I think that TS is not working properly, at least in the Trending version of Phoenix I am using.

Attached are two chart shots, one showing a few hours of activity with TS=0, the other TS=5. The TP and SL are both 20. The settings are optimized from visual inspection of indicators and one run of backtesting.

In the TS=0, both sell transactions go out in profit within the same bar as the transaction was initiated. Profit is 20 both times. However in the TS=5 chart, the trades hang around until they go negative, and close on losses. Shouldn't the TS have salvaged the profit, or even a bigger profit beyond the 20 pips of the TP? I am very confused. Is this why backtesting the reversed Phoenix has been so inconsistent?

The report shows the log of events with TS=5. Somebody help!!

Attached Images

	test-ts0-tp20.gif (18.5 KB, 167 views)
	test-ts5-tp20.gif (19.2 KB, 137 views)

Here's the report file for the above message.

Attached Files

TestTS5-TP20.htm (187.5 KB, 39 views)

In Phoenix 5.7.2 I could not confirm this bug. It appears your Phoenix modified version has an issue where the code doesn't confirm the new SL is an improvement over the old SL. The code in 5.7.2 is working properly. I'm working on porting EASY over to Phoenix6.

As I mentioned in another post, the Envelope signal (#1) in Phoenix is problematic. Please compare the attached figures. You will see that with the default setting for GBPUSD, the band produced by the Envelope signal is barely better than a line. There is only up and down, and no in between. The object of the Envelope signal is to create a no trade channel, not a single line dividing buys from sells.

With a more realistic setting (0.025-we ignore the period for the time being) you get a nice band which has an interesting property. It widens when the price line is oscillating (i.e., flat) to prevent any trades (since trades only occur above or below the band) and tightens up when the curve is trending up or down in order to open up trade possibilities.

This function is essentially the same as the Price Action Channel, although the latter may have some advantage over the simple Envelope indicator. Price Action Channel uses a calculated average price (a la Heiken Ashi) which may better reflect the movement of the currency than the simple High and Low we have been using with the Envelope signal. It might make sense to replace Envelope with Price Action Channel, although Envelope may offer some advantage. Because it is not a smoothed average, it follows the curve better at the top and bottom of a trend slope. Accordingly it might help filter out trades when the curve is exhausting itself and about to reverse.

Of course here I am speaking in terms of a trending system. For the time being, I will keep using both until it becomes apparent that one of them is not serving any purpose.

Attached Images

	envelopeperiod0023.gif (11.1 KB, 143 views)
	envelopeperiod025.gif (11.6 KB, 135 views)
	pac5-0.gif (14.9 KB, 129 views)

Wau! I have no words to describe, how I become without words

Result are really really great!

Pass 200, over 24.5 pips per trade with less than 8.5% drawdown?!?
Or pass 115 with 17 pips per trade with only 6.65% drawdown!!!

This is impressive! This sound something like cash machine...

I understand you completely. Hope this EA helps you reach your dreams. You deserve it!

Good work!

Mario

I just finished writing some insights from my studies of the fractals into Phoenix. I didn't add the fractals, I added the emergent property that often matches the fractals. The emergent system is based on Gann's observations back in 1910 of commodities and stocks for determining Support and Resist levels. Gann's work was extended by others over the years, and then I finally made it work with forex. I checked a few days of data, and it lines up with the fractals about 90% of the time. However, the fractals are a much more complete picture than Gann's way of calculating data and Gann's method doesn't work for substantial price movements whereas the fractals do.

alamanjani, you should have received a copy of the demo code, please let me know what you think of it. Some data sources it just can't loose money no matter what crazy settings I use. Other data sources and it looks confused. :/

Phoenix 6, Beta Advanced and Simple has been released in a new thread.

I confirmed this bug and updated. It was my misuse of Mode_Stop.

You may not have used this part of my StopLoss code.

Daraknor

I got errors from the following code because I didn't set the MinLot to the right value.

if(lot<U_MinLot) { lot=U_MinLot; Print("lots switched to min ",lot); } //Dmitry_CH Add 5.7.1
if(lot>U_MaxLot) { lot=U_MaxLot; Print("lots switched to max ",lot); } //Dmitry_CH Add 5.7.1

I did it in the following way,so I would not have to set these variables myself, and it worked.

Add this to the init section.

U_MinLot = MarketInfo(Symbol(),MODE_MINLOT);
U_MaxLot = MarketInfo(Symbol(),MODE_MAXLOT);

This information is entered, by the currency for each broker.
I just kept the extern variables as they are so there would be no change to the settings files.

I created a 5_7_3 version of the code, which I will publish tomorrow.

5.7.3 sounds great, it would be nice to have a new maintenance release.

We can use the MinLot/MaxLot settings from broker if the value is zero, but it would also be nice to give people the option of a manual override. Something like:
if(U_MinLot==0) U_MinLot = MarketInfo(Symbol(),MODE_MINLOT);
if(U_MaxLot==0) U_MaxLot = MarketInfo(Symbol(),MODE_MAXLOT);