Phoenix 2007 (new thread) | Page 39 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Phoenix Forex Manager trading journal 25 replies

Phoenix Foundation 1 reply

Can someone fix the Phoenix EA? 0 replies

Automated Trading Championship 2007: New Challenge! 33 replies

Phoenix is here! 848 replies

Attachments: Phoenix 2007 (new thread)
Exit Attachments

Phoenix 2007 (new thread)

  • #761
  • Nov 12, 2006 10:19am Nov 12, 2006 10:19am
Quoting BenQ
Wackena,

In your attachement to this post, how did you get the results to partition out by pairs?
The attached excel file is created by:
1. Opening the DetailedStatement.htm with Excel.
2. Select all data, starting just below the Deposit line and stopping just before the p/l line.
3. Click sort in the top menu Data drop down.
4. Check "no header row".
5. Sort by column E - ascending.
6. Click OK.
7. Select first sorted cell in column E.
8. Click Subtotal in the top menu Data drop down.
9. Click OK at next prompt window "MicroSoft Excel cannot determine..."
10. Select column E in "At each change in".
11. Select only column N in "Add subtotal to".
12. Click OK.

The rest is manual, such as pair win/loss/profit spreadsheet. I delete name and account number and add blank rows before and after each pair total line.

Wackena
  • #762
  • Nov 12, 2006 11:06am Nov 12, 2006 11:06am
Howdy,<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>

Just got two ideas (was called out of my standby this morning @ 3:15 –SIGH- so must have done something to the old brain ;-) ) Will only post idea 1 2day since it’s long enough trying to explain this one as it is…

To start of, I don’t know if the following already has been done/tried/invented by someone else… if so, sorry didn’t want to ‘steal’ it. Just haven’t seen it yet (still a rookie..).<o></o>

Here goes…<o></o>

I think it is safe to say that after all the optimizing that has been done in the past the TP and SL of Phoenix are pretty much set and reliable (give or take a few pips). My idea is based on when best to EXIT a trade. At the moment <?xml:namespace prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" /><st1lace w:st="on"><st1:City w:st="on">Phoenix</st1:City></st1lace> is either using the SL or the TP. Unfortunately sometimes the exit is a bit early and we could have made a few more pips if in some cases the tp was higher..<o></o>

What if we therefore would incorporate a sort of ‘leading tp'. No idea if something like that already exists but don't know how to call it otherwise (sort of the opposite of a trailing stop..)

Let me try to explain.

For this example lets assume we are LONG.
There NEEDS to be the following condition: current price/profit comes within 5 pips AND ONLY THEN (SO NOT BEFORE) of the originally set TP, which is usually 42 pips. If it then keeps pushing this TP up to prevent it from reaching it.
If we now, at the same time attach a trailing stop (=upwards moving SL that NEVER moves down again to protect profit) of f.i. 5 pips BELOW the current price; it would in theory keep on following a continuing trend upwards until the juice really runs out.<o></o>

We will then only get an exit of the trade at the (hopefully!) higher SL.

Come to think of it… this would make an actual TP obsolete all together because it will never reach this point (being pushed upwards).. An added bonus of this could be that it will prevent brokers to ‘play there tricks on us’ as soon as they see we are about to reach our TP...<o></o>

Just an example in numbers which hopefully makes it a bit clearer what I’m trying to eplain..<o></o>

1) You go long @ 1.0000<o></o>
2) @ 1.0037 (42 pip profit minus x (in this case x=5)) change the current SL into a trailing stop (so protect profit and hopefully increase profit) of f.i. 5 below the current price (again open to discussion) so therefore it starts it's ‘existence’ @ 1.0032 (=1.0037 minus 5).<o></o>

If now the trend continues (see attached picture for an example were I think it could have worked nicely) it will keep following the market until it comes back down and exit at the newly formed trailing stop..<o></o>

The obvious downside of this is of course that if unfortunately after reaching 1.0037 the market turn against us we will only have a profit of 32 instead of the otherwise 42 pips...<o></o>

As I said before the value of 5 pips is only used for this example and aren't based on any backtests. Will have a look in the charts the next few days to see if my above theory makes sense at all whatsoever (told you, I’m ‘a bit’ tired..). <o></o>

And PLEASE give me your feedback!!! Positive and/or negative! Trust me I can take it!

Cheers,

Vince<o></o>
Attached Image
  • #763
  • Nov 12, 2006 11:17am Nov 12, 2006 11:17am
Vince:

You took the words right out of my mouth....

