Phantom 6 System | Page 58 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Jag's Phantom 6 System 295 replies

Phantom 6 - Manual System 8 replies

Phantom Entry on ODL Japan. 2 replies

Tradeview CAD/JPY "Phantom" spike 0 replies

Phantom high, phantom low 0 replies

Attachments: Phantom 6 System
Exit Attachments

Phantom 6 System

  • #1,141
  • Apr 1, 2011 6:23am Apr 1, 2011 6:23am
Hi again friends.
While some of you are testing the new interface, I am thinking about how to make this EA better, because I really like it, and I believe it has potential.
In the question of the UI, I think about to add a line showing the Max/Min PL for the current basket, tracking if the current basket was close to reach the profit target prior to go to the negative zone, by instance.
Maybe useful too, to add some buttons (like a menu) that allow the user close all the basket, or reset phantom values right now, or, including, trade the highest pair right now. But this last question can wait, I believe.

In the trading logic, I should say that I'm worry about trading without net. I say, this EA don't use StopLoss, because we know that it is hedging continuously pair after pair, reducing the possible loss to a minimum. But, what about a disaster like a shutdown of the VPS? Then, the EA can't hedge the first trade, and if it goes against us the loss can grow greatly (Murphy's law, you know). I think that we should add a StopLoss to each trade, so far away as you want, we can discuss the quantity, but the possible loss should be stopped.

Other major questions on the trading logic maybe the use of Break Even and Trailing Profit points. I see that in some cases, a unique trading order reach our target profit in few minutes. This means that the momentum is big, and ... what should we stop the profits when we can double it, by instance? Then my proposal is that in a level set by the user, the EA don't take profit but trailing a stop loss a quantity of $ down the current profit. This way we leave to run profits until the market would want.

Another minor logic change maybe that the user can set up the trading interval, freeing it to not be just every hour, in the first minute (adding a parameter IntervalMinutes, by instance).

I don't know what these changes can impact the EA performance.

What do you think about this?
  • #1,142
  • Edited 8:42am Apr 1, 2011 7:36am | Edited 8:42am
Quoting xere
But, what about a disaster like a shutdown of the VPS? Then, the EA can't hedge the first trade, and if it goes against us the loss can grow greatly (Murphy's law, you know). I think that we should add a StopLoss to each trade, so far away as you want, we can discuss the quantity, but the possible loss should be stopped.

Other major questions on the trading logic maybe the use of Break Even and Trailing Profit points. I see that in some cases, a unique trading order reach our target profit in few minutes. This means that the momentum is big, and ... what should we stop the profits when we can double it, by instance? Then my proposal is that in a level set by the user, the EA don't take profit but trailing a stop loss a quantity of $ down the current profit. This way we leave to run profits until the market would want.
I have not been using VPS and I shut down my pc at night (from 16:00GMT to 2:00GMT) and I have not faced any such problem with 2a. It has continued to perform well.
About profit trailing, I feel it is mot essential. I have seen 2a sometimes even reach a profit 7% or 8%. When the momentum is good, we should encash it.
"Put all your efforts in becoming effortless"
  • #1,143
  • Apr 1, 2011 10:35am Apr 1, 2011 10:35am
Quoting IamTHAT
I have not been using VPS and I shut down my pc at night (from 16:00GMT to 2:00GMT) and I have not faced any such problem with 2a. It has continued to perform well.
About profit trailing, I feel it is mot essential. I have seen 2a sometimes even reach a profit 7% or 8%. When the momentum is good, we should encash it.
Then, could you tell me, please, what set up you have in the bot? What ProfitPic and MaxMarginToUse? Because I think I have a tight ProfitPict as 0.25 and MMTU 0.25. My idea is not to risk too much, being safe with an account with 5000 $ or so and 1:500 as leverage (until now just demo account, of course).
  • #1,144
  • Apr 1, 2011 10:43am Apr 1, 2011 10:43am
Quoting IamTHAT
I have not been using VPS and I shut down my pc at night (from 16:00GMT to 2:00GMT) and I have not faced any such problem with 2a. It has continued to perform well.
About profit trailing, I feel it is mot essential. I have seen 2a sometimes even reach a profit 7% or 8%. When the momentum is good, we should encash it.
Interesting - you're shutting down during the US session. I would have thought that the Asian session would be the session to skip. What do you do with open baskets when you shut down?
  • #1,145
  • Apr 1, 2011 10:55am Apr 1, 2011 10:55am
Quoting xere
Then, could you tell me, please, what set up you have in the bot? What ProfitPic and MaxMarginToUse? Because I think I have a tight ProfitPict as 0.25 and MMTU 0.25. My idea is not to risk too much, being safe with an account with 5000 $ or so and 1:500 as leverage (until now just demo account, of course).
I am running it on FXPro with 1:500 leverage with MMU of 0.3 and profit Pc of 10. I close manually whenever profit stalls for few min. ( This is where we need some sort of indicator) Couple of days back I have closed a basket with 7% profit. Even today One basket closed with nearly 3%.
"Put all your efforts in becoming effortless"
  • #1,146
  • Apr 1, 2011 11:02am Apr 1, 2011 11:02am
Quoting TheMaxx
Interesting - you're shutting down during the US session. I would have thought that the Asian session would be the session to skip. What do you do with open baskets when you shut down?
Yes, midway through the US session, can't help it I am on GMT +5.30. asian session is practically of no use, but now with jpy flying around, it can't be ignored. I keep the basket open as it is, even on week ends, that has not bothered me much. May be when we go live, we need to close the basket on week ends atleast..Today my DEMO has made a profit 4% today, and the floating is around 1.5%.

