Pathfinder DAX 4H

Pathfinder DAX 4H

Hi guys,

I want to share one of my DAX trading ideas based on simple daily, weekly and monthly high/low crossings. I observed that simple cross over and cross under of daily/weekly/monthly high/lows in combination with a multiple smoothed average and some simple filters could be a profitable approach. On the long side the cumulation of orders could be a performance booster for this system.

Comments and suggestions for improvement are welcome.

Have fun



Many other instruments and continuously updated versions are available in the dedicated forum topic of this automated trading strategy, everyone can read and participate here: Pathfinder trading strategy forum topic

Share this

Risk disclosure:

No information on this site is investment advice or a solicitation to buy or sell any financial instrument. Past performance is not indicative of future results. Trading may expose you to risk of loss greater than your deposits and is only suitable for experienced investors who have sufficient financial means to bear such risk.

ProRealTime ITF files and other attachments : How to import ITF files into ProRealTime platform?

New! PRC is also now on YouTube, subscribe to our channel for exclusive content and tutorials

  1. Wisko • 07/04/2016 #

    Hey Reiner,
    im Backtest sieht es so aus, als ob Trades nur zum Open einer neuen 4h-Kerze eröffnet werden. Wie läuft das im Live-Trading? Wird der Trade eröffnet, sobald der CrossOver/Under der Signallinie passiert? Was passiert, wenn der Kurs um ein High/Low Level schwankt und es innerhalb kurzer Zeit mehrere CrossOvers/Unders gibt?
    Ich hab auch noch eine Frage zur Tradingzeit: Trades werden nur zwischen 8 und 22 Uhr gestartet, ich gehe also davon aus, dass auch nur dann die CrossOvers/Unders zählen und nachts stattfindende Crosses ignoriert werden, oder? Wie sieht es mit den Ausstiegen aus? Wenn nachts ein SL/TP erreicht wird, oder die jeweilige Anzahl der Kerzen erreicht wird, wird dann direkt verkauft oder erst am nächsten Morgen um 8?
    Danke schonmal für die Hilfe

    • Reiner • 07/04/2016 #

      Hi Wisko,
      Der Trade Entry erfolgt im DAX nur zur Eröffnung der Kerzen um 9, 13, 17 und 21 Uhr und wenn genau! zu diesem Zeitpunkt ein Crossover/-under vorliegt. Crossings ausserhalb des Trading Windows führen nicht zu einem Trade. SL/TP ist im Life Trading bisher nur im Trading Window eingetreten, ich vermute aber, dass dies generell auch ausserhalb des Fensters passieren könnte, diese Fälle sind aber sehr selten und du kannst diese dann morgens um 8/9 schliessen. Tritt die Erreichnug der maximalen Haltedauer ausserhalb des Tradingfensters ein, wird morgens um 9:00 die Position glattgestellt.
      Auf welche Plattform willst du Pathfinder migrieren?
      Gruß Reiner

  2. Wisko • 07/04/2016 #

    OK, falls es also vorkommen sollte, dass zB der Cross der Signallinie über das Daily High während einer langen bullischen Kerze erfolgt und bei Beginn der neuen Kerze demnach kein neuer CrossOver vorliegt, findet kein Trade statt? Die Signallinie müsste sich doch (wie andere MAs auch) während die Kerze wächst auch bewegen, oder? Hast du das mal beobachten können?
    Ich würde den Pathfinder gerne auf die MT4-Plattform übertragen, da dort Microlots (0,1 €/Punkt) angeboten werden und dadurch ein deutlich präziseres Risikomanagement möglich ist.

    • Reiner • 07/04/2016 #

      die Prüfung der Bedingungen erfolgt ausschließlich beim Open der Kerze z.B. 9:00:00, in deinem skizzierten Szenario wird nicht gehandelt

  3. Wisko • 07/04/2016 #

    Alles klar, danke

  4. ALE • 07/04/2016 #

    this strategy is formidable, the jpeg above is of the first version?

    • Reiner • 07/04/2016 #

      Yes, description and picture is still from V1. We shoult make an update. Current version is available in the forum.

