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

Reiner

 

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 • 230 days ago #

    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 • 230 days ago #

      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 • 230 days ago #

    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 • 229 days ago #

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

  3. Wisko • 229 days ago #

    Alles klar, danke

  4. ALE • 229 days ago #

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

    • Reiner • 228 days ago #

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

  5. volpiemanuele • 218 days ago #

    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

  6. Wilko • 215 days ago #

    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,

    Wilko

  7. Wilko • 214 days ago #

    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 • 214 days ago #

      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 • 214 days ago #

      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

      Flo

  8. Wilko • 214 days ago #

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

  9. GAMMA • 181 days ago #

    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 • 181 days ago #

      Hi GAMMA,

      I didn’t offer a Pathfinder version for sp500 because backtests were very poor.

      best, Reiner

  10. GAMMA • 181 days ago #

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

    thank you in advance.

  11. naqviak • 179 days ago #

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

    • Nicolas • 179 days ago #

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

  12. Mr Mackey • 172 days ago #

    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

    Simon

     

    • Reiner • 172 days ago #

      Hi Simon,

      Thanks for your comment and welcome. You’ll find the last news and releases for Pathfinder here in the forum http://www.prorealcode.com/topic/pathfinder-trading-system/

      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 • 172 days ago #

    Thank you Reiner for responding, I can only find v4, am I looking at something wrong?

    Pathfinder not to be used for DFB?

    • Reiner • 171 days ago #

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

  14. demoz • 161 days ago #

    Did anyone test this out with a live account?

  15. volpiemanuele • 133 days ago #

    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 • 133 days ago #

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

    • Reiner • 133 days ago #

      Hi Emanuele and welcome,

      You’ll find the latest versions at the first post in the related forum https://www.prorealcode.com/topic/pathfinder-trading-system/#post-13565

      Best, Reiner

  16. volpiemanuele • 130 days ago #

    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 = false
    ONCE longPositionMultiplier = 1
    ONCE shortPositionMultiplier = 1

    • Reiner • 130 days ago #

      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 • 130 days ago #

    Ok, thanks

  18. Master_Code • 120 days ago #

    Bonjour,

    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.

    Merci.

     

  19. Master_Code • 120 days ago #

    Bonjour,

    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 • 104 days ago #

    Hello reiner,

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

     

    • Nicolas • 104 days ago #

      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 • 104 days ago #

      Hi djtaktik and welcome,

      I have answered your question in the related Pathfinder forum because I want show you a picture.

      https://www.prorealcode.com/topic/pathfinder-trading-system/page/55/#post-24973

      Best, Reiner

  21. danver34 • 85 days ago #

    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?

avatar
Register or

Likes

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

+58 more likes

Related users ' posts
ALE
1 week ago
CSR strategy DAX 1 D
CSR strategy DAX 1 D
12
Strategies
ALE Hello GuysI'm trying to fix a problem that I missed during the probacktest, I update you as ...
Nicolas The code has been updated in the attachment of the post with the last version which contain ...
ALE Thanks as always Nicolas
juanj Here are another small tweak (just check Flatafter time as I am in a different timezone); /...
juanj Also change the ATR multiplier to 2 instead of 3. Slightly better drawdown.
GraHal I started he thread below as I get the results shown on image below. Please help on the Foru...
Francesco78 Hi Keemax, I dont have it on real at the moment, in any case the strategy is very long term ...
Francesco78 I did a little bit of work on that and now the results looks better and more stable. Please...
Francesco78 I did a little bit of work on that and now the results looks better and more stable. Please...
Francesco78 Ciao JR  Francesco here not Ale :)) thanks so much for checking the code, I will have a lo...
JR1976 Ok sorry Francesco :-)  
Francesco78 These are the comparison with my lates version and yours, I think it improves by all the mea...
bjoern You are absolutely right, thank for your comment. Already working on an optimization while u...
bjoern Added a range filter for entering positions. Performance seems to be much better, even with ...
Nicolas Thanks!
ALE :-) Wim Thank you!
GDKLockout Hi, Im new to this forum and automated trading, so please excuse the nooby question: If you...
ALE Hello because this trailing stop start next candles if condition is verified.  For more in...
Nicolas What are you calling a finished strategy? This one has been coded by a request on forums, si...
clanli Hi Nicolas Thanks for a fantistic piece of code. I tried the following with really good res...
rama where is Z variable in the program? i want to have look back period says 3 candles. I want...
Elsborgtrading I have moved the development topic https://www.prorealcode.com/topic/development-of-initial-...
Elsborgtrading Sometime along the way since I coded the Indicator something has change in PRT- at least I t...
Elsborgtrading Btw last code was a 10 min TF ver. 
Derek Nice strategy. Have you tried adding a stop loss since there are a few sharp drawdowns? I ...
Piston_Broke Non so .... da qui la mia domanda iniziale :-)
Piston_Broke Hi Derek. I have tried many different ways to apply SL's to this and similar versions of th...
Nicolas Indeed, if you are not willing to loose, you will always win. Averaging down losers can carr...
David You're always safer going Long especially with averaging down on an index as the probability...
Oskar Bergvall  I noted Davids and Nicolas remarks. Could it be possible to make an indicator for contraria...
bluetime6 Hello Nicolas. Can i ask you for something? I backtested your strategy in Dax 4H Timeframe...
Francesco78 Thank you Nicolas,  I have copied your nice indicator, but it shows in the lower part of th...
Abz  hi , you need to add the indicator from the "price" menu in chart
Francesco Thank you Abz!
Lotar
3 months ago
Lotar Hi Plop61, thanks for your comment. To make the code for short positions is possible. But ...
JvdG Hi Lotar, your strategy raised my interest; I did some tests and made a few adjustments fit...
Nicolas You can send it to contact@prorealcode.com and I'll attach it here. Thanks for your sharing!
johnymon Hi Reiner, Many thanks for the  superb strategy.I just tried to back-test on prerealtime bu...
Reiner Hi Joy John, Navigator will allways sell at 9:00 at the first tdom and will accumulate a lo...
dajvop Thank you for an excellent system Reiner. Started it April 7 and it really showed what it ca...
SAM
3 months ago
Elsborgtrading By adding the inside bar breakout failure candle, which is the next bar after the inside bar...
Wing Yes, investigate as much as you want. For more insight, you can view the linet1, linet2 etc....
CKW Hi Wing, Thanks for your sharing. I am still trying to breakdown & understand your code...
Wing Hello CKW. No, the parameter, 7 in this case, is used when calling the RSI indicator to ide...
Nicolas For everyone information, a discussion thread is ongoing here: https://www.prorealcode.com/t...
larouedegann Bonjour Nicolas, comment se fait-il qu'avec cette stratégie stop loss à 5 pips, PRT mettant...
Nacho Buenos días Raul, tengo puesto este sistema desde hace unos días en una cuenta demo en la ve...
pascal3431 Salut Doc , Je ne vois pas ce qui fait office de stop dans ce code .. un retour dans le nua...
Doctrading Bonjour, Il s'agit juste d'une stratégie où on est toujours en position : soit à l'achat / ...
pascal3431 Bonjour, après quelques essais sur EUR/USD au M15 en rajoutant(car sinon on est en perte) u...
Doctrading Nice Job ! Here is my version of the code :  KBO = 0 Tenkansen = (highest[9](high)+lowe...
gefinance Thanks for this code. The only thing left missing is the time lapse, otherwise, lots of old...

Top