2a has been stable somewhat, whereas I was running AddSix1a on another demo which blew up, after a fantastic run. That is what scares me.
"Put all your efforts in becoming effortless"
  • #1,147
  • Edited 11:46am Apr 1, 2011 11:29am | Edited 11:46am
Hello,

after new tests (see below for the URLs) I decided to change Phantom_AddSix_02a with Phantom_Add_Six_03b.

But I want to merge some other "old" features and something new and easy, for more testing, that I will explain later. I already added the "PhantomCheckMinute" feature, coded for Phantom6_01xR, as well as I'll add xere's new graphic interface (only if he will not do it by himself).

xere, with this post I ask you the permission to merge your code into this attached EA. Or do you have time to do it (I can wait, of course)?

What I want to add ASAP is a single, simple coded line that gives new value to the "PhantomCheckMinute" variable, after every closed basket; it should check what are the last 2 digits of the time string, sum 5 and round up to next 0; if the total >= 60, subtract 60.

I think that the right place where to insert this new line is around line 916 (at line 917 there is my comment about what to add too).

Of course it is also possible to set this feature on and of, but I'd just like to test if this "accelerated EA" performs better or not.

TheMaxx (or anyone wants to help me on this), can you do it, please?

I thank you in advance.

Good weekend.
/achab

Test 31:

Phantom_AddSix_02a, 31a, 31b and c;
Phantom_AddSix_03b (modified), 31d, 31 e and f.

http://www.mt4i.com/users/achab-test-31a/stats
http://www.mt4i.com/users/achab-test-31d/stats
http://www.mt4i.com/users/achab-test-31b/stats
http://www.mt4i.com/users/achab-test-31e/stats
http://www.mt4i.com/users/achab-test-31c/stats
http://www.mt4i.com/users/achab-test-31f/stats
Attached File(s)
File Type: mq4 Phantom_AddSix_03b_0.4.1.mq4   51 KB | 240 downloads
  • #1,148
  • Edited 3:41pm Apr 1, 2011 3:09pm | Edited 3:41pm
Dear achab: Of course you (really everybody), can edit the 03 release because it's not mine, it's OURS (in fact, the first idea was from MetaCoder, if I remember well. From here, I thank him). I just give you the interface. Feel free modifying the code, just be cautious not to make poor performance, or, maybe better, allow that the user can set up the new capabilities on/off.
All right, I have seen what the parameter CheckMinute does... Simply changes the minute the EA works sending trades. Not the first minute, but another. I have seen it's easy to code. If you want, you can edit the 03 easy. Would you? Or do you want that I do it? Right now, I prefer to rest for a while, taking a vacation for some days (last days were full of activity to me). Feel free doing you the job. I explain: I just add some global vars in order to track the new data (I did comment the lines as "Added by xere") and some code to track them, and add a couple of functions called DrawChartComments and DrawText at the bottom, and an "if" statement in the end of the start function that switches between the two interfaces. That's all.
Thanks for all (I see my job is in value...)
  • #1,149
  • Apr 1, 2011 3:15pm Apr 1, 2011 3:15pm
Quoting IamTHAT
Yes, midway through the US session, can't help it I am on GMT +5.30. asian session is practically of no use, but now with jpy flying around, it can't be ignored. I keep the basket open as it is, even on week ends, that has not bothered me much. May be when we go live, we need to close the basket on week ends atleast..Today my DEMO has made a profit 4% today, and the floating is around 1.5%.

