Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Trading is a mind game - by Forex Mind Games 2 replies

Can an expert advisor run with a Screensaver on? 4 replies

Expert Advisors or Human Traders? 8 replies

How can we set stop loss on Expert Advisor? 0 replies

Can someone create for me an easy Expert Advisor? 0 replies

No Expert Advisor can replace human mind

  • Post #1
  • Quote
  • First Post: Jun 24, 2011 11:15am Jun 24, 2011 11:15am
No Expert Advisor can replace human mind.
This is my belief after many years of dealing with market.

There were times when I was winning whole week and was about to increase
lots but next week I was losing. Why? Because market changed its tactics because people got used to it and market needs their money. Human mind can adapt to it but no EA can do this.

I feel sorry for people who chase for 'profitable' EAs for they waste their time.
So, what to do? Get an EA that gives custom signals to cell phone and then human mind should decide what to do. Yes, persistent people gets constant improvement.

The market is created by human mind and must be understood only by human mind.
Any thoughts on this? Please share.
  • Post #2
  • Quote
  • Edited 12:33pm Jun 24, 2011 11:40am | Edited 12:33pm
Quoting Expert44
No Expert Advisor can replace human mind.
This is my belief after many years of dealing with market.

There were times when I was winning whole week and was about to increase
lots but next week I was losing. Why? Because market changed its tactics because people got used to it and market needs their money. Human mind can adapt to it but no EA can do this.

I feel sorry for people who chase for 'profitable' EAs for they waste their time.
So, what to do? Get an EA that gives custom signals to cell phone and then human mind should decide what to do....
Hello Expert44,

I completely agree that EA's in the long run can not out perform the human mind.

However, EA's are useful in making life easier for traders in only managing trades after the decision has been made by a human to execute a trade. Thus I am a big proponent of money management EA's that chaperon trades while you are away from your machine. There many good ones online and I highly recommend, particularly new traders< to evaluate this concept in place of looking for auto trade EA's

Cheers,
Hello Dr. Pip,
Yes, other than signals, EAs manage trades BUT the final decision to make a trade should be done by human. On the other side if a person trades in fast and volatile market then EA can make a trade BUT it should be checked immediately by the trader for possibility to exit or adjust stop loss.

I believe that real traders work with signal/managing trades EAs and update them constantly AND trade manually.
Quoting Expert44
Hello Dr. Pip,
Yes, other than signals, EAs manage trades BUT the final decision to make a trade should be done by human. On the other side if a person trades in fast and volatile market then EA can make a trade BUT it should be checked immediately by the trader for possibility to exit or adjust stop loss.

I believe that real traders work with signal/managing trades EAs and update them constantly AND trade manually.
Agreed!
Yes, i am also of the opinion that no EA can replace the human mind. we can see, observe read whereas most EA cant.

Nevertheless, EA's also have some key advantages over the human mind.

 

  1. They completely remove the psych aspect of trading especially when trading live.
  2. They can implement as many filters as the programmer can include to filter bad trades while we humans cant. This is principally because the more indicators we use, the more concentration and focus we need. making us easily exhausted and confused.
  3. EA backtesting allows for quick assessment (provided it was done correctly)
  4. EA's can easily manage several trades at once. whereas humans cant



BTW i have been observing most of the traders i have interact with that are seemingly successful are Programmers and most probably use EA's.

As a result of the time constraint i have and the observation stated above, i have been learning MQL4 recently so that i can convert some of my profitable systems to EAs. I have been hitting so many road blocks but i know nothing good comes easy.

So what do i suggest?

Try monitoring your EAs (if you have) and make sure they are not opening bad trades based on your own analysis of the market.

Quoting Expert44
No Expert Advisor can replace human mind.
This is my belief after many years of dealing with market.......
Any thoughts on this? Please share.
Good EAs are not designed to replace human minds, I repeat "Good" EAs are never designed to replace human mind. They are just a tool to speed up the process and perform mechanized tasks like any machine or computing euipment does.

