Profit All Day with Retracement - EA development

Hi All

this is the thread where we can discuss and try out Cubby's Profit All Day methodology with the Retracement enhancement proposed by ReaM.

Edit 1st March 2011 - Other enhancements to the original Profit All Day methodology will also be incorporated.

Created a seperate thread so as not too mess up the original thread and it's readers with this modified version.

In addition the attempt will be to create an automated EA to perform the trading.

Regards
John

Hi All

well the skeleton for the final EA has been created:-

this by no means the final version - DO NOT USE IN LIVE!!!

on this version there is no mm as this code can always be added once the logic for opening and closing orders is there

there appear to be some strange quirks where it opens & closes orders immediately, i am not sure whether this is due to the fact that my modelling data is not perfect or there is an actual bug.

I will need to spend the time to reload the dukascopy data into mt4.

I tested using the defaults that are there on EURUSD (1hr) and since the 20th January it had about 639 pips profit.

it would have been more but for these strange open & closes which cost about 4pips a time

In addition the logic for the closing of trailing orders simply uses an MA cross - i think as fibonacci is used for the entry, we should maybe use fibonacci for the trail

All in all, i think this is something that we can use as a starting basis.

Let's see if together we can actually create an EA that works with relatively low drawdowns

Regards
John

hi good idea on the thread.
Ill contribute when I can. The EA for now does not interest me but keep up the good work as I know some people would love an EA. Will post my trades as they come.
Regards
Autotrader

Thanks for your nice work!
I will follow up this thread.

Hi

attached is the strategy tester report, unfortunately i don't have sufficient data at the moment to give it a full test on lots of data

Highlights - Profit factor 3.00
Max Drawdown 2.02%
Profit trades 77.27%
Net Profit 736.18

The strategy tester report is in the zip file, couldnt work out how to upload it onto the thread as htm;

This isnt using money management and the default settings

I have updated the EA here as well

And to ensure of no accidents i have added code that will stop it running on a live acct.

Regards
John

Hi All

as i believe it will be some time before we can get version of the EA that can be run in live, i have created an interim manual indicator solution.

The indicator should be placed on the timeframe on which you will time your entries, for example M5.

The parameters are as follows:-

xi_Trigger.TF - the timeframe whose fractals you are looking for breaks on
xb_UseOtherFractal - if when the fibonacci is created it should use the other fractal as it's top/bottom
xi_Candles.Back - if not using the other fractal, how many candles back should it look for the top/bottom
xd_FibLvlx - the fibonacci levels being used - set to 0 if you dont want a level
xb_ShowPrice - whether the price should be shown on the fibonacci
xb_ShowExtensions - show the fibonacci extensions
xb_CloseAlerts - switch on the alerts for when there is a close on the entry tf above/below a fibonacci level
xc_Color.xxxx - self explanatory

The levels of the fractal are shown on your entry timeframe, as soon as one is broken an alert will be sent and a fibonacci automatically placed.

Note if a new fractal comes for the same level ie. upper/lower the fibonacci will be re-drawn

IMPORTANT - if you manually adjust the fibonacci on the screen - the Alerts will still come for the original values!!!

Regards
John

The goal is to enter at a better price than the original PAD.
The SL remains the same, your entry is earlier. TP remains the same, but with a better entry price, I guess, about 1:1 RR.

All original rules of PAD apply, make sure you stay with the trend, use higher TF or squeeze the time line.

WE RECOMMEND H1 TIMEFRAME FOR TESTS! Do not go lower that H1!

The 50% retracement does not have to be on the entry candle, and not from the high of the entry candle to the swing low?

One way is to put fibs between the fractals. But it is not correct.


So, the end line: On the screen between the swing low and the high of the entry candle is how you put them.

Ah, I see. Thanks

Hi

I don't particularly like MA's, especially when comparing at different TF's to the one you are trading, they are so open to opinion.

I use a modified version of the strategy which replaces the MA's 3,5 & 13 on the higher TF with CCI.

I have changed the CCI parameter on the trading TF to be 120. I have left the original levels of 90 & -90.

I refer to the 3 zones created as

>90 - Buy Zone
<-90 - Sell Zone
-90 to 90 - No Trade / Reversal zone

So the rules generally remain the same over 90 buy is valid, below -90 sell is valid, however because we are using a much higher CCI there is no requirement to check the next highest TF for trend confirmation.