2a has been stable somewhat, whereas I was running AddSix1a on another demo which blew up, after a fantastic run. That is what scares me.
Thanks for your help, IamTHAT. Your approach to the EA is interesting. In fact you let run the basket until it reaches a great profit or you close it by hand when you think it's enough. But this reinforces my thought of develop a kind of Trailing in order to not become our profits into losses, without human monitoring (as your previous experience with 1a...).
  • #1,150
  • Apr 1, 2011 4:05pm Apr 1, 2011 4:05pm
Quoting achab
Hello,

after new tests (see below for the URLs) I decided to change Phantom_AddSix_02a with Phantom_Add_Six_03b.

But I want to merge some other "old" features and something new and easy, for more testing, that I will explain later. I already added the "PhantomCheckMinute" feature, coded for Phantom6_01xR, as well as I'll add xere's new graphic interface (only if he will not do it by himself).

xere, with this post I ask you the permission to merge your code into this attached EA. Or do you have time to do it (I can wait, of course)?

What...
Few months before I replace Phantom_Add_Six_03b with Phantom_AddSix_02a for testing....

My observation about Phantom_Add_Six_03b is if it hits 3-4 cut lossers or do not close basket in 2 days then it goes out of control.. and you have to wait for very long time to recover or it may not recover and you will get margin call.

Right now I have stopped experimenting with phantom EAs. Only one ea I am testing is Phantom_Addthree_02a. Its same as Phantom_AddSix_02a but only for 3 pairs and its not like art phantom. Settings are pp=1 mmtu=0.2, Reset Hours=150 (no reseting). Clossed 3 baskets in last week.

Its my final testing with phantom ea
  • #1,151
  • Apr 1, 2011 6:44pm Apr 1, 2011 6:44pm
Quoting xere
Thanks for your help, IamTHAT. Your approach to the EA is interesting. In fact you let run the basket until it reaches a great profit or you close it by hand when you think it's enough. But this reinforces my thought of develop a kind of Trailing in order to not become our profits into losses, without human monitoring (as your previous experience with 1a...).
Xere,

Few pages back we had discussed a bit about a trailing equity stop for the phantoms. Not for individual trades but for the account equity. I use a conservative profit pct of .25. Mostly because I like quicker closures and I am unable to monitor my EA's during the london/us session because of my "real" job.

I like the idea of being able to set a higher profit percent for the times the basket reallly takes off, but yet still lock in a profit if it should reverse before PP is hit. I only know a little about programming so I dont have the expertise to code up something like this.

My idea is to trail the equity by a certain percentage as the profit goes up. For example set original profit percent to 3%. At 1% profit EA sets equity stop to .5%, at 2% profit sets it to 1% etc. So that would be a trailing equity stop of 50%. Is it possible to code something like this?

Of course, only if you are interested in such an idea.
  • #1,152
  • Apr 1, 2011 11:20pm Apr 1, 2011 11:20pm
I have run "Add Six 03b" since 28 Dec 2010 till now .. never got any drama, and it runs auto pilot ( no intervention ) .. balance start 5000 and current balance 5,944 and floating P/L $-6.19 .. and i use 10 cents per-trade

Quoting Password
Few months before I replace Phantom_Add_Six_03b with Phantom_AddSix_02a for testing....

My observation about Phantom_Add_Six_03b is if it hits 3-4 cut lossers or do not close basket in 2 days then it goes out of control.. and you have to wait for very long time to recover or it may not recover and you will get margin call.

Right now I have stopped experimenting with phantom EAs. Only one ea I am testing is Phantom_Addthree_02a. Its same as Phantom_AddSix_02a but only for 3 pairs and its not like art phantom. Settings are pp=1 mmtu=0.2,...
Attached Image (click to enlarge)
Click to Enlarge Name: statement.gif Size: 5 KB
  • #1,153
  • Apr 2, 2011 2:05am Apr 2, 2011 2:05am
Quoting thorn
My idea is to trail the equity by a certain percentage as the profit goes up. For example set original profit percent to 3%. At 1% profit EA sets equity stop to .5%, at 2% profit sets it to 1% etc. So that would be a trailing equity stop of 50%. Is it possible to code something like this?
Of course, only if you are interested in such an idea.
I tried one EA from T101 thread (Account profit protect EA) which did not work properly. May be Steve's mptm should be tried.
If doing it manually,monitor the chart of the pair(s) which is contributing to the profit, and if there is any steam left. It was easy last week with jpy pairs.
"Put all your efforts in becoming effortless"
  • #1,154
  • Apr 2, 2011 3:26am Apr 2, 2011 3:26am
Quoting Password
Few months before I replace Phantom_Add_Six_03b with Phantom_AddSix_02a for testing....