We all can manually add 1345+1278+56789 but have invented calculator to do this task for us. Calculator does not know which figures to add, we provide these figures it just adds whatever we want to add. With the invention of computers we have assigned some more tasks to these machines, including arriving at the figures that we want to add.

Development of Artificial Intelligence is not at a stage where one could design an EA that replaces human mind.

Wish I was able to develop an EA for my method. My life would have been much easier.
The Thief of Wall Street
I like the term Artificial Intelligence. From here we can start. Lets say that this term can replace 70% of the human mind. On the other side human mind cannot match to its speed of calculations and other actions. Therefore using Artificial Intelligence + human mind is very needed in trading or maybe necessary. Again the conclusion is that one cannot exist without the other at least on the long range of time.
Quoting Chicky
Good EAs are not designed to replace human minds, I repeat "Good" EAs are never designed to replace human mind. They are just a tool to speed up the process and perform mechanized tasks like any machine or computing euipment does.

We all can manually add 1345+1278+56789 but have invented calculator to do this task for us. Calculator does not know which figures to add, we provide these figures it just adds whatever we want to add. With the invention of computers we have assigned some more tasks to these machines, including arriving at...
Hello Chicky,

I agree with your thoughts as well, but...most novice traders do not share our thoughts, they are more interested in an auto trade robot that they can unleash onto their favorite instrument and lay back while the money roles in. This is the mentality we are trying to educate to abandon that trail of thought and start looking at EA's as a tool to enhance their trade making decision and trade management instead of a stand a long tool that you set it and forget it.

Ultimately we are all agreeing in this thread.

Cheers,
I feel the main failing of EAs is their inability to deal with fuzzy zones of support and resistance which can often occur in multiple and complex configurations, and which are a major factor determining what the price will do next. Has anyone seen an EA that can do this?

For any newbies reading this I'd advise you to forget about EAs and focus on training yourself to be the ultimate trading robot .
Quoting sambo83
For any newbies reading this I'd advise you to forget about EAs and focus on training yourself to be the ultimate trading robot .
Semiautomatic EA is better than fully automatic and better than manual trading.
Subsignals are a great helper.
i really like your post about ea cant beat human trader and i agree, there is 100% some private ea that is good but no one will sell to you but still i think its better to trade by yourself because you can see things that most ea cant...
Quoting Expert44
No Expert Advisor can replace human mind.
I believe there is no better customer for a market maker than one who is willing to blindingly use EAs as "fire-and-forget" money making machines. They really are money making machines - for their market makers.

As I already wrote in another thread - take a guy that believes he can fire up an EA and simply make money long term, without any involvement and challenge him to find a forex strategy, which can be traded in a purely 100% mechanical manner - because a 100% automatic EA can only be based on such a strategy - what, you'd think should be quite obvious.

Those people still seek a holy grail of an EA exactly for the wrong reason. They are unable to operate a manual trading system due to discretion (i.e. knowledge and experience) it requires, yet they a willing to blindingly use an EA, which is based on that system.
Quoting Giansar
I believe there is no better customer for a market maker than one who is willing to blindingly use EAs as "fire-and-forget" money making machines. They really are money making machines - for their market makers.

As I already wrote in another thread - take a guy that believes he can fire up an EA and simply make money long term, without any involvement and challenge him to find a forex strategy, which can be traded in a purely 100% mechanical manner - because a 100% automatic EA can only be based on such a strategy - what, you'd think should be...
there are strategies that can be programmed into ea's and are working over a very long time. Having said that, I never have seen a consistently profitable one online.

People should get used to the thought that if they want an ea, they need to do a tremendous amount of market research and program it by oneself.
Quoting Expert44
Semiautomatic EA is better than fully automatic and better than manual trading.
Subsignals are a great helper.
I would have to agree with you. If your trading style would benefit from subsignals then EAs are extremely useful. I guess this is more likely to be the case if your trading opportunities are infrequent and you don't want to be at the computer all the time or if you're looking to trade lots of instruments and need an EA to filter them for you.
So, we cannot live without at least semiautomatic EAs.
Quoting Expert44
No Expert Advisor can replace human mind.
This is my belief after many years of dealing with market.
You are right. EAs are made by humans can't be better than their creators.

