Can be automated trading be profitable?

It seems that no one wants to pick up that gauntlet..but here's one thing you can use the EA for, micro accounts. And you can use it to test out the EA. Do remember, if a product does not offer a refund, it's not worth wasting your time with it and will, in all probability, turn out to be a scam.
A real product, will be listed in any of the online shopping portals with a 30 day or a 60 day refund. So read the small print before running off to purchase the EA..

when cybernetics become trading champs and all people adopt them,then for the first time P.A will become totally flat...since equals will cancel out each since the market is moved by human emotin-greed and fear and the market game will be over.

I think the fields of Data Mining and Machine Intelligence (both of which contribute tremendously to medical research -- genetics, etc) are often overlooked.

Given time, I think the level of automation will increase. It would seem to me most people/programmers have never heard of Data Mining, Machine Intelligence and other related disciplines let alone actually studied them..

My aim thus far is to have something semi-automated, whereby I receive alerts and trade execution in the end is my decision.

Thank you Benkrah,
I prefer to stick to automated trading for now, but I'm reading and learning stuff about Forex, so soon I'm going to try the "semi" thing :-)

Do you also use the automatic option? For now I'm selecting the strategies with the highest T-Score. I have 3 strategies working at every given time and it seems to work well. (I made more than 500 pips in a week)
Any other parameters you recommend looking at?

Thank you again for you help and have a great week!

Yup! You would have thought that traders would figure this one out but so far, they're determined to hunt down the holy grail. EA, at best, can be used to check strategies but that's about it. And even that, you'll need to be a coder to decode the EA in order to take it apart.

Yes. Unfortunately, no one is going to hand it to you on a plate, so you'll need to look hard for any exploitable behaviour and be able to program and test it.

Just Google 'Sergey Aleynikov' or 'Renaissance Technologies' and you'll see there are people that are consistently profitable with automated trading.

I don't trust automated trading at all. I would much rather sit in front of a computer 12 hours than trust an auto bot. Semi auto is the way to go!

i think if the EA is programming correctly, it can be as profitable. Not easy to get a profitable EA online.

i dont believe with EA i''ve spent thousand of dollar but nothing works, i think better do it manually

Sure it's possible, there are people who have consistent profitable robots trading for them - that's a fact. One example is high frequency trading, which is highly profitable if done right.

But from my experience, it's much much harder to make a robot that works than learning how to trade and profit by yourself. To make a winning robot you probably need experience in analyzing complex systems (using statistics, signal processing, etc...), and then you also need to know how to fine-tune the robot. Then you also need to make sure it's profitable before trading with it, which means you may have to put it to trade some months on demo before even going live (and if it fails, you have to fine-tune it and go through the same again).

Our brain is much more powerful than a computer, so even if you know how to trade it's very hard to put it on a robot. Our brain can make a qualitative analysis very easily, while that may be very hard to program.

I also was under the impression that robots and programs required a person to know coding and to be able to code expertly if they wanted to change and modify a robot?

That automatically counts me out, if it's true. lol

As already mentioned, there's High Frequency Trading, Quantitative Analysis, etc.

I can't remember the exact statistics/markets (i'll try dig it up), but in some markets, automated trading outnumbers discretionary trading by a ration of 3.5 to 1..

Surely automated trading can be performing .. but you need to "teach" your program something like "sense" - if that works, a robot can be much more efficient than a human - that's a matter of fact.

A human brain is far more powerful than a computerchip (because of technical reasons i wont explain further) BUT a computer is more efficient in a smaller range of involved factors and therefore it's much faster to do calculations.

IF a computer had to process as much information as our brain does (visual, emotions, etc) a computer will instantly kill itself - it is not capable of that.

But if you manage to formulate some rules into a programm a computer will be ways faster than a human.

Let's compare the market with the weather (because it's roughly the same in terms of being an chaotic system) - who is better in predicting a weather forecast? It's the human because it can nearly instantly add some variables to his equation of understanding what happens.
A robot lacks this feature and is jailed into his fixed equation.

Just for your interested: Even today NO computer was able to exactly reproduce climate and weather in the past because the equations still lacks of some information.

On the other hand, if you found a mechanical system for yourself which works in, say, 60% of time for a win (where you always! act for your rules - then a robot will make you rich, because it can trade this system 1000s of times per day.

But the thruth is, there is no such system, which will actually work - it may work for some time but because of the chaotic nature of the market parameters will change over time and the robot will fail .. and he wont fail 1 time like you, he will fail 1000s of times per day.

