My idea how to use EAs | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Anybody use Bollinger Bands that use fibonacci ratios?? 15 replies

What indicators do Hedge Funds use or do they even use one? 12 replies

One MT4, 2 EAs 2 replies

NewsTracker to control EAs around newstime 4 replies

EAs and losing connection 3 replies

My idea how to use EAs

  • #1
  • First Post: Jun 28, 2011 1:52am Jun 28, 2011 1:52am
Everyone knows two things: 1. EAs don't work forever. 2. They can be profitable during a short period of time. So why no to use EA this way: you buy new trading system and use it until it pays itself 4-5 times, then you stop using it no matter what! No matter what, it make more... you are 100% sure that sooner or later it will fail. This is some thing like play in casino, you should stop before you lost everything.

I think this concept can change the understanding of EAs. There are many good EAs out there, but they good for limited time.

Buy - make money - put it on the shelf!
  • #2
  • Jun 28, 2011 1:59am Jun 28, 2011 1:59am
Uh, what guarantee is there that the EA doesn't stop working the first time you use it? Or any trading "system" for that matter....

In fact, you can draw down infinitely from the first trade you take.
  • #3
  • Jun 28, 2011 2:33am Jun 28, 2011 2:33am
Quoting tdion
Uh, what guarantee is there that the EA doesn't stop working the first time you use it? Or any trading "system" for that matter....

In fact, you can draw down infinitely from the first trade you take.
In my personal experience, trusted EAs make money for some time...
  • #4
  • Jun 28, 2011 3:32am Jun 28, 2011 3:32am
to my idea, i think the best thing to do is to find/create EA's which are not so much trading or make very low profit, and then combine all these "low performers" in 1 EA. It might do the trick...
  • #5
  • Jun 28, 2011 3:43am Jun 28, 2011 3:43am
Quoting ido370
to my idea, i think the best thing to do is to find/create EA's which are not so much trading or make very low profit, and then combine all these "low performers" in 1 EA. It might do the trick...
This might work.
  • #6
  • Jul 3, 2011 10:11am Jul 3, 2011 10:11am
I think that good idea is to progress in one direction and develop own EA. The if market changes, just update the EA and improve it.
  • #7
  • Jul 3, 2011 10:43am Jul 3, 2011 10:43am
I just find one thing why EA can make money. Because their users implement their good mm with high discipline. Without a good mm and discipline, EA's validity is just a temporary.

Let's think how EA will solve an unlimited dimensional space (random market) with just a limited dimensional tool (parameters).

It's looked like how the fed used to try solving the multidimensional economic crisis by using just two dimensional instruments (increase/decrease rate)
  • #8
  • Jul 3, 2011 10:51am Jul 3, 2011 10:51am
Why not just use the EA to effect an entry? A computer cannot EVER think faster than a human. YOU decide when is the best time to deploy the EA, on a hourly chart to make your entry and then manage the trade from there. I've been using this approach for the last 6 months and find it very profitable. Wait for a trend to develop on the day chart then deploy an EA to enter. It calculates my position size automatically and places the stop & limit orders accordingly. Then I just manage the trade by watching the same daily chart for an exit, and or move my stops up to protect the position.

In my experience EA's fail more because of bad exits and brainless management, not poor entries.

My way just checks the charts once every day. Then gat back to real life.
  • #9
  • Jul 3, 2011 11:03am Jul 3, 2011 11:03am
Very Good Point,

The Right tool at the right man will leverage Efficiency and Effectiveness.

Quoting gspajon
Why not just use the EA to effect an entry? A computer cannot EVER think faster than a human. YOU decide when is the best time to deploy the EA, on a hourly chart to make your entry and then manage the trade from there. I've been using this approach for the last 6 months and find it very profitable. Wait for a trend to develop on the day chart then deploy an EA to enter. It calculates my position size automatically and places the stop & limit orders accordingly. Then I just manage the trade by watching the same daily chart for an exit, and...
  • #10
  • Edited 11:41am Jul 3, 2011 11:26am | Edited 11:41am
There are certain inconsistencies in the market that I find hard to believe that an EA system can detect. It takes thousands of hours of screen time to figure out what those inconsistencies are. The people who build EAs don't know what those inconsistencies are and that is why their EA only works for a short while or why it produces mixed results. IMO, people create EAs with the intent to sell them to an uninformed/lazy public. If there is a consistently profitable EA out there, you probably won't hear much about it in any public forum. My advice would be to put away the EAs and put in some screen time.
  • #11
  • Jul 3, 2011 12:59pm Jul 3, 2011 12:59pm
Quoting josef
Everyone knows two things: 1. EAs don't work forever. 2. They can be profitable during a short period of time. So why no to use EA this way: you buy new trading system and use it until it pays itself 4-5 times, then you stop using it no matter what! No matter what, it make more... you are 100% sure that sooner or later it will fail. This is some thing like play in casino, you should stop before you lost everything.
Interesting idea but it doesn't sound to me like breaking bank in a casino once and never gambling again. It sounds more like breaking bank in a casino one night and then on the next night taking the money to another casino and hoping for the same outcome.
I really do not see this working in a long term - and that is after all what trading is all about, right?
I think that if you keep using EAs indiscriminately you will eventually stumble upon an EA, which is completely unsuited for current market conditions and it will clear up your account in no time.
I can't imagine giving my money to an EA without knowing how it operates and checking whether it is design for the market I want to apply it to.
May be an idea for you - to assemble a set of EAs, which you know very well, you know how they work and you know what you can expect of them. Then you can use each EA in a market, in which it will manage the best - whether it is trading or ranging or any other market.
  • #12
  • Jul 4, 2011 1:03am Jul 4, 2011 1:03am
Quoting Giansar
Interesting idea but it doesn't sound to me like breaking bank in a casino once and never gambling again. It sounds more like breaking bank in a casino one night and then on the next night taking the money to another casino and hoping for the same outcome.
I really do not see this working in a long term - and that is after all what trading is all about, right?
I think that if you keep using EAs indiscriminately you will eventually stumble upon an EA, which is completely unsuited for current market conditions and it will clear up your account in...

