Simple “Dax Breakout” trading strategy code

Simple “Dax Breakout” trading strategy code

Hello everyone,
Here is a small breakout strategy.
Here, I tested with the following settings:

– Range (channel) defined between 09H at 9H30

– On the DAX, 15 minute timeframe

– We play the breakout of high / low of the range

– Stop Loss: the other side of the range

– Take Profit: equal to the stop loss

As you can see, the code is simple.

The backtest is positive, even if it is not optimal.

Indeed, the max drawdown of € 6,548 (for an initial bet of € 10,000), for a capital gain of 51.80% per year. is too high.

We can decrease the contract size by 2 (max drawdown of € 3,274 for a capital gain of 25.90% per year), but the capital growth curve is still quite irregular.

However, this strategy seems logical and reliable, so I think there is much way to improve it.

It also works on CAC40 (or may some other indices).

But paradoxically it doesn’t work well on forex (I wanted to write it for forex, in order to use it automatically).


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?

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

  1. Eric • 04/08/2016 #

    Have you tried to double the size if the first trade is a loss?
    I mean if the first breakout up is false then double the size on the next breakout down
    i would try it myself but i cant code it

  2. Doctrading • 04/08/2016 #

    Hello Eric,
    Of course it is possible. But for my backtests I don’t use martingale.
    Martingale is an easy solution to make you believe that your strategy works all the time… But the only time you loose, you loose much or all your money !
    Backtests without martingale are true results.
    Best Regards.

    • Nicolas • 04/08/2016 #

      Hello Doc, do you have tested the strategy with transactions costs also? Thanks.

  3. Doctrading • 04/08/2016 #

    Hi Nicolas,
    You are right, as it is in 15min timeframe, transaction costs are important.
    I will make the test later 😉

  4. Doctrading • 04/08/2016 #

    The same code with 1 point spread gives the following :
    A little ugly at beginning…
    I made no optimization for this code.Later, I will try to create a good breakout code (like “London Open Breakout” for GBP/USD, GOLD, DAX, etc.)

    • johnymon • 04/08/2016 #

      // Definition of code parameters
      DEFPARAM CumulateOrders = False // Cumulating positions deactivated

      j1=(DHigh(1) - DLow(1)) * .236
      j2=(DHigh(1) - DLow(1)) * .50
      // Conditions to enter long positions
      indicator1 = close //realtime price
      c1 = (indicator1 CROSSES OVER DHigh(1))

      IF c1 THEN

      // Conditions to enter short positions
      indicator3 = close
      c3 = (indicator3 CROSSES UNDER DLow(1))

      IF c3 THEN

      // Stops and targets
      SET STOP pLOSS j2
      // Stops and targets

      Hi Doctrading,
      I started using proreal time via my IGMarkets website and trying to code a similar breakout strategy  system  .I am finding difficulty in running  the system between only dax trading hours(07:00 to 19:00 UK time).I take previous day high/low and if  today’s price breaks  yesterday’s high/low , I bought /sell with target as (23.6% of yesterday’s high -low difference) points. The stop loss for both is (50% of of yesterday’s high -low difference) points.I wrote the code as attached.Could you please review the code and include the  trading hours restriction.  
      Thank you in advançe for your valuable time.

  5. Brage • 04/08/2016 #

    Thanks for sharing. I’m new to autotrading.
    I thought this was a 30min breakout strategy with one position per day.
    Testing this on PRT/IG works ok with index (swedish omxs30) but when testing on stocks I get several positions during 9.30 and 18.00. Mostly one bar positions. What do I do wrong? Do I need to change something when testing on stocks?
    Best regards

    • Nicolas • 04/08/2016 #

      Hi Martin, are you testing on a daily timeframe?

  6. janverhoeff • 04/08/2016 #

    Hi, if i want to do this whit 2 contracts then i only have to chanche N=1 to N=2??
    greetings Jan

    • Nicolas • 04/08/2016 #

      Hi Jan, yes “n” is the contracts size, you are right.