My observation about Phantom_Add_Six_03b is if it hits 3-4 cut lossers or do not close basket in 2 days then it goes out of control.. and you have to wait for very long time to recover or it may not recover and you will get margin call.

Right now I have stopped experimenting with phantom EAs. Only one ea I am testing is Phantom_Addthree_02a. Its same as Phantom_AddSix_02a but only for 3 pairs and its not like art phantom. Settings are pp=1 mmtu=0.2,...
Hi Password,

after my several tests I have demonstrated to myself that often the biggest difference in Phantom's performances depends from the broker: which one(s) did you use for your tests?

The ratio between MMTU and PP you use is the opposite I use: yours is "0.2", mine is "4.0", actually on my Alpari real account and "5,86" on my AAAFx real account (I drop the infos about demos).

I prefer to have high average of closing baskets (actually 1,87/day on AAAFx since February 2011 and 5,49/day on Alpari since 2 days) instead to wait, watching the EA while running.
But, your 3 baskets per week mean 3% of gain (weekly), 12% of gain per month as average: great result!

I will be happy and proud if I'll have constant gains about 3-6% monthly, balancing the gains with the risk to blow up.

Finally, I have no idea about what happened to your Phantom_Add_Six_03b tests: could you share them, please?

Have a nice weekend.
/achab
  • #1,155
  • Apr 2, 2011 3:38am Apr 2, 2011 3:38am
Quoting sapiemas
I have run "Add Six 03b" since 28 Dec 2010 till now .. never got any drama, and it runs auto pilot ( no intervention ) .. balance start 5000 and current balance 5,944 and floating P/L $-6.19 .. and i use 10 cents per-trade
Hi sapiemas,

and welcome here. I have read your post and watched your graph balance with interest; could you say me/us:

 

  1. Is this your real account?
  2. Which broker do you use?
  3. Which are the settings you use (MMTU, PP, Leverage and their changes, just in case)?
  4. Could you, please, share your statement, too?


Your monthly average of gains is interesting (around 6.25%), very near what I am looking for, we can say "my final target".

By the way, sorry for my many questions (maybe too much), but feel free to answer only if it is comfortable to you.

Good weekend.
/achab

  • #1,156
  • Apr 2, 2011 4:01am Apr 2, 2011 4:01am
Quoting xere
Dear achab: Of course you (really everybody), can edit the 03 release because it's not mine, it's OURS (in fact, the first idea was from MetaCoder, if I remember well. From here, I thank him). I just give you the interface. Feel free modifying the code, just be cautious not to make poor performance, or, maybe better, allow that the user can set up the new capabilities on/off.
All right, I have seen what the parameter CheckMinute does... Simply changes the minute the EA works sending trades. Not the first minute, but another. I have seen it's easy...
He again, xere.

Thank you for your answer: it was to be expected. Maybe my question was useless, but a question more is always better than one less, IMHO.

Ad abbundanziam, if you read the first lines containing the comments, you'll see mine about the propriety of these EAs.
I agree with you also when you write about "Added by xere": it's what I already did in there.

The only "problem" I can see now, is the future confusion about the releases (when they will be many), that's why I update the version number every time I modify it and make it print on the graph the version number and date.

Also I am a bit tired and busy in these days, but I will make these updates in next days, if you will not do it before me (it's better to communicate each other when we will start to do it). I prefer e good job that needs time than to make things hastily.

At last, but not the least, yes: your job is in value for sure!

Good weekend and vacation!
/achab
  • #1,157
  • Edited 5:46am Apr 2, 2011 5:10am | Edited 5:46am
Quoting xere
Hi again friends.
While some of you are testing the new interface, I am thinking about how to make this EA better, because I really like it, and I believe it has potential.
In the question of the UI, I think about to add a line showing the Max/Min PL for the current basket, tracking if the current basket was close to reach the profit target prior to go to the negative zone, by instance.
Maybe useful too, to add some buttons (like a menu) that allow the user close all the basket, or reset phantom values right now, or, including, trade the highest...
xere,

we must have a closer relationship because you think and write too similarly to what I do.

Well, this will be a long post, so sit back.

Your idea about "adding a parameter IntervalMinutes" is quite the same I have in mind, if I have understood it properly.
It would be perfect if we could merge this last new option with the "Break Even and Trailing Profit points" ones.

