When backtesting P.CONTEST I can't get the result the contest is showing (28 476.80 Equity). I wonder why...backtesing always shows way worse results.
Anybody else experiencing this?
Anybody else experiencing this?
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
Quoting tradeigelDislikedWhen backtesting P.CONTEST I can't get the result the contest is showing (28 476.80 Equity). I wonder why...backtesing always shows way worse results.
Anybody else experiencing this?Ignored
Quoting HendrickDislikedHi Tom!
Well, it seems that you are one of those that have better results with Mode 3 than others. I think we will keep this Mode <?xml:namespace prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" /><st1:metricconverter w:st="on" ProductID="3 in">3 in</st1:metricconverter> for now. Tomorrow is Sunday so I have the whole day to investigate Mode 3 because I think it needs improvement! Thanks for your contribution Tom (what are your “testing-plans” for next week?)<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>
<o></o>Ignored
Quoting edcheongDislikedHi Babarmughal,
Just check the very first post's attachhment of this thread. The idea is to use the first post for the latest updates.
edIgnored
Quoting HendrickDislikedDid a lot of experiments today to further developing Phoenix. I think I found a nice addition. Imagine this: as soon as the first trade (TP=42, SL=84) is a certain amount of pips in profit (say 32 pips) that will be the sign for to place a second trade in the same direction. But now with a TP of 100 and a SL of 100. I did some backward testing for this. Phoenix_5_Pure.htm with the usual settings and Phoenix_5_Double.htm with the second trade option and with a setting that closes a trade after 62 hours. Look at the difference.Ignored
Quoting billworldDislikedComparing these results Classic looks better to me for the following reasons.
1. Even though Classic (Mode 1) yields a lower return when trading the same number of lots as Mode 2, it's max drawdown is 1/2 that of Mode 2. So, if you double your lots on Mode 1, the overal return doubles and is actually better with Classic ($36.2k vs $35.5k).
2. The Win % of Classic (80/83%) is better than Mode 2 (67/73%).
3. The Profit Factor of Classic (2.36) is better than Mode 2 (1.75).
All these things point to Classic being the better performer.Ignored
Quoting HendrickDislikedJust a quick recap of the development of the next version of <st1lace w:st="on"><st1:city w:st="on">Phoenix</st1:city></st1lace>. <st1:city w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:city> 5 has some nice additions. One of them is opening a second trade (in the same direction) when the first trade is a certain amount of pips in profit. The first trade is using a low TP and a much lower SL than a TP of 42 and a SL of 84. Consider this first trade as a scout that is exploring the market. If the market is not right for <st1:city w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:city> the first trade will end in a loss. This loss is however limited due to a low SL. But when the scout is in profit the second trade (with a higher TP and SL) will open with a fair chance of becoming a profitable trade (and so is your first trade).<o></o>
<o></o>
Other additions are:<o></o>
<o></o>
- closing trades after a certain amount of hours (seems to work for a couple of pairs).<o></o>
- moving your SL to breakeven when your trade is in profit.<o></o>
<o></o>
... And so on and so on.<o></o>
<o></o>
I think <st1:city w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:city> will be more profitable in the long run. But one thing: I need your support!! <o></o>
<o></o>
And now a very small critical remark:<o></o>
<o></o>
When the new <st1:city w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:city> thread just started there were a couple of visitors that were very enthusiastic. I haven't seen these visitors lately. And that’s not surprising at all, because they are "the takers": People that are willing to try a new EA for a week and hops to another EA when the first one isn’t profitable within the first couple of days. Well, goodbye to them, we don't need them! And that's why my love and my gratitude goes to the regular visitors of this thread. They understand that developing a good Expert Advisor takes a lot of time. I hope you guys will continue your support for the coming time!<o></o>
<o></o>
And let be honest, in the Contest <st1:city w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:city> is still occupying the fourth place out of 260 EA's! So it has some potential. And together we will develop <st1:city w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:city> further into a very good EA. Let's go for it!!!<o></o>Ignored
Quoting SkyhookDislikedAnyways, as far as Phoenix goes, I'm finding that I'm getting better results on my demo account with Phoenix running in mode 1 over the others. While running in classic mode, it's cleaning house. Mode 3's "easy-out" losses just keep dragging things down, hurting the average. Then again, this is only my mileage, your mileage may vary.Ignored
Quoting billworldDislikedI'm curious on how Classic v4 and Classic v5 compare with one another. I'll likely go ahead and set up some forward tests on this and see how if fares after a few hundred trades. .Ignored
Quoting pabloskiDislikedHendrick,
I just have a question about the signals Phoenix generates. During this week I've seen that depending when I oppen the EA, the buy/sell position opened varies. For example, this week I opened the EA in one computer and a few minutes later it opened a sell position in USD/JPY. Then I decided to start Phoenix in my second computer and a few moments later it opened a buy position in USD/JPY.