It'll be an interesting week for P5 with all of the red flags on the calendar, we'll see. In the meantime, I battling a massive headcold and have about 10 different drugs in me right now...even this is getting me lightheaded...
  • #764
  • Edited 12:09pm Nov 12, 2006 11:30am | Edited 12:09pm
Quoting Palmer
Vince:
You took the words right out of my mouth....
Meaning? You already tried this? In that case, as posted in the beginning of my post didn't want to 'steal' anyone else’s idea! Kudos to you then! Just haven’t come across it yet!
Funny to see more then one person can have the same idea's across the world! I think the have a term for this phenomenon. Think they call it "collective memory".

Vince<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>
  • #765
  • Nov 12, 2006 1:10pm Nov 12, 2006 1:10pm
Quoting VinceTheBeast
Howdy,<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>

........

If now the trend continues (see attached picture for an example were I think it could have worked nicely) it will keep following the market until it comes back down and exit at the newly formed trailing stop..<o></o>

The obvious downside of this is of course that if unfortunately after reaching 1.0037 the market turn against us we will only have a profit of 32 instead of the otherwise 42 pips...<o></o>

......


Vince<o></o>
Hoi Vince,

If I understand correctly, this would be a delayed TrailingStop?
So the TrailingStop would only be activated as soon as price gets near the TP?
Do you (or maybe onyone else) have any experience in trading this way?
The idea sounds good!

Groet uit een nat NL
(Greetings from a rainy Netherlands)
  • #766
  • Nov 12, 2006 1:43pm Nov 12, 2006 1:43pm
Quoting HerbertH
Hoi Vince,

If I understand correctly, this would be a delayed TrailingStop?
So the TrailingStop would only be activated as soon as price gets near the TP?
Do you (or maybe onyone else) have any experience in trading this way?
The idea sounds good!

Groet uit een nat NL
(Greetings from a rainy Netherlands)
Hi Herbert! Good to see one more fellow countryman! Well, the weather down here isn't much better I'm afraid :-(<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>

I think your way of summarizing makes it much easier to understand! I tend to be able to explain something relatively simple, the hard way ;-)
Added to the above, I would, once near this TP, and after activating this delayed trailing stop, remove the actual TP to make sure it isn’t being hit early.

Maybe I interpreted Palmers post incorrectly but I think he was hinting with his "Vince: You took the words right out of my mouth...." that he already has tried/invented this in the past. Were you Palmer?

In that case could you share your findings please? Would be much appreciated!
Would be a shame to invent the wheel for a second time!

Time for bed!
  • #767
  • Nov 12, 2006 1:49pm Nov 12, 2006 1:49pm
I was getting at the notion of locking in profit at a certain point...that's all. Please note that I am currently wacked on Tylenol, Alka-Seltzer, Mucinex, and Zicam....my thought process is worthless right now....I didn't mean to say that I was stealing somebody's idea....peace.
  • #768
  • Nov 12, 2006 2:18pm Nov 12, 2006 2:18pm
Hi everybody

Ive heard alot of people had success combining mode 2 and 3 together. How does one go about combining modes like that? Ive read the whole thread and I dont remember seeing how it was done. If it has, I apologize. I know alot people posts questions that already have been answered.

Thanks
  • #769
  • Nov 12, 2006 6:49pm Nov 12, 2006 6:49pm
I have been demo-ing pheonix for 3 weeks starting with 10k. I have been up as far as 11.5 k and am now back down to 9.5k.
  • #770
  • Nov 12, 2006 7:09pm Nov 12, 2006 7:09pm
This is my first ever forward testing (and first demo trading) so it is no suprisse I have big mess. I used several different EA, running minimum two at a thime, sometimes, three. So data and numbers arent to acurate. I was messing too much around.

I have statement, but it is probably not usefull. I have there few manual trading, and PhoenixBooster with 8 losses in a row and few other EA's.

Anyway, I collected together numbers for two main EA's I was using last two weeks.

P v4_2_03 generated $1.102
P v5_6_01 generated $1.233 (more than half trades in mode3)

On the same demo account starting with $5.000

Phoenix is IMPRESIVE!!!!

Mario

P.s. just in case I will add statement, but it is more or less useless.
Attached Image
Attached File(s)
File Type: zip FXDD - MetaTrader 4.zip   10 KB | 458 downloads
  • #771
  • Nov 13, 2006 12:43am Nov 13, 2006 12:43am
Quoting Palmer
I was getting at the notion of locking in profit at a certain point...that's all. Please note that I am currently wacked on Tylenol, Alka-Seltzer, Mucinex, and Zicam....my thought process is worthless right now....I didn't mean to say that I was stealing somebody's idea....peace.