  7. Jan • 04/08/2016 #

    Thanks for the quick answer.
    Is it allso an idea to add a trailing stop extra to the system. start at the take profit zone?

    • Nicolas • 04/08/2016 #

      It is Doctrading’s strategy, maybe he could answer better than me to this question 🙂 Sometimes trailing stop just kill profit more than let it run, it needs some test here!

  8. Jan • 04/08/2016 #

    This morning the system startet and worked good whit 1 contract. halfe an houre ago i deleted the system and added a new one whit n=2.
    After a few minutes it started too sell 2 contracts.  it wasn’t in the ranges of 09:00 houres 09:30.  ho is this possible?

  9. Doctrading • 04/08/2016 #

    Of course this strategy needs improvement.
    It’s only good after 2014 (don’t forget to backtest it with spread ! the screenshot doesn’t show spread).
    Trailing stop would also be a nice idea. But first I test on several years with take profit, in order to see if it’s viable.
    Personnage, I wouldn’t use this one in real trading, as it is profitable only since 2014.

  10. Hockeytrader • 04/08/2016 #

    Hi! Thanks for this code. Im willing to make some adjustments but before that i need the code to be translated to english. Somebody who can help?

  11. Doctrading • 04/08/2016 #

    Hello Hockeytrader,
    I don’t understand you, the code is already in English, isn’t it ?

    • Hockeytrader • 04/08/2016 #

      Hi! Thanks for answer. Haut, bas, achat & vente are the words i’m thinking about.

  12. Doctrading • 04/08/2016 #

    “Haut” = “high”
    “Bas” = “low”
    “Achat & Vente” = “Buy & Sell”
    Google Translate is your friend 😉

  13. raj • 04/08/2016 #

    Thanks for sharing. I have tried to use this code for a strategy I want to test on forex. The idea is to create an overnight range from the previous 8 hours and trade the breakout of this range from 9am onwards. Below is the code just for the long side of the stagey with 15min bars, hence (32) as this is 8h. I want it to go flat at 13:00, have added a trailing stop and a pip target. This are not optimised in any way, just first ideas to test.
    Anyhow, it does not seem to work. For some reason it does not take trades on certain days and sometimes take multiple trades on after another. I think one of the problems is the range is constantly being updated from the previous 32 bars, whereas I want it set rigid after 9am (highest/lowest 1am-9am). I thought the part ‘IF time = 090000’ would set this but not 100% sure, I am very new to this.
    Thanks in advance!
    Defparam cumulateorders = false
    Defparam flatafter = 130000
    n = 1
    IF Time = 090000 THENtop = highest[32](high)long = 0ENDIF
    if Time >= 090000 AND Time <= 130000 THEN
    IF long = 0 THENbuy n share at top stopENDIF
    If longonmarket THENlong = 1ENDIF
    SET STOP pTRAILING 15set target pprofit 30

  14. Glen Marquis • 04/08/2016 #

    Only exit at other side of the 9.00-9.30 range, Dax moves far more “usually” than that!Perhaps try this with a 50+ target & 20 point stop trail.

  15. Hockeytrader • 04/08/2016 #

    I’ve had this thought that if the first trade goes wrong and we end up with a loose. Is it possible to make the second trade double size?

  16. Doctrading • 04/08/2016 #

    Hello, it is still possible, but I don’t like martingale.

  17. cupcake • 04/08/2016 #

    Hi, I’m trying to get my own system together that is very similar to the OP’s.
    It’s in essence a range breakout, but my range is the opening 5m bar of the day.
    stops and targets are similar theory to the OP’s as well,  Stop Loss: the other side of the range, Take Profit: equal to the stop loss
    I just can’t for the life of me amend this code to work, or get my own code.
    Is there anyone that’s able to help me with this at all?
    Thank you in advance.

  18. Doctrading • 04/08/2016 #

    Do you just consider the first 5min bar at 9AM as the range ?
    Yes, it is quite easy to backtest, but results will not be good. Because it is a too small timeframe, because of slippage, because of spread.