So I had a buy and sell position for USD/JPY in two different computers at the same time!
In the comments on the Championship website, someone called "Freaky" wrote this:
Hi Hendrick,
I use your EA since 2 weeks. It gives me 8 TP in a row! GREAT.
I've made an indicator showing the signals made by Phoenix.
Here it goes:
If you see the picture and if those signals are right, it seems that Phoenix can give two or more opposite signals with little time diference. And this is similar to what happened to me during last week.
Could you clear this to me?
Thanks a lot!Ignored
Quoting corner_hDislikedHi there,
I feell like I am tuned to this threads after reading it for a few hours
and looking for a little feedback
Thank you for clarifying that post about you 'going private' and I understand now that
inspiration you have created will stay where it is now.
I found the place with your buddy buddies (FF) and it is inspirational I'll tell you
About ideas for new developments in Phoenix
I just have this wild gess that you already maybe doing this in Phoenix
as you mentioned about 'your little secret'. Sure averybody have to have some
and I do not want to break it.
I just had these in may head and do not know how to (MQL)
.
So... It is about natural laws again.
As you have signal1.. signal2..signal 3....
I am gessing that combination of these sinals let you follow natural laws of repeatable patterns
and it is just have to be timed in sync with real pair price movement in 'wave like' fashion
It is great and will lead to great success as long as we in sync.
I am maybe totally off on this but this is the reason to have all timeframes signals to see whole picture
as m15 maybe missing the view of some big turn.
Still looking for your feedback on higher time frame Signal(s) within m15
As I understand Phoenix development will continue ...thank you very much
My post #620 about this on f-tsd did not really touched anybody
Just tell me if it is so 'out of line' that I will not become annoying and go some place else
One thing to say 'No' and another to ignore it
Maybe it is not possible in MQL technically??
thanks again hendrick
I am at your feet
and you can write to corner_house at fastmailDOTfmIgnored
QuoteDislikedWell, I think you're absolutely right about the synchronization 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> with "time". I think that the Contest version is now completely in sync and this explains also why others don't get the same results. For me it's still a mystery how to get <st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City> in sync. It has maybe to do with the time you put the Phoenix EA to your chart. I've noticed that the best time to do so is Sunday before the market opens or start <st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City> an hour after a big news event. So picking the right time to start <st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City> seems to be very important.
Quoting Trader83DislikedHendrick, I've been studying during last week this problem of when Phoenix should be put on a chart with the help of two terminals and 16 charts. There can be difference if you put EA now and 15 minutes later, for example. However, this difference dissapears in 2 or three trades. After 1 trades, now Phoenix 5 on my computer opens positions exactly as Phoenix 4 Contest in the championship.
This can be explained in the following way. The signals for entering positions appears very often, however if Phoenix has already opened a position, it ignores all signals until this position is closed.
So, you have one position. When you put another Phoenix it opens position, that is different to the first one. However, later, the first position or second one is closed, and if there are not any new signals for entering and the last position is also closed, new positions will be opened at the same time by the same signals.Ignored
Quoting jalpDislikedHi Billword
Correct me if I'm wrong but as far as i know, Classic v4 and V5 are exactly the same except that V5 has less pairs optimized.Ignored
double maF1, maF2, maS1, maS2;