Peace! :-)
Just a misunderstanding then! Just me being.... well... me.... ;-)
That’s the problem of English not being my first language I guess...<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>
Hope you get well soon!
Vince
  • #772
  • Nov 13, 2006 2:01am Nov 13, 2006 2:01am
To be honest I’m a little overwhelmed by all these proposals to change/improve <?xml:namespace prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" /><st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City>. A lot of you have very good ideas but what I’m really afraid of is that changing to much of <st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City> will finally leave us with a completely useless Expert Advisor. Believe me; I’ve seen this happen more than once with other EA’s. I truly believe that at a certain point you have to stop making changes. Too much of these and the “spirit” of <st1lace w:st="on"><st1:City w:st="on">Phoenix</st1:City></st1lace> will be restrained too much to do its work.<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>

<o> </o>

I agree that <st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City> is not suitable for all kind of markets. But if the market is right then <st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City> will do its work. Look at the results of the Contest and you know what I mean. The Contest version made 23 trades so far with only 3 loosing trades. Look at the Balance. Do you really want to change that? <o></o>

<o> </o>

You always have the option to switch <st1lace w:st="on"><st1:City w:st="on">Phoenix</st1:City></st1lace> off when its performance gets worse and put it back on when things are changing for the better.<o></o>

<o> </o>

Best solution will be to let 2 opposite EA’s work together. When one of them fails the other one, more suitable for the market at that moment, will kick in and continue the work. It’s not too hard to make an EA that will complement <st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City>. I like this idea so much that I will make a separate project for this. <o></o>

<o> </o>

To make a long story short; at the moment I think it’s not right to make extensive changes to the code of <st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City>. On the other hand some of your proposals are making sense and I would really like to test these changes to see if they can improve the performance of <st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City>. But I can’t do this on my own. For now I’ve only limited time to work on <st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City>, so if there’s a programmer out there that can judge those proposals to see if they are useful or not then this would be very helpful. I just can’t do this on my own! <o></o>
  • #773
  • Nov 13, 2006 2:06am Nov 13, 2006 2:06am
it just speaks volumes about human nature. improving an EA that made 200% profit in lika a month. greeeeed.
  • #774
  • Edited 3:26am Nov 13, 2006 2:15am | Edited 3:26am
I try Phoenix_4_Contest.mq4 in interbank (demo) and Phoenix_EA_v5_6_03.mq4 in Alpari (demo).
Result that Phoenix_4_Contest.mq4 is more profitable..
I don't know why, in my demo account Phoenix_EA_v5_6_03.mq4 is not really profitable..
For me, Phoenix_4_Contest.mq4 is beautiful !!

Thank you Hendrick!

And Congratulation, you are in the top ten again !
http://championship.mql4.com/2006/pa...ants/Hendrick/

Viva Phoenix ! Viva Hendrick !
  • #775
  • Nov 13, 2006 2:20am Nov 13, 2006 2:20am
Quoting Hendrick
.....
To make a long story short; at the moment I think it’s not right to make extensive changes to the code of <?xml:namespace prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" /><st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City>. On the other hand some of your proposals are making sense and I would really like to test these changes to see if they can improve the performance of <st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City>. But I can’t do this on my own. For now I’ve only limited time to work on <st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City>, so if there’s a programmer out there that can judge those proposals to see if they are useful or not then this would be very helpful. I just can’t do this on my own! <?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>
Following the Contest Ratio / Results, and all tests/works (night and days) made by HENDRICK, i think it's a judicious decision.
Keep It Rocking Hendrick.
  • #776
  • Nov 13, 2006 3:00am Nov 13, 2006 3:00am
Hi all. Do I understand correctly that setting ver 5_06_03 to mode 1 will give me the classic version? If true is that with the contest settings as well? I read the manual but have been confused a bit by some posts here so I wanted to clarify.

I got pretty beat up last week using mode 3 and lost almost all my positive equity. Wouldn't you know it was the first week I tried Phoenix on a live account too.

BTW: I agree with what some have said about getting Phoenix in sync. It seems once you get it the right way it doesn't miss and once it is going the wrong way you have to wait for the trades to cycle out.

Thanks
  • #777
  • Nov 13, 2006 3:59am Nov 13, 2006 3:59am
Quoting traderlismiss
Hi all. Do I understand correctly that setting ver 5_06_03 to mode 1 will give me the classic version? If true is that with the contest settings as well? I read the manual but have been confused a bit by some posts here so I wanted to clarify.