There is one caveat, if the CCI was in the Buy or Sell zone and crosses back into the No Trade / Reversal zone, this is also an opportunity to Buy / Sell on the break of a high / low. This allows you to profit from reversals which you would normally miss using standard PAD.

Note - you can only enter a trade once on a reversal into the zone, if the trade does fail, you have to wait again for the CCI to break the normal levels.

I will be posting some charts to help clarify the point.

Important - CCI 120 works well using the 5m TF, you may need to experiment with settings for other TF's.

Regards
John

We will start with a perfect example of using the reversal zone to trigger entries

if only every trade could be like like these

NOTE:- The second buy trade would NOT have been taken if the the first reversal hadn't taken it down into the sell zone

that looks amazing

Right now I am looking at that OTF indi, it's cool

Hi

most trading methodologies mainly are based on the entries, leaving when to close up to the trader, or as in the case of the original PAD they use arbitary TP values that have no real trade reasoning behind them.

Most people when seeing a potential trade opportunity coming up they look at the reasons why they should take it, if it goes against them, then they look at why it wasn't a good trade to take.

I do this in reverse, when I see a potential trade opportunity I first look at all the reasons why I shouldn't take the trade or at least take the trade at reduced risk.

One of my main considerations is the risk to reward factor, the original PAD system has a 2:1 Risk/Reward ratio, which is fine - however for me, defining a take profit based on your stop loss doesn't make any sense in relation to how far the market (hopefully) will move in the direction of your trade.

Lot's of people (myself included) use Pivot Points (whichever ones float your boat standard/fib/camarilla etc) as one method to define potential hazards to the profitability of your trade.

In addition I use previous Highs & Lows achieved at different timeframes.

As my screen was getting rather cluttered with all the support & resistance lines, I designed a new indicator that lists simply the Highs & Lows of other timeframes.

Timeframes to be included are selectable, as well as if you only want the last High/Low achieved or the last 2 High/Lows achieved.

The potential support/resistance areas are listed on the right of the chart as well as with little line markers 2 bars in front of the zero bar.

When looking at a new trade, I am still looking for a minimum of a 2:1 risk / reward ratio, however to calulate the reward I look to where the next possible support/resistance level may be, whether that be a Pivot level or a previous High/Low, if I can still get the minimum of my 2:1 then I will enter the trade.

Occasionally this stops you entering what could have been profitable trades but more often stops you entering trades where there was little chance of your take profit being hit.

Regards
John

Thanks a lot for your effort!
btw, I tested the pad retracement EA on some pairs, seems the results aren't very promising. It may need some refinement or the idea behind it needs modification. for me, the original method is good enough. still, Thank you very much for creating all these good indicators and EAs.

JMW,

Can you alter the EA and add the original strategy code? I would like to use it as I miss a lot of opportunities when I leave my desk for a few mins or so.

Thanks,

Hello

I'm using my own modified version of Cubby's original, if anyones interested this is my setup.
I only ever use the H1 or above as I find the lower time frames just too unreliable.
I'm only looking for a 123 reversal.
There must be at least 2 bars between 1&2 and 2&3.
I use Fibonacci for 2 indications, at least a pull back to the 61.8 level (using 1 as the 0% and 2 as the 100%) and also for profit levels, I'm looking for a 200% up from the original swing (100% profit from entry point 2), if the 200% (from point 1) is hit and the trends looking strong I move up my stop and let it ride.
I want to see the CCI below or above the 90/-90 at the point of entry and ideally a cross of the ADX in the direction of the trade and between the 20/40 level although this is flexible, if it looks near I'll take it.
I also want the ma's looking in the right direction (because I'm trading on a reversal I'm not looking at the ma's in the higher time frame).
If all these conditions are met I feel pretty confident of a 100% profit, using 1 as the stop (+/- 3 points and spread)
Attached an ideal trade (notice how the price pulls back to nearly touch the 61.8 then carries back down)

Hi,

I have a version with the original strategy code (mainly used for comparison) - it is not profitable at all at lower TF's (almost causes a margin call).

I will be posting these test EA's, however they are only for indication of whether or not the basic strategy is sound.

Regards
John

hmm my tests with the EA ended up positive most of the time, my modeling quality is 90% though, not 99%.

oh yeah, I read that 1 2 3 thread, they put fibs and wait for the price to go at least to the 62% level.

So I guess youve entered where the price went through the 100% level at the number 2 when it retraced off 3?

You took the trade slight off to the left from the middle of the screen, right?