  19. cupcake • 04/08/2016 #

    it’s the first 5m bar on the 8am GMT open.
    Manually I would go long 2 points above the high of that first bar, with a stop 2 points below the low, and visa versa for the short entry.
    I’m just struggling to code it at the moment.

  20. Doctrading • 04/08/2016 #

    I will try later.If it is profitable, I will share the code.
    Do you take 2 trades a day even if the first is winning ? Or only one ?Regards,

  21. cupcake • 04/08/2016 #

    if the first hits the target then that’s the only trade.
    If doesn’t hit target, then opens the reverse trade when stopped.
    No trades at all after 12pm if long or short hasn’t triggered

  22. Hockeytrader • 04/08/2016 #

    Been testing the code for four weeks now. One optimisation i’ve been thinking about is to open a position when a bar closes over/under day h/l. who knows how that code would look?

  23. Doctrading • 04/08/2016 #

    The code should look like this : 
    Defparam cumulateorders = false

    // Timeframe : H1

    IF TIME = 080000 THEN
    HAUT = highest[1](high)
    BAS = lowest[1](low)
    achatjour = 0
    ventejour = 0
    amplitude = haut-bas

    Ctime = time >= 080000 AND time <= 120000

    IF Ctime THEN
    IF achatjour = 0 THEN
    buy at HAUT + 1 stop
    ELSIF ventejour = 0 THEN
    sellshort at BAS - 1 stop

    IF longonmarket THEN
    achatjour = 1
    IF shortonmarket THEN
    ventejour = 1

    set stop loss amplitude
    set target profit amplitude

    But there is some issues : 

    I did’t set yet that if the first order is winning, there will be no second order
    it is winning without spread… but loosing with 1 point spread

    Feel free to improve this code.regards,

  24. Doctrading • 04/08/2016 #

    Just don’t consider the // timeframe   and // 05H to 09H   breakout.It was from an ancient london open breakout code.

  25. Hockeytrader • 04/08/2016 #

    Thanks. Will try it out.

  26. cupcake • 04/08/2016 #

    so I’ve tried to amend the OP’s code to look at a smaller range for my breakout, but I can’t get it to take any trades at all
    Defparam cumulateorders = false
    Defparam flatafter = 180000

    n = 1

    IF Time = 090501 THEN
    phigh = highest[2](high)
    plow = lowest[2](low)
    amplitude = phigh - psell
    pbuy = 0
    psell = 0

    if Time > 090501 AND Time <= 180000 THEN

    IF pbuy = 0 THEN
    buy n share at phigh stop

    IF psell = 0 THEN
    sellshort n share at plow stop


    If longonmarket THEN
    pbuy = 1

    IF shortonmarket THEN
    psell = 1

    set stop ploss amplitude
    set target pprofit amplitude


  27. Hockeytrader • 04/08/2016 #

    Hey. Been comparing the strategys performance versus the DAX volatility index. i’ve not done all 100 000 bars yet but so far the strategy is profitable when the $VDAX is somewhere between 15 & 25. Gonna work on this tonight and see if a can give you some kind of report.

  28. Dave • 04/08/2016 #

    Hi, I’m new to coding and have been trying to modify the code a little to backtest an idea I have but am struggling!.my idea is to take the first 8.00 uk m15 candle and create a sell order 25 points above the high and a buy order 25 from the low incorporating a simple 25 sl and 50tp.Any help would be greatly appreciated.

    • Nicolas • 04/08/2016 #

      Better use the forums for coding assistance please. You’ll get more results there for sure.

  29. Dave • 04/08/2016 #

    Apologies – only just learning the site layout.
    Maybe you could delete the post?

