Cable Update (GBPUSD)

One can use EA :

if controlling emotion is not possible.

If a person has some work outside and not able to do manual trading.

If a person wants to earn some money thro properly developed EA in a perticular session, for example London open to london close.

One must get after sales support service if purchased in the market.

This is highly doubtful.

And i just saw the calender this week.

Account blowing week.

Lots of red news events.

Experienced and newbies may avoid trading before news event in order not to blow the account.

I am contiuning this discussion to point out the mistakes in logic being presented, not for trolling.

You are comparing the actions taken by an EA versus a human but are including conditions that the human "sees" that are not programmed into the EA. So of course if the human is allowed to use a much larger rule set than the EA, there is no other outcome than for the EA to perform worse than the human. If the human and EA are on equal footing as to the rule set being used, then the human will be the worse performer due to human error and speed differences in execution of the rules. (How many times have you clicked "buy" when you meant to click "sell"? How many times did you set the SL or TP in the wrong place or enter the wrong number of lots because you entered the wrong value? EA does not make those mistakes.)

A premise was given that for an existing positive edge, rules are invoked by the trader to decide when/where to enter a trade. So, whether those rules are simple or complex, and whether or not the edge continues to be valid and yield positive results under the latest trading environment were not being addressed. And so, every rule that the human trader is using in his decision making can indeed be programmed into the EA. The problem lies in the human's ability to translate his rules into code. In addition, mention was made of the edge evolving over time. And even this can be through a change/improvement of the rules within a static trading environment, thereby improving the net results, or through a change in the rules brought on by observation that the net results are decreasing, suggesting that the trading environment has changed.

The fact that an EA "worked" for some period of time but then failed for another period of time does not change the fact that those rules were developed by the human trader who determined that they were "good" rules. And when it is stated that "the market has changed", I posit that the market has not changed at all but rather, the market has simply moved into a different phase than it was in earlier. (In North America during the middle of Winter with the temperature below zero for many days, then several days of several degrees above freezing occur, no one says the Winter season has changed to Spring but rather that a temperature change has occured, albeit not a FREQUENT occurrence.) As a simple example, when viewing a single TF and it is obvious there is an up trend, then the trader's rule set dictates only buying, and the buy entries are determined by the rules. So, when a large retracement occurs that is outside of the bounds of the rules set, it is decided that OOPS! THE MARKET CHANGED because the buying rules for entries are never met during this retracment phase. Such a terrible EA this is! Time to throw out this EA with those buying rules and try something else. An alternative to throwing out the EA, considering that it was "working well" earlier, would be to examine this new phase and develop some rules to deal with it and program them into the EA.

No trade is supposed to be taken between 3 pm est and 8 pm est ( until Tokyo opens )
On some and one day, for some unknown reasons, some pair moves 50/60 pips in that time duration, then when EA is put on when tokyo opens, it will malfunction.
That is simply wrong. The EA will continue to function exactly how it is programmed and not make any trades during that time period no matter how much price moves in the non-trading period. There will be no malfunction of the EA when the Tokyo session opens. The fact that the trader sees a missed opportunity is not relevant to the functioning of the EA. However, if the trader sees this unexpected price movement and determines that he would like to try to capture such missed opportunities, then he can make changes to the rule set and adjust the EA coding.

i highly doubt if the person who is inputting to EA is fully knowledgeble. All are almost half knowlegeble in this field.
This is also irrelevant to the working of the EA as programmed. That speaks to the fact of the deficiencies surrounding moving the trader's rule set into an EA. It is difficult to impossible for most traders to turn trading rules into an EA because they lack programming skills. And similarly, they lack the ability to describe their rules in such a way that a skilled coder can build an EA that matches the trader's rule set.

Human intervetion is a must even if EA is developed.
That is ONLY if you intend to do something that is not programmed in the EA! So I will assume that you expect every EA is so defficient that human intervention is always required. And if so, then as I said at the beginning, you are trying to present the EA as defective when the actual case is that you give the EA and human two different rule sets to use.

Experienced and newbies may avoid trading before news event in order not to blow the account
If an EA is used to insure being flat and to block entering trades over a selected window around the news event, then no way to blow the account.