I got pretty beat up last week using mode 3 and lost almost all my positive equity. Wouldn't you know it was the first week I tried Phoenix on a live account too.

BTW: I agree with what some have said about getting Phoenix in sync. It seems once you get it the right way it doesn't miss and once it is going the wrong way you have to wait for the trades to cycle out.

Thanks
I need clarification also. On backtesting, if I put the mode to 1, and adjust the settings such as stop loss and take profit I am OK. But when I type in 2 for the mode, and clear the take profit and stop loss in mode 1 and transfer to mode 2, backtesting does not work. Ditto for mode 3.

Does mode 2 work off of Mode 1, or is it totally independent? Thanks.


Also- Does anyone know of a good, reasonably priced programmer for EA's?
"Its not where your at, but where your coming from"
  • #778
  • Edited 5:41am Nov 13, 2006 5:31am | Edited 5:41am
Quoting traderlismiss
Hi all. Do I understand correctly that setting ver 5_06_03 to mode 1 will give me the classic version? If true is that with the contest settings as well? I read the manual but have been confused a bit by some posts here so I wanted to clarify.

I got pretty beat up last week using mode 3 and lost almost all my positive equity. Wouldn't you know it was the first week I tried Phoenix on a live account too.

BTW: I agree with what some have said about getting Phoenix in sync. It seems once you get it the right way it doesn't miss and once it is going the wrong way you have to wait for the trades to cycle out.

Thanks
Melissa- I have found that also about the cycles. I am getting ready to go live with a small amount. I am going to look at the trading championship and wait until Hendrick (Harry) is not "in" a trade. I will then start the EA. This way I will be on the same cycle. I am also going to use the traditional EA from this post.
http://championship.mql4.com/

Matt

Hey Palmer: How is your cold? Man I have tried to learn MQL programming and it is testing my patience. Same thing with learning C. Don't know if I am cut-out for that. I guess I need to drop it for a couple of days.
"Its not where your at, but where your coming from"
  • #779
  • Nov 13, 2006 6:03am Nov 13, 2006 6:03am
Maybe because the contest version uses different settings for USDJPY? (That's my understanding anyway, see my earlier post on this.)

Quoting bellefemme
I try Phoenix_4_Contest.mq4 in interbank (demo) and Phoenix_EA_v5_6_03.mq4 in Alpari (demo).
Result that Phoenix_4_Contest.mq4 is more profitable..
I don't know why, in my demo account Phoenix_EA_v5_6_03.mq4 is not really profitable..
For me, Phoenix_4_Contest.mq4 is beautiful !!

Thank you Hendrick!

And Congratulation, you are in the top ten again !
http://championship.mql4.com/2006/pa...ants/Hendrick/

Viva Phoenix ! Viva Hendrick !
  • #780
  • Nov 13, 2006 6:36am Nov 13, 2006 6:36am
I began 4 demo accounts yesterday running Phoenix 4.x/5.x at 2 brokers as follows. All installations began running within 30 minutes of one another. I don't recall exactly the order in which they were started up, so the A-B-C-D sequence doesn't necessarily represent the first started to the last started.

A) IBFX P5.63 (default Classic settings; mode1, max risk .30, micros, decrease factor 1)

B) IBFX P4 Contest version (default settings)

C) FBFX P5.63 (default Classic settings; mode1, max risk .30, micros, decrease factor 1)

D) FBFX P4 Contest version (default settings)

Interestingly, this morning I check and see the following.

A) Placed a long entry at 117.23 at approx. 01:30 GMT which exited when TP was hit at 117.65.

B) Placed no trades. Remains flat.

C) Placed a short entry at 117.33 at approx. 07:00 GMT which is open right now at a -60p floating loss.

D) Placed a short entry at 117.33 at approx. 07:00 GMT which is open right now at a -60p floating loss.

Conclusions
-------------------------------------
- Oddly we were long at one broker and short at another at the same time (although entries were offset by about 6.5 hours).

- Install B didn't make any trades even though the contest version at MQL4.com did make a trade during this time (long entry at 117.15)

- C) and D) installs placed entries at the exact same point even though I thought the settings were different for the Contest version vs the relased versions (or is there a settings file which the contest version is reading which overrides the default settings in the EA, Hendrick please clarify)

Now, to test the theory that installing and starting a fresh EA at the same time as when the MQL4.com contest version remains flat, I've now installed an additional new installation of the Contest 4 EA as currently the MQL4.com contest version is flat. We'll see if it gets synced up with the MQL4.com version at the get-go.
1 trader viewing now