That's the risk which is involved when you have a robot that does nothing more then you will do but faster :-)

Many things will change if someone will someday write the first really self-lerning Artificial Intelligence... :-)

Uh yea that's obvious, you need to know coding to code a robot

Hehe, yeah, I guess I phrased that question quite ignorantly... lol

I believe EAs are still in the beginning stage. Stock marks and forex had been around for years but personal computers as well as trading through personal computers had only been popularized in the past decades. Yes, there had not been a successful EA ever made that had made consistent profit ever since, however, it is too early to tell that EA will not work to drive consistent profit.

I believe a successful trader will adapt to environmental changes which includes technology. When technology creates opportunities for automated trading, people's natural greed will create a demand to exploit such a new opportunity. When the demand grow to a certain stage (which it is growing at the moment), I believe it is possible for automation to beat the market. People was once before, skeptical about online purchases due to security issue. I see this problem is slowly claimed down when people beginning to accept and enjoy online security of transactions.

People can actually build complex artificial intelligence and robots in which the robot infact does its decisions and learnings. What stops us from thinking that one day, forex EAs may incorporate such AI and judge with a certain degree of human emotions into trading automatically?

Simply put, I had a hard time beating the master level of a chess game, not to mention this chess game is free of charge. I believe it will be much harder to beat if the game is selling for millions of dollars.

Bullshit, all I do is create EA's that are successful over the long haul. And they don't have to be complicated. They just have to incorporate market logic.

I don't think, that an EA will ever be able to have a 100% win ratio (over unlimited time) - it is mathematical possible with an unlimited amount of margin .. but that's it.

Because an EA would need to have all paramters necessary to "calculate" the market, which he can never have. As I stated in another thread there are huge factors which you will never know and neither process in a reasonable amount of time.

The strength of an EA is: It can trade permantly. It really doesn't need to be extremly complicated as long as it makes profit.

Let's assume a scenario where a strategy makes a little bit of profit, but it needs the trader to be permanently online. Not possible. An EA can do that with ease, over years.

And that's the thing about it - even the smallest win results in a good amount of money if realised over and over again.

But don't believe, that a computer can predict chaos without having all input that it would need to precalculate them. And it will never have all input.

do forex automated trading systems work? Will you be able to take some of these systems sold on-line and amass a great fortune -or even only a monthly income- during a small amount of time? I have to tell you that the answer to this is no and the proof of this is all around you. I have been in the expert advisor industry since it started almost 4 years ago and I can tell you that I don't know a single person who has been able to achieve to live from an income received from commercial automated trading systems.

There are mainly two reasons why this has been the case. First of all, automated trading systems being sold are not created to be profitable in the long term, they are created to be sold and to create a system to sell is very different than to create a system to actually generate a long term income. You will find that most commercial systems are based on unsound trading tactics showing very little evidence of profitability and market adaptability. A higher danger is that most people are not aware that statistically significant results (years!) are needed to proof any like hood of long term profitability and they rush to buy systems that promise extremely high profits but that in the end deliver little or nothing. These systems which are only able to achieve profits in the short or medium term (results which are statistically meaningless) show their true face in the long term and wipe people's accounts.

The second problem is that most people who come into automated trading are under capitalized and feel that they can make a living starting with something like one or ten thousand US dollars, some expert advisor sellers even claim you can do this from 50 or 100 US dollars. The truth is that in order to achieve this you would need to achieve a long term performance that would be extremely high, something that the market is not going to allow because of the self-limiting character of market inefficiencies. It is no coincidence that the best traders and hedge funds in the world achieve average yearly profits of 20-30% and not higher. The fact is that achieving higher profits puts equity at a very significant risk due to market exposure, something that expert advisor sellers want you to believe does not happen.

Now, I am not saying that automated trading systems do not work. As a matter of fact, I live from trading these systems myself and I can tell you that there are systems which are able to produce money in the long term. These experts are designed with adaptability and capital preservation in mind and produce much less than what your regular EA seller promises you, but they actually do work with a high like hood of surviving in the long term (10-20 years). However automated trading is a difficult task that requires a lot of knowledge and understanding, something you should be prepared to do if you really want to profit in this market.

So in the end, do forex expert advisors work? I don't think you will be able to get from your 500 USD to 1 million in a few years but through adequate investment and the use of expert advisors with sound trading strategies you will most likely be able to achieve a long term performance of 15-25% per year with conservative risk and this in the long term will help you build wealth and diversify your other investments. Does this sound more realistic? Certainly making millions of dollars in a few years from small investments simply cannot be done and a lot of patience and a lot of knowledge and training are required to achieve long term profitability.

Very good post! Exactly what I am thinking as well. The thing is, it depends how much risk you can personally stomach. If you can stomach drawdowns in the realms of 60 - 80% then the returns can increase dramatically, while still staying profitable in the long-term. Problem is that most people can't stomach such drawdowns, that is also why insane returns will stay a theory for most people.