  5. volpiemanuele • 07/04/2016 #

    hi, can i have the latest version of the strategy ? I wont to try it in FTSE 100 cash on IG market. What is the time frame that you suggest me to use ? 4H? And the time zone ? 8 -22 the same of DAX?  What are the variable that I must to optimize ? Thanks Emanuele

    • Reiner • 07/04/2016 #

      Welcome Emanuele,
      You will find the latest version of Pathfinder FTSE H4 V5B2 here in the forum
      Pathfinder was developed for H4 only, trading window for FTSE is 9-22 CET, variables are up to date, please adjust position size to your account size.
      best, Reiner

  6. Wilko • 07/04/2016 #

    Dear Reiner,
    Thank you for posting an interesting system, indeed. I’m looking at it and trying to analyse potential weaknesses. I was a bit sceptical to the numbers you are using for the triple smoothing. I suspected that you had merely curve-fitted, but so far I have run the optimisation myself, and it appears that neighbouring numbers work almost equally well, although your chosen numbers are “optimal”. But this result would indicate that the notion of triple smoothing is not necessarily curve fitted. What bothers me, though, is that PRT backtest, for all its’ benefits, has a couple of serious drawbacks that can give you very different results if/when you go live. One of them is that it appears (at least to me) that PRT backtest is unable to handle backtesting a system that can trigger both a long and short position. I can’t see that the code will in itself not trigger both a long and then a short on a different signal cross-over, or vice-versa. So my question is; Have you tried to disintegrate the system into its different “parts” ie treat each signal separatetely in a system for each signal, to see if you get the same results when you add them up?
    Kind regards,

  7. Wilko • 07/04/2016 #

    Dear Reiner, dear fellow forum members,
    After writing the previous post, I have myself decomposed the system into several independent systems and backtested them separately without the seasonailty boosting. What I found was that each of the parts/strategies independently generated not many trades over the full available database. In other words, allthough the outcome for each part/strategy may be positive, it is based on few observations, meaning the statistical expectation that historical results would repeat is weak. If you merely add a lot of weak parts/strategies together, this does not in itself guarntee the end-result is a strong (or rubust) system. Therefore, I would issue a warning: The parts/strategies incorporated in this system need to be tested independently on a much bigger dataset in order to be able to say with any statistical significance that the integrated system is likely to repeat past performance. In my book, one needs at least 100 observations for each of the different parts/strategies. To avoid misunderstanding, with parts I mean one part/strategy buys break of monthly high, the next part/strategy buys on break of weekly high, and so on. Pathfinder is comprised of 7 different parts/strategies. On top of that filters and seasonality boosts are applied, not uniformly, increasing the risk of curve-fitting. IMHO.
    There are some nice ideas in the code that have been well applied, and for these ideas I thank Reiner!  I’m not saying the integrated system won’t work, I’m saying the certainty that it will work live is not high enough for me. I hope my findings can be of some help to fellow forum members. Please be careful out there! (Quote h/t the tv-series Hill Street Blues from the 80’s)

    • Reiner • 07/04/2016 #

      Hi Wilko,
      Thanks for your feedback and welcome. I will move your comment to the realated forum for further discussion. I think many people are really interested to underand a little bit more your arguments.
      best, Reiner

    • flowsen123 • 07/04/2016 #

      Hey Wilko,
      that is an interesting point. could you upload the results of the “single” test that you made? if I understood you correct, you see this strategy as several strategies that are combined and you tested them separatly?
      Thank you

  8. Wilko • 07/04/2016 #

     Sure. I’d be happy to answer any questions. /Wilko

    • Reiner • 07/04/2016 #

      please answer here

  9. GAMMA • 07/04/2016 #

    thank you for this system, I try to run on down/nd  very good, but when I try on sp500 it not take any trade? try crude oil not good luck ,
    looking forward  to hear from you

    • Reiner • 07/04/2016 #

      Hi GAMMA,
      I didn’t offer a Pathfinder version for sp500 because backtests were very poor.
      best, Reiner

  10. GAMMA • 07/04/2016 #

    i can understand that sp 500 other animal  , do you have for crude oil or brunt oil ?
    thank you in advance.