I think all you write is correct, and yet the fact remains that part of your perception is unconscious, i. e. you are confident that you know all the rules to apply but actually you are not aware of everything you perceive.
After years of watching price you develop a sort of intuition what will happen next. (Think: psychology/theory of Gestalt)
Pressed, you will give five criteria why, what happens next, does happen - then you programme an EA and discover (in back- and forward-testing) it doesn't work the way it should.
The reason for this may well be: you don't consciously know, what you know intuitively (think: riding a bike...)
So: while you are 100% right theoretically; practically things may work out differently.
Just saying...

Fomc this week. Dollar started it's move the Thursday before as usual. Odd that it appears down this time after about a year of up. This is scary for the markets IMO as a weak dollar may spur the markets up for Xmas. While great for Xmas this would technically signal a sell off next year. IMO a prosperous year will require Dollar strength in the coming month.
Either way. Swingers get ready.

true

Hi Guys,
For those of you that know me from Steve Hopwoods trading forum. I have been dealing with Ea's for many years now and believe me, Steve Hopwood has the cream of the crop when it comes to coding Ea's and many talented coders there. You name it and it will be coded. An Ea is only possible if it can replicate an indicator which has code. I believe that Navk has a point in saying that an Ea cannot do what the eye can see when it comes to levels because it cannot be coded. Everything else can be coded as Dave eluded to. Time of trade, lot size stops/profits and time of entry are all possible to be coded. By the way, Ea's work best on longer time frames. I am yet to see a good scalping Ea. Why do I have indicators on my chart? One very simple reason is because I have persued this for more than 10 years now and have tried and gaining progress to get an Ea to replicate my trading system. If you can get the holy grail indicator you are on the the right path. There is no easy route.
Cheers.

ah interesting, you even ignore the news calendar for your decisions and act purely technically. thanks for your answer and pics.

Who is going to give inputs to EA developer ?

You or me or he or she ?

This is where the problem lies.

All of us are half baked.

So the EA developed by EA developer will also be half baked.

And I or anyone will agree with you when some one show proof some EA constantly works and makes profit.

If at all one is available, I am sure it will be worth a million.

Having said , the half baked EA will be useful for :

1. Those who can not control emotion

2. Those who does 9 to 5 jobs and uses EA for additional income or to loose what he earned in 9 to 5 job.

3 . Those who know nothing about trading but has desires to make million in 1 month.

I am planning to trade CPI after seeing on market reaction.

EA can not

If one such EA is developed, please let me know.

I will borrow money to purchase the EA.

Please.

I agree with you to some extend. I suppose the “levels” you mentioned are highs/lows, price structures, etc.
ADR, pivot, etc are easy to code.

I have experience in coding Price Levels in indicator/EA, price levels associated rules are so difficult to be coded since the indicator/EA has “time” element.

Levels are easy to be seen with eyes such as those be tested as support/resistant multiple times. But we have so much flexibility in determining which ones are the “levels” (intuition).
In coding we need rigid rules, then it becomes very complicated. How many times as support and resistant to justify? For how long in history? How frequent? How to define a retest (eg volatility/ reject for how many pips)?

There are a few approaches in coming up code-able rules to determine price levels. Each has its strength and shortcomings. There are price level indicators, divergence, break-of-structure alerts available, some are pretty good at identifying levels.

The way to go is to find a rule set that works for your strategy and trading method.

trade well.

Thats a bit selfish don't you think Navk. There are traders here that cannot sit all day looking at trades because of their personal circumstances, commitments etc. but there are routes for traders. All cannot be here all day,
Cheers.

yeah, I agree all can not sit on trading for various reasons. This includes me too.

I mentioned this in my previous post, EA can be good for part timers and who can not control trading emotions.

Having said, all are points and arguments only.

Until now, to my knowledge, no good EA available.

If such EA is available, i am sure.... people will go ahead purchase that EA at any cost.

of course, not everyone can sit in front of the chart all day.
but unfortunately I don't know a good EA that could be left unattended.
maybe it exists.

This is my experience in using EA s.

In the year 2009, that is the time, first time, i got interest in trading.

I bought EA ofr 100 CAD. Cheeeep !!

On a 10000 account, it was making 1 to 1.5 k regularly. every day.

Then, the account was about to be vanished, it became arounf 3.5 k account, i threw that EA and started leaning manual Trading.

And until now....the learning continues.

Any EA that is effective and sold so that many people are using it will be rendered useless by the trading AI's that will simply start to move the market price against it.

Markets always be a changing, unless the EA can change with it then no use, well that's my belief could be wrong 50/50...