Not sussed it all yet (one eye on a film doesn’t help! 🙂 ) but it has proven one thing to me … the same code CAN produce vastly different results on CFD from results on UK Spread Bet (SB)!
Trade on attached is the SAME date & time but the trade develops so different on CFD than SB … due to use / misuse of pipsize.
For EURUSD …
Pipsize on UK SB is 1
Pipesize on CFD is 0.0001
Grahal said Results seem too good to be true?
Did you have check the box tick ???
can I see your codes for to try understand
Did you have check the box tick ???
can I see your codes for to try understand
Yes tick by tick box was enabled … first thing I checked also! 🙂
Attached are the 4 versions. Version Ro(3) is a misuse of the pipsize (I think?). I was trying suggestions made by Roberto, but I probably misinterpreted what he said.
Let us know any findings / thoughts please.
Ro2 and Ro3 attached due to limit of 4 files per post
- Fractal-BOut-EURUSD-1H-Ro2 is correct with REINVEST=0 (should you set it to 1, you’d have to remove P from the final SETs or change the SL & TP calculations at lines 38-39 by adding * POINTSIZE
- Fractal-BOut-EURUSD-1H-Ro3 is correct in ALL cases, provided you remove the P in the final SETs, since all SL’s & TP’s are now expressed in price difference rather than pips. With REINVEST=0 it is exactly the same as the file above, but after removing P it should be preferred since it will work with ANY instrument
As Nicolas pointed out, POINTSIZE/PIPSIZE is not necessary with instruments where the ratio price to pip is 1 (like Dax), but necessary when it is 10000 (like currencies). To make code ALWAYS work, use POINTSIZE/PIPSIZE whenever possible and let ProOrder do the math.
Have fun GraHal with your toys (you can still use your grandchildren’s, occasionally)!
I prefer TP’s greater than SL’s and found out that changing them to
TakeProfit = 70 * pipsize //change
StopLoss = 40 * pipsize //change
grants better results.
wp01Participant
Master
GraHal,
Thank you for your postings.
In version Ro3 there are multiple trades where it is months in position. In attached overview you see two examples.
In 2011 there is even an example of a position of 7 months (not shown). Due to the overnight costs IG charges i do not think that that is the intention.
Grahal and Roberto
Thank you for your messages
Backtests on gbp usd seem to me good
Pipsize, pointsize, Nether it becomes complicated for me
In version Ro3 there are multiple trades where it is months in position.
Version Ro3 is flawed for the reasons Roberto states re TGL / TGS etc … this brings the Take Profit down to near zero and so the strategy stays in for weeks / months until it gets back to a small profit. Correct me if I have it wrong and you see it different?
(the downside is that over those weeks / months, trades going the other way may be missed!?)
I did not understand the code / strategy, but I do now after me going to sleep thinking over Roberto explanation. This does help I find as one’s brain works on it overnight and I wake understanding it better. Thank you Roberto for being patient and explaining two ways.
I apologise for confusion due to me … playing with my toys (as Roberto put it! 🙂 🙂 ).
When I can’t get a strategy understood in my head (often due to numerous loop backs in the code? ) then I make a change (logical / semi-logical) and check results difference and try and suss out the reasons why. My trial and error method resulted in version Ro3.
I hope other less than expert coders can learn from the contributors of the last 12 posts or so? If nothing else, my questions have revived this Thread / Strategy and it can be seen as very robust and well worth further tests on different Instruments / Markets and Timeframes.
Many Thanks
Ciao. Recentemente sono passato a ProRealTime da VisualTrader e sto cercando di imparare la programmazione….. Sto studiando il tu Trading System con le originarie modifiche di Elsborgtrading. Chiedevo per favore se qualcuno poteva spiegrami perchè ci sono giorni interi in cui il sistema non entra in posizione e cosa significano queste due righe di codice “Capital = 10000” e “MAXpositionsize=5000
MINpositionsize=1”. Grazie
Please speak english on the english forum.
Thanks.
Hi guys, if some of you are still interested here is a backtest of the version 4 of EURUSD 1H, it seems to be plateauing
Hi Morgan, please update your country flag in your profile 🙂
Plateauing, but not losing! It is usually a sign to make a new optimization, because market has changed its behavior in the last periods and it was not expected by how the strategy was designed.
Hi there,
I’m from Taiwan and thanks for all the program you guys provided.
I tried to run the program but can’t find out a best result like yours.
May I confirm what the parameter I need to fine tune?
the value CP, target profit/lossprofit or others?
Thank you!
Mason
Can anyone help with coding a fractal indicator that will draw a horizontal line based on Williams fractal that would help quickly indicate a break of market structure, where a break of market structure is based upon a candle close above the previous fractal? The horizontal line starts at the upper or lower fractal and continues into the future until a future candle closes above or below the respective up or down fractal. This is the first building block to the strategy. The next building block would be an entry below a 50 percent pull back from the break of market structure based upon a candle close past the fractal break of market structure. I have attempted to code such an indicator but new to coding.
Thanks