  11. naqviak • 07/04/2016 #

    Is this strategy converted to Ninja trader 7/8 by someone? If so, can I find the link?

    • Nicolas • 07/04/2016 #

      Sorry, this website is dedicated to prorealtime programming, I don’t any Ninjatrader stuff around here! do you??

  12. Mr Mackey • 07/04/2016 #

    Greeting Reiner, Simon here from London…
    Just joined this site, even though I have no clue to programming, what a great site… I was intrigued by your code…
    I like using DFB spreadbetting here in London for tax reasons so I backtested pathfinder v3 for DAX30 and found fantastic results, I compared it to v4 and the results wasnt nearly as good, so here are my questions:
    I am running v3 on a demo account and it opened a position at £2 per point, how do you change these figures?
    The open position has put a stop nearly 500 points away? I do not understand this as it is using the margin?
    Same for the limit, nearly 500 points away
    When I activated the autotrading, proorder asked for my position size, I inputed 10 per point, but the bet is still 2 per point?
    Only opened it yesterday and the first order is, so far, doing well!!!!!
    Thanks in advance

    • Reiner • 07/04/2016 #

      Hi Simon,
      Thanks for your comment and welcome. You’ll find the last news and releases for Pathfinder here in the forum
      In the meantime we have version 6 running. All settings are controlled by the code and have nothing to do with the margin. Pathfinder is pyramiding the position and the position size is related to a saisonal multiplier. The requested position size in ProOrder is a maximal value in a sense of an overall limit for the whole strategy. By default Pathfinder is trading maximal 15 contracts. I recommend that you go through the forum entries and create a post if something isn’t clear. There are also some UK guys subscribed with the same IG accout features to answer your questions.
      best, Reiner

  13. Mr Mackey • 07/04/2016 #

    Thank you Reiner for responding, I can only find v4, am I looking at something wrong?
    Pathfinder not to be used for DFB?

    • Reiner • 07/04/2016 #

      Simon, what is DFB? Please ask further question in the related forum topic.

  14. demoz • 07/04/2016 #

    Did anyone test this out with a live account?

  15. volpiemanuele • 07/04/2016 #

    Good morning,
    I would use the strategy on my IG account on FTSE 100 futures, rometime (9/22). Someone have already optimized this strategy on this instrument? Please could you give me a link of the ultimate version of the strategy. If no one has optimized the strategy on this market please could you tell me which parameters I have to optimized and the best time frame? Thanks emanuele

    • Bandido • 07/04/2016 #

      Emanuele, vai nel forum e troverai di tutto e di più 😉

    • Reiner • 07/04/2016 #

      Hi Emanuele and welcome,
      You’ll find the latest versions at the first post in the related forum
      Best, Reiner

  16. volpiemanuele • 07/04/2016 #

    Tanks, if I want to enter in a market only with 1 lot do you confirm that I cam modify only this parameters? Also the seasonal logic is bypassed with these parameters? 
    DEFPARAM CUMULATEORDERS = falseONCE longPositionMultiplier = 1ONCE shortPositionMultiplier = 1

    • Reiner • 07/04/2016 #

      Emanuele, please ask in forum in the future. Set position size and maximum position as well. Please be aware that Pathfinder’s performance based on grid orders and without you have another system with much higher drawdown. I’ve posted a Pathfinder DAX version for small accounts (4k).

  17. volpiemanuele • 07/04/2016 #

    Ok, thanks

  18. Master_Code • 07/04/2016 #

    Je ne comprends pas bien ce qu’apporte la V3 par rapport à la V2, car le ratio nb Gain/ nb Perte est légèrement inférieur.

    • Nicolas • 07/04/2016 #

      Un topic de l’auteur du système est disponible ici:
      De nombreuses autres versions et améliorations sont disponibles à cet endroit.

  19. Master_Code • 07/04/2016 #

    Est ce normal qu’un ordre ne passe pas en backtest, mais passe en réel ?
    Ca m’a crée une perte très douloureuse.
    Merci de votre partage.

  20. djtaktik • 07/04/2016 #

    Hello reiner,
    How do you explain that pathfinder have good results on DAX but not on the CAC ?