Register or


avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar
Related users ' posts
superfalcio Buonasera, I downloaded the " Sistema HA - ITA 40 - 1 HOUR " but when i run that, it says a ...
Aaron Bennett Thanks for this @lupo32, I've taken a copy of sistema-ha-dax-1-hour.itf this morning and sim...
Aaron Bennett ITF:
Roland57800 Bonjour Noobywan, si je comprends bien l'utilisation de ce code que je trouve particulièr...
Noobywan Bonjour, quand la question a été posée dans le forum à celui qui avait fait la requête en am...
superprof salut merci pour ton indicateur. Pour bien comprendre , si suis acheteur à quelle comment j...
Nicolas A discrétion, bon courage dans ton apprentissage du trading.
Manu L. Bonjour Nicolas, j'ai créé un screener avec l'aide à la prog en TF weekly mais je voudrai...
Newbie999 I want to use this code but limit my scan to stocks priced to a maximum of 10c. Any ideas? t...
Nicolas mycondition = close <= 0.10
B31700 Bonjour, Je suis débutant, c'est mon premier post. Félicitations pour le screener. Je vou...
Oxxoz Re bonjour Vivien, Il me smeble que 3 mois, c'est 60 séances. Cordialement,
Vinks_o_7 thanks ...and It can be used on a monthly and weekly basis by changing just the duration in ...
Randy33 Very very nice! How can i change the price in the code? Like to screen only smal caps (Price...
nicko Many Thanks Viv. It looks good
Nokis Bonjour Vivien, Comment je peux faire pour adapter ton code afin: De cibler les sociétés q...
toendidi Hi Vonasi, thankx for the indicator I find it very usefel. I am trying to study the breakot...
Canis Majoris Hi , Helpful indicator ...
Maxime Csls Bonjour Vonasi, Bravo pour ton travail ! Est-il possible de créer un screener quand le prix ...
FXmike hey my friends, thank you for this great code. my problem is he make no trade open. backtest...
FXmike Can i put a Action that my start contract is smaller than 1 ? 0.3 or 0.5 ? In wich Position...
phoentzs I wrote the code for M15 back then and also variants in H1 and M1. Everything works so far....
Jan Wind Thanks for sharing ! The Gain/Loss ratio is a bit low, 1,16, hardly offset the risks take...
drive whats the period ?
Ciurno981 il suffit de changer le nombre de contrats à acheter
Ciurno981 Does anyone have the data at 200,000 units, why would I like to see over 5 years?
Ludwig Bonjour Merci pour le code, je souhaite intégrer un break even et mette en place un réinves...
Dom Hello, commence le trading et découvre par la même occasion le ce n...
Nicolas Merci, ça fait plaisir !
Be-n Bonjour tout le monde ! Dans l'indicateur de tendance, j'ai du mal à saisir la nuance entre ...
luxrun Vonasi, is it possible to have a screenshot of an index chart (eg djia) and below the breako...
Vonasi I can't add any images here now that the library post has been listed. Perhaps you could sta...
Vinks_o_7 Thanks again Vonasi !
frenqle I have it on 1 minute but still 13 april is the latest its doing
robertogozzi The error you reported has been fixed, now the strategy enter trades regularly in AutoTradin...
frenqle Done!! thanks Roberto, it is running again!
sycomore Ça ne fonctionne sur aucune valeur, je dois être trop con...
Nicolas il faut appliquer l'indicateur sur le prix.
AntoGH Je crains de ne pas comprendre... Cela indique la probabilité de la prochaine bougie ou donn...
samsampop Hi robertogozzi - thank you very much for sharing this strategy. I have performed various ...
robertogozzi Thank you samsampop.
Dotan Hello guys I really appreciate this coding effort but can I use this code for Mt5 Forex Trad...
Gubben So far so good. After a long drawdown it was winning four times in five weeks. Since 24 jul...
pat95162 Salut Pourquoi je n'ai pas les même résultats que vous
UkCoopDownUnder Any hints for us, on which variables we should be optimising ? As of November 2020, the cu...
TempusFugit Interesting idea and with a lot of room for improvement.... maybe a trailing stop? Thanks fo...
Jan Wind Unglaubich das dieser Strategie Gewinne generiert ! Es erschient leider nicht so gleichmaßi...
Peter I recognized a possible problem with the timeframe / Defparam variables. As far as I know, t...