the brokers will let you access you to this kind of information

Also, MQL5 is here to help brokers to make money, not traders

Yes buddy, you may have said the most brutal, obvious, and inconvenient truth:

- Metaquotes (a russian company registered in Cyprus) makes its money selling systems for brokers.

Traders are not it's clients - outside this community's market that, of course, isn't its main source of revenue.

God knows if they earn commissions from brokers over the users orders. There's an obvious conflict of interests there, if that's the case.

They control the servers, "our" terminals, the programming language we use to code our EAs (they don't even need to decompile the EAs, since they also control the MetaEditor), they own the field, the ball and the hot dogs too. It's the business model of the century! And the brokers pay them to install everything! Imagine that, you being paid to build your own tower from where you have the most clear vision of the large majority of the retail market. Cool eh?

Now, truth be told, they do not hide this from anyone. The deal is clear for everybody and nobody is forced to sign in.

Most brokers have optional client terminals from other providers (that probably falls in the same arrangement, but...). Everyone should know what kind of fight is getting into. If the rewards are worth the risks, most of us go.

I'm not saying they use this advantage to profit from us retail traders in any unfair manner - they don't even need to! Actually, the best strategy would be to make sure this environment that you control works in the most fairly and trustworthy way. Being trustful it expands, and so your vision of the market.

I really like the Metatrader environment. I really believe that Metaquotes does a great job on it.


Regarding the original question, I suggest you all to think of an EA as an ongoing piece of work. The market conditions change all the time. It evolves and so must the EAs. It must be constantly updated, refactored, adapted to the current market conditions. The core strategy can even turn useless because of a market change. So, any product off the shelf is doomed in the long run. A seller should not advertises it as a perpetual solution and the buyer should know that it will require constant supervision, frequent parameters optimizations and other kinds of work.

There's no money making machine for sale. I mean, a true one. What makes money is work not a device.

 
Marco vd Heijden:

There are different types of users.

Whether an EA can be made profitable usually boils down to the user that is using it.

If it's profitable, the user usually or most likely knows why it's profitable, what it does, when it does that, why it does that, and what it shouldn't do and when it should not do this or that,

And all these elements are then cooked into one concept, that consists of combined tuned functions with their exact parameters.

Think of a user that does not know anything about the EA he is using, what are the odds he will get it to work properly ?

And on the other hand look at a user that knows exactly what it should do and when and why and for how long and when to stop.

I tend to draw a correlation with a car.

Everybody can buy a car, but to drive it successfully, you need to know how to make it operate successfully.

This is done by practice.

Knowledge is not given or bought, it is acquired.

It's naive to say that nothing works.

Especially if you don't know how to use it. 

And it's also naive to say that nothing works, simply because when something works, it will leave a decent trail throughout history.

And there are many of these trails so it shows that the statement that nothing works, is false.


I can only think of one thing here,

Someone bought some cars over the past 5 years, but never got his drivers licence.

Very good explanation and i hope people give it a thought ;)