    • Nicolas • 07/04/2016 #

      You should join and read the forum thread about this strategy. There are plenty of different parameters to adapt from an instrument to another. Since I’m not the author of this code, Reiner could probably answer with more details though.

    • Reiner • 07/04/2016 #

      Hi djtaktik and welcome,
      I have answered your question in the related Pathfinder forum because I want show you a picture.
      Best, Reiner

  21. danver34 • 07/04/2016 #

    is this version the definitive one or from the original one have there been modifications to the code ? Also I see that members have used it on different time fames any time frame that was favourite or is 4H still the best for dax?

Register or


avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar

+67 more likes

Related users ' posts
JanWd Good evening John, Thanks for your new strategy ! Is Exponentialaverage [1] (close) ...
JohnScher Not every day 1 trade, but it works every day
JohnScher I think it is, but my programming skills are very limited, so I worked with the 2 EMA´s
Glen Marquis Sorry John I have no time.
Geronima Ortiz Is this trading system expected overnight? or is it my problem? I see long at 21 hours, wh...
Dymjohn Hi John You can cut the number of trades down by about 23% by introducing a "BuyPrice" STOP...
juanj If I run the backtest with a zero spread the backtested result look similar to your screensh...
Eric If you have a high percentage winners and the trades are closed with take profit the spread ...
Casenova I Agree with you JaunJ, and Yes Eric, what you say makes sense too. Optimizing the Stop Loss...
Daniel da Costa Thanks Doc! Where can you find a service for Dax with a 1 point spread?
Doctrading Hello. I suggest you : ProRealTime CFD, or IG. Best regards,
JanWd Hey Doctrading, I tried your code, optimising the A and V and the RSI period, see below. ...
vlongobardo67 Ciao Fulvio, ho scaricato ed importato il file. m,a se lo faccio girare sul miniDax non ...
FULVIO09 Attualmente non c'è sufficiente volatilità : la condizione "C0 = AverageTrueRange[1500](Clos...
vlongobardo67 Ma io intendevo in backtest ! Scusa non l’ho menzionato.
beeb Hallo Dominik, das mit dem deutsch schreiben kommt mir sehr gelegen. Komme auch aus Deutschl...
vlongobardo67 hi Beeb, I imported the file in my platform but is does open any order. What is wrong , wha...
3 months ago
juanj Kinda reminds me of Renko Trading
Real Pro Hi Nicolas, I'm having trouble changing the colours according to the instructions you gave. ...
Vonasi Thanks for the compliments ALE. If you use any results from it in a strategy then please sha...
CKW this is Awesome tool. thanks Vonasi
Vonasi ....and thank you for the compliment CKW.
Vonasi There is another very minor bug fix. The indicator description when you hover over the speec...
Reiner great work:-)
Vonasi Thanks Reiner. that is compliments indeed from the seasonality expert. I have a second vers...
Yannick TradingDayLong = dayofweek = 1 or dayofweek = 2 or dayofweek = 3 or dayofweek = 4 or dayof...
Marlon Hey JohnScher, your code works fine, but my results aren't as good as yours. Even if I use ...
DarioMazza Bha... this code dont work. gg
5 months ago
ALE P01 , it’s the first patner of the indicator, I hope to add others in future.
ALE We can discuss about the strategy here:
Alai-n Merci pour ce travail...
patmaba merci Alai-n. C'est ma première participation sur prorealcode, merci pour à toi pour ton co...
Yannick Hello There is a mistake in trailing stop code, I think that this should fix the problem s...
Paul About the trailing-stop the way it's coded. That works for backtesting only, just look at th...
Jonny sorry paul, but system says that is not possible to put many stop order combined...what have...
6 months ago
TempusFugit Thanks Bin, nice code, I can use several ideas of the way you build it
DarioMazza Thanks Bin, great concentration, i understand the first idea, but may u indicate me the asse...
Bin Hi Mazza This robot is optimized for DAX30.
Nicolas What do you want to do exactly? Because dots are plotted afterwards..
rama I want to place buy or sell order just after the last man , what is the code to use ? i tr...
Nicolas But it should, please open a new specific topic in the ProOrder automatic trading forum.