Right now in the industry there are 10-15 working profitable EAs, the rest is already old or simple was made just for selling, not for making money.

And these 10-15 EAs will not run forever, so it's time to trade and time to stop trading.
  • #13
  • Jul 4, 2011 3:15am Jul 4, 2011 3:15am
I believe in EAs. I think that one of the main reasons why EAs don't work for so much time is that the creators usually code ideas based in short term experiences. I mean, they dont create an EA based in concepts, but create an EA base in what they are seeing in the market in that moment. I dont know if I explained myself here, but if someone creates an EA based in concepts like TRENDS, FIBONACCI, Suport and Resistance, Elliot Wave theory, price action and avoid the news time, i can't see why an EA won't work in the long run.
I think that the main reason is because people create EAs based in the wrong concepts, but there are some concepts that will never (NEVER) change. For example the idea of TRENDS will never change. While the idea of trends is in place, so will the idea of Fibonacci. I think these concepts work because they are based in the core secret(it's not really a secret) of the market that is Supply and Demand!
Sorry about my English, but I think you understood!
  • #14
  • Jul 4, 2011 3:25am Jul 4, 2011 3:25am
I believe that a profitable EA is designed to work only in some market conditions. If those market conditions are not in place, then the EA should take no trades at all. Its impossible to code all the market conditions in one single EA. For example, I am a trend trader, so I think about coding an EA to work only when there is a trend in place. So, in my opinion, the best EAs dont trade every day and if this EA is really based in concepts, it could be applied to any pair. Now, I dont thing its possible for an EA to trade every single day and still be profitable.
  • #15
  • Jul 4, 2011 3:39am Jul 4, 2011 3:39am
Oh yeah? 10-15? How do you know that? Just curious where you get that figure, other than out of thin air.

Quoting josef
Right now in the industry there are 10-15 working profitable EAs, the rest is already old or simple was made just for selling, not for making money.

And these 10-15 EAs will not run forever, so it's time to trade and time to stop trading.
  • #16
  • Edited 3:57am Jul 4, 2011 3:45am | Edited 3:57am
Every single breakout, there is no way to know whether it will retrace or continue breaking out.

So, that is the risk.

Either you bet for the breakout, or against it. Either you bet that a retracement will continue in the direction of the major trend, or that it is a new top or bottom.

And you don't know when a trend will turn into consolidation, or vice versa.

The only thing that you can really do is get lucky with a string of winning bets.... all because the market gave the conditions you bet on.

These are the irrefutable facts that determine that trading is gambling. Sometimes it's nice to just look at the facts for a minute.

Quoting BrazilPips
I believe that a profitable EA is designed to work only in some market conditions. If those market conditions are not in place, then the EA should take no trades at all. Its impossible to code all the market conditions in one single EA. For example, I am a trend trader, so I think about coding an EA to work only when there is a trend in place. So, in my opinion, the best EAs dont trade every day and if this EA is really based in concepts, it could be applied to any pair. Now, I dont thing its possible for an EA to trade every single day and still...
  • #17
  • Edited 4:14am Jul 4, 2011 3:53am | Edited 4:14am
Knowing nothing of this subject, and talking shallowly,
and given you had an EA to manage entries and exits:

If you had a bunch of EA's, each based on a different approach
-- Elliott Wave, stochastics scalping, cog channel, MACD trend trading, etc./whatever --

would you not need an EA to tell you which of these EA's to use at the proper time?

...or this would be compounding too much risk, perhaps, as the last poster seemed to say.


correction: Oh wait, this idea'd be like one of those traffic light indicators for multiple indicators: All lights go, then trade.
  • #18
  • Jul 4, 2011 3:55am Jul 4, 2011 3:55am
Quoting tdion
Oh yeah? 10-15? How do you know that? Just curious where you get that figure, other than out of thin air.
My Air: personal experience, reading forums, following the threads about EAs, following the results of other forum members, doing my own researches. Don't trust review sites too much.
  • #19
  • Jul 4, 2011 3:07pm Jul 4, 2011 3:07pm
You can't know whether the breakout will be a real one or just a fake breakout, but I still believe that trading is more about money management than any other think. You don't need to be lucky to be profitable in the market. You need to have a profitable trading system and a very very strong money management. Now, why do I still believe in EAs? Because I think that it's possible to create an EA with a specific trading system, to work only in a specific market condition and a very strong money management. I didn't find such EA yet, mainly because if I created an EA like that, I wouldn't tell anyone. But I am in the path of creating it...in fact I already have the specific EA, I onlye have to implement the money management.

Good Luck


Quoting tdion
Every single breakout, there is no way to know whether it will retrace or continue breaking out.

So, that is the risk.

Either you bet for the breakout, or against it. Either you bet that a retracement will continue in the direction of the major trend, or that it is a new top or bottom.

And you don't know when a trend will turn into consolidation, or vice versa.

The only thing that you can really do is get lucky with a string of winning bets.... all because the market gave the conditions you bet on.

These are the irrefutable facts that...
  • #20
  • Jul 6, 2011 10:29pm Jul 6, 2011 10:29pm
How did you get this number?
0 traders viewing now