stevehopwoodforex.com
https://www.stevehopwoodforex.com/phpBB3/
Print view

James16: the EA
https://www.stevehopwoodforex.com/phpBB3/viewtopic.php?t=377		 Page 3 of 8
Author:  garyfritz [ Thu Mar 01, 2012 9:14 pm ]
Post subject:  Re: James16: the EA
Dietmar, your GU test shows about 90% wins -- but the EU test shows 10% wins !???
Author:  forextrader-radioman [ Thu Mar 01, 2012 9:28 pm ]
Post subject:  confirmation!
garyfritz wrote:Dietmar, your GU test shows about 90% wins -- but the EU test shows 10% wins !???
Hy garyfritz :)

yes, I see that too ...

in EU
10 winning trades with average profit of 10 x 541,24 $ and
94 loss trades with average loss of 94 x 23,43 $

so the single losses are as small as possible ... with that settings was the best performance in that past 9 years ... in D1 timeframe we have only some trades each year. Here in the past years we have had only 104 trades. I tried also to find best (or working) settings in H1 timeframe (just to see more "action") but that was not working for longer period of years.

In GU where other settings better ... you can run a test in EU with the GU settings and see the (poor) results ... and vice versa ...

time to go to bed now :)

all the best,
Dietmar
Author:  SteveHopwood [ Thu Mar 01, 2012 10:00 pm ]
Post subject:  Re: James16: the EA
Guys, please refrain from becoming deflected from useful stuff by backtesting results. They are uniformly bollocks.

To a coder, backtesting is useful in checking the accuracy of the code. To traders, they are useless, for all the reasons discussed again and again in this and many other forums.

If the strategy the EA trades is good, and the EA's coding reflects the strategy accurately, and the strategy does not depend on too much 'hidden stuff' such as 'trader instinct/training/experience' etc, then the EA will be profitable. If any of the previous thingies are missing, the EA will fail. Backtesting in ST is irrelevant to this.

Please do not get involved with a discussion about backtesting. I shall delete such posts from now on, so posters are wasting their time typing out their posts.

:D
Author:  gaheitman [ Thu Mar 01, 2012 10:47 pm ]
Post subject:  Re: James16: the EA
Surya buana wrote:halo everyone this is my first post ,
i learn a lot in this forum, many thanks to Steve and others

i have an idea that estimate lowest or top price for a day using RSI

we need indicator for that to make it easy to calculate

previous Bar price close - currrent Bar price close examp 1 : 1.3420 - 1.3450 = 30 pip

RSI previous Bar= 45 - RSI current Bar = 55 examp 2 : RSI 45 -55 = 10

examp 1 divided examp 2 ( 30 / 10 ) = 3 pip

conclusion : 1 RSI = 3 pip

if RSI 75 = 75-55 = 20 * 3 = 60 pips so the top price estimate = 1.3450 + 60 = 1.3

can someone make this indi it will be very useful
I posted an indicator in another thread that places lines to show different levels of RSI that might work for you. My question for you is how do you know what the RSI will be at the end of the bar?

Here's the indicator. http://www.stevehopwoodforex.com/phpBB3 ... t=30#p8456

George
Author:  SteveHopwood [ Fri Mar 02, 2012 12:08 am ]
Post subject:  Re: James16: the EA
I have added a strategy to James. Please see 'Automated strategies' in post 1.

:D
Author:  forextrader-radioman [ Fri Mar 02, 2012 7:59 am ]
Post subject:  Re: James16: the EA
SteveHopwood wrote:Guys, please refrain from becoming deflected from useful stuff by backtesting results. They are uniformly bollocks.

To a coder, backtesting is useful in checking the accuracy of the code. To traders, they are useless, for all the reasons discussed again and again in this and many other forums.

If the strategy the EA trades is good, and the EA's coding reflects the strategy accurately, and the strategy does not depend on too much 'hidden stuff' such as 'trader instinct/training/experience' etc, then the EA will be profitable. If any of the previous thingies are missing, the EA will fail. Backtesting in ST is irrelevant to this.

Please do not get involved with a discussion about backtesting. I shall delete such posts from now on, so posters are wasting their time typing out their posts.

:D
Hy Steve

Thank you for the advice, thought that would be helpful for someone here ... I will delete that nonsense backtest results and settings about James16EA for the different pairs by myself.

I wish you and yours a nice weekend,
Dietmar
Author:  Surya buana [ Fri Mar 02, 2012 6:53 pm ]
Post subject:  Re: James16: the EA
thanks George

each bar close there is a value of RSI examp:
EU H1 alpari 12.00 bar close 1.3240 Rsi = 24.59
13.00 bar close 1.3233 Rsi = 23.22
1.3240 - 1.3233 = 7 Rsi 24.59 - 23.22 = 1.37 1 RSI = 5.1 pips

if we calculate each bar and make average we can estimate if RSI 20 = 23.22 - 20 = 3.22 * 5.1 = 16.4 pips
EU Rsi 20 = 1.3233 - 16 = 1.3217

this what i mean
sorry steve i just want to explain
Author:  SteveHopwood [ Sun Mar 04, 2012 10:24 pm ]
Post subject:  Re: James16: the EA
Latest update in post one.

Tmhl took an incorrect buy on Friday and I just got around to looking at the code. There is a basic bloop, fixed in post on1.

Coders, find
if (MathMax(high3, high2) - MathMin(high3, high2) > (Tmhl.Pips * Point) ) SendLong = true;

and replace it with

if (MathMax(high3, high2) - MathMin(high3, high2) > (Tmhl.Pips * Point) ) SendLong = false;
Author:  SteveHopwood [ Wed Mar 07, 2012 5:23 pm ]
Post subject:  Re: James16: the EA
Latest update in post 1. No new system yet, but Zeljko spotted a bloop in one of the functions.

Coders, go to CountOpenTrades() and find this code block:

Code: Select all

      //Magic number check
      if (OrderMagicNumber() != Bovb.MagicNumber
          //Next MN condition
         ) continue;
and add this extra line of code - && OrderMagicNumber() != Tmhl.MagicNumber, so the block is edited thus:

Code: Select all

      //Magic number check
      if (OrderMagicNumber() != Bovb.MagicNumber
          && OrderMagicNumber() != Tmhl.MagicNumber
          //Next MN condition
         ) continue;
Nice one. Zeljko.

:D
Author:  fxozgirl [ Thu Mar 08, 2012 7:33 pm ]
Post subject:  Re: James16: the EA
EURJPY BVOB

Buy Trade taken on 5/3, trade went +64 and then went into a bit of a spiral down, only to come back up last night and currently sitting at +84

I have been playing around with TP and decided to put mptm on with BE +5 at 30 pips and JS at 20, so this trade has now closed out at +60 pips.
I will monitor this but it seems most trades get to at least +35, so will see how this goes.
eurjpy_bovb002.gif
All times are UTC Page 3 of 8

Powered by phpBB® Forum Software © phpBB Limited