I was hoping to start adding Signal based exits (RSI and ParaSAR) that work in this release, but I'm going to save it for next time since they aren't really part of the Phoenix Strategy.
Trading hours that start AT at a time and continue TO a time are now working. I have a few features I know don't work but the error messages were vague and the code seems to make sense. For example, why OrderLots() doesn't return the lotsize for closing a trade is something I will need to resolve later on, with a lot of debugging.
If optimization is done in P6 simple, the entire settings will now be displayed when the settings are copied out. This is important if we want to know what settings were used for an optimization. True/False settings still aren't shown, but we're already at the limit and MT4 doesn't support that. We'd have to set them as integers, with values of 0 or 1 to make them display.
I'm switching gears a little and working on the ECN code and the lotsizing functions. ECN is now storing price data, although only in a rudimentary form so far.
I'm still working on some settings that make the Advanced system 'advanced' and hope to have it done later tonight. Here is P6 B4 Simple, it is not settings compatible with prior versions and it has USDCHF by default - a currency with settings that have been profitable in live trading.
Please keep in mind that P6 B4 has new settings enabled by default, including no Friday afternoon or Sunday trading. These settings are specific to the GMT+2 FXDD data. If you are using IBFX or another broker, you will need to adjust the hourly trade settings.
Presentation + Regret + Suggestion
I would like to present myself to you. I was working in firebird (which phoenix came from) since 04/2006 just with a friend of mine. Two months ago I decided to look at the firebird community's work. We are very happy about so good work we found here. Congratulations people.
I had just finished to read all threads about firebird and phoenix, here in forex-tsd and at forex factory. It was almost 7,000 posts. I'm really tired and I collected a great material. For the next week I'll relax in a farm for vacation and think deeply in the essence of our EA. So I'll be able to contribute with ideas and comments to the community.
First of all I'm really regret to be away of the community so much time. I felt that I could had colaborated too much. I need to ask you to excuse me.
For the second regret, I'm sorry because I'll not be able to contribute with tests. I have a weak PC and a poor Internet connection. I hope you understand.
Almong all the material collected at almost 7,000 posts I would like to share a special one (at least for myself) just to start sharing.
This techinique is able to replace all the features we had implemented about "given a trade signal, when is the great momment to entry the trade". The chances to reach SL become lower and the TP can be higher.
I know that you have too many interesting things at TO-DO list, sorry for add one more. But I really think it is going to be very good for our EA.
Hi emilson, welcome.... or should I say welcome back! lol
Don't worry about testing, we have good people that can do the testing. Right now, daraknor and Pcontour seems to be the only coders we have, so I'm sure your coding addition will be greatly appreciated.
Have a relaxing time on your vacation and looking forward to your help with Phoenix
Welcome Emilson, always great to hear from old hands coming back. I also experimented with different TP levels, and I think it might be good testing to set SL as a function of TP, and then just optimize TP separately. It prevents some of the silly curve fitting in backtest, but only some.
One of the key concepts I'd like to address in P6 Advanced is the idea of a "viability window" for a signal. If we're dealing in small intraday movements, why wait 2 days? The signal was missed, we should exit and move on to the next trade. It seems like the TP18 SL40 policy is very similar in mentality but only applied to price.
I also pioneered some different TS strategies developed from suggestions in the thread, a lot of research and experimentation. The result is an average profit with TS than by TP and a higher Win:Loss than with TP/SL.
I organized Phoenix 6 in a way that the core code doesn't need to change for new Exit Strategies, Entry Filters, or Entry Signals. I'm going to make a few changes and then Phoenix-as-code can run multiple strategies including Phoenix-as-system plus alternatives side by side.
daraknor, the link you have to P6B4 on the first page is wrong, it is linking to P6B3 at the moment.
I'm recommending people who were running all previous versions of P6 Beta upgrade to this version. I added a signal that Wackena suggested, so P6 no longer enters trades during a consolidation (no accidental breakout trading) and I optimized the data for USDCHF. The signal (Filter 1) is on by default.
Parabolic SAR as an exit strategy is now working properly, but I don't recommend using it with our counter trend strategy. I limited the trade to only close when the trade is in profit, but this was only a slight improvement in performance.
The main reason to upgrade is that some brokers increase the MODE_STOPLEVEL during news trading. FXDD changed the minimum point to adjust stops to 15 pips, up from 4 pips on majors. Because of this, the sliding trailing stop didn't work, and I manually adjusted the trade to the best of my (allowed) ability during the trade. The new version will automatically correct the MODE_STOPLEVEL scaling during a trade for both TP and SL pip adjustments.
I complained that the MODE_STOPLEVEL adjustments were not made on the demo server, and I suggest that others do the same. Demo should match Live in every way possible. Sorry I didn't catch this earlier - please watch moving SL & TP during news releases to make sure it is working properly. You can see if the MODE_STOPLEVEL has been adjusted by doing: CTRL+M, right click a currency pair, click SYMBOL, Properties, Stops Level.
Phoenix 6 Beta 5 testing
Beta 5 in test USDCHF M15 - Kudos Daraknor!
Thank you very much for doing testing GetDown!
I did some optimizations last night (several rounds of testing over different dates and honed in on some favorite settings). Here are the new settings for P6B5, I don't recommend these with any prior versions because it relies on the new filter for the trades.
I'm making these settings live with PhoenixFund. I'm going to keep a test running using the old settings so I can compare side by side. Live is using these new settings though.
Should I forward test Phoenix6 Simple beta5 by default set or the p6b5 usdchf best.set?
I'm testing the beta 5 with my IBFX Mini live account with your posted settings.
IBFX time is GMT. Does this mean I have to change all the time settings from your preset -2h?
Thanks for your great work, hope we can help a bit.