Could you imagine the Phantom EA that starts trading at "00" time, goes rapidly to the first main target (PP), override it because the strong momentum, continuing to performing up to the second target and, if the momentum is still strong, put there a trailing stop loss for continuing the gain. When it will be stopped, the EA has gained much more than now. If this basket is closed after, for example, few minutes (and it happened to me many times, depending on the high ratio between MMTU and PP I use) I imagine its ability to watch the time and to set itself to change the "PhantomCheckMinute" exactly how I have explained in my previous post (sum 5 and round up to next 0; if the total >= 60, subtract 60) to open a new trade as it does normally now. More trades, and more safe, equal to more positive performance, IMHO.

Of course I candidate myself for many, different stressing tests.

This is my priority list:

  1. Add a grafical indicator (very similiar to "Equity_bal_traker" but without the bug I have experienced) because I like to have an immediate idea of what's going on just watching the monitor (at a glance);
  2. Add to "PhantomCheckMinute", already present in this Phantom downloadable here, the feature to change its starting time by self and to be set on or off (just because it should be easier than other mods);
  3. Add your Trailing Profit feature;
  4. Add your Break Even feature that overrides the PP (PP will be only the first target);
  5. Add your line showing the Max/Min PL for the current basket, tracking if the current basket was close to reach the profit target prior to go to the negative zone;
  6. Add a day and time of starting trading and a day and time stopping trading (I like the idea to stop its work after it closes the basket on mid Friday, because I want to think to other when resting or during the weekend and to start on Monday while I am still sleeping to watch what it did in first 2-4 hours of the morning, drinking my cappuccino);
  7. Add the stop loss (at the bottom because tests demonstrated how it works fine and because just in case of a "shutdown of the VPS", it's possible to run MetaTrader also from another PC, office or home).


Now I have a question: how to merge the new features to each Phantom? They are enough different...

Again I wish you a best rest.
/achab

  • #1,158
  • Apr 2, 2011 5:49am Apr 2, 2011 5:49am
Quoting IamTHAT
Yes, midway through the US session, can't help it I am on GMT +5.30. asian session is practically of no use, but now with jpy flying around, it can't be ignored. I keep the basket open as it is, even on week ends, that has not bothered me much. May be when we go live, we need to close the basket on week ends atleast..Today my DEMO has made a profit 4% today, and the floating is around 1.5%.

2a has been stable somewhat, whereas I was running AddSix1a on another demo which blew up, after a fantastic run. That is what scares me.
That's what scares me too, and any fan of Phantoms, I guess.

If you take a look to my tests (the number 3, doubled and blown up, 27a, doubled and still running or 23, tripled in less than one month and still alive, for example) you can notice that Phantoms can have amazing performances. But, how long?

More you ask from it, before it will blow up (statistically speaking, let's keep the finger crossed).
That's why I am using low PP.

Which is your month target, averagely speaking?
  • #1,159
  • Apr 2, 2011 7:20am Apr 2, 2011 7:20am
Quoting thorn
Xere,

I like the idea of being able to set a higher profit percent for the times the basket reallly takes off, but yet still lock in a profit if it should reverse before PP is hit. I only know a little about programming so I dont have the expertise to code up something like this.

My idea is to trail the equity by a certain percentage as the profit goes up. For example set original profit percent to 3%. At 1% profit EA sets equity stop to .5%, at 2% profit sets it to 1% etc. So that would be a trailing equity stop of 50%. Is it possible to code...
Yes to your two questions: yes, it is possible and easy to code a trailing like you say, and yes, I am interested in the idea. I ask to you read the last comments posted by achab (because they can be the start of a collaboration between all of us), wait for a few days and wait for our new comments. My will is to make an almost "serious planning" of including all modifications we are thinking about, not too much rigid in time terms, but slow and safe.
Good weekend!
  • #1,160
  • Apr 2, 2011 7:23am Apr 2, 2011 7:23am
Quoting achab
xere,

we must have a closer relationship because you think and write too similarly to what I do.

Well, this will be a long post, so sit back.

Your idea about "adding a parameter IntervalMinutes" is quite the same I have in mind, if I have understood it properly.
It would be perfect if we could merge this last new option with the "Break Even and Trailing Profit points" ones.

Could you imagine the Phantom EA that starts trading at "00" time, goes rapidly to the first main target (PP), override it because the strong momentum, continuing to...
achab,

Fantastic! I like your capability of sorting the ideas. Give me a couple of days and wait for my proposal about making a planning with every questions listed.

Good weekend!
1 trader viewing now