But... they do their job at night, when the humans need to sleep.

Josef
  • Post #17
  • Quote
  • Edited 5:57pm Jul 18, 2011 8:04am | Edited 5:57pm
I use indicators that alert me to potential setups. But the decision on whether - and how - to enter the trade is manual. Given the number of variables that I look at (currency strength, S/R quality, time of day, imminent news events, etc), it is too complex to be left to an EA. And I feel more comfortable trusting my own judgment.

I use a couple of EAs (that I wrote myself) to manage my open trades. Saves me many hours at the computer and also ensures that exits are made objectively and emotionlessly. Unless a trader wants to make adjustments to his trading plan as each trade unfolds, there is no reason to exit manually.
I wish EAs be on iPhone MT platforms. Looks like they move slowly in this direction.
Quoting Chicky
Good EAs are not designed to replace human minds, [b]
We all can manually add 1345+1278+56789 but have invented calculator to do this task for us.
No seriously...

I think the issue is defining the steps we take in arriving at trading decisions in a different way so that they could be replicated using an EA.

It's not that there is a lot of complexity required but I think we process a lot of very simple information 'unaware' at the time we're actually doing it.

In my case, an EA would just need to do the same as I do on demo and the opposite of what I do on live.

The EA would be to trading what the calculator is to adding your numbers.

Regards, Paul.
  • Post #20
  • Quote
  • Edited 6:29pm Jul 18, 2011 6:18pm | Edited 6:29pm
Quoting pah
No seriously...

I think the issue is defining the steps we take in arriving at trading decisions in a different way so that they could be replicated using an EA.

It's not that there is a lot of complexity required but I think we process a lot of very simple information 'unaware' at the time we're actually doing it.

In my case, an EA would just need to do the same as I do on demo and the opposite of what I do on live.

The EA would be to trading what the calculator is to adding your numbers.

Regards, Paul.
Paul,

I agree that even the most discretionary technical strategy could potentially be coded. It would be a case of sitting down with the trader, putting a basic design together, and then incrementally adding all of his nuances, over a long period of time. There would likely be basic rules, then exceptions to these rules, then exceptions to the exceptions, and so on.

There are other reasons why the process would not necessarily be trivial. Some things that appear to be visually simple (support and resistance is a good example) are not necessarily that easy to reduce to a mathematical formula, especially if the trader is analyzing subjective abstracts like the 'quality' of the S/R. The same could be said for trendlines, chart patterns, multiple time frame analysis, and perhaps different aspects of trade management. And then there are other more abstract nuances, for example the trader might not trade 'today', because today is an options expiry day. Or correlations, perhaps, e.g. he might make a decision on whether to trade a certain pair based on what the stock indices did, or the price of gold.

From a design/programming standpoint, then, it would be a very interesting and challenging exercise! I imagine that the 'finished' product would result in many hours' work, and several hundred lines of code. I say 'finished' in inverted commas, because some traders might make changes to their strategy as market conditions continue to evolve. For example, the current recession might necessitate adjustments to an approach that worked successfully 5 years ago, as Europe and the US continue to battle their financial problems.

If the trader uses fundamentals in his decision making, then there would be added complexity, ranging from difficult to impossible. An example of 'difficult' would be to get the EA to read the results published on a news calendar, and then interpret some results (actual vs forecast; actual vs previous) in context with others. What is more significant, a 0.2% rise in CPI, or a 0.4% rise in GDP? As an example of 'impossible', the EA obviously couldn't be made assess the content of a Bernanke speech.

For the above reasons, I'm happy to trade in the semi-automated manner that I described in my earlier post.

David
0 traders viewing now