Fractal breakout intraday Strategy EUR/USD 1H –

• 10/07/2017 at 3:57 PM #48559
  Vonasi

  I guess there must be a difference but I’ve only ever developed and tested on DFB so not able to be much help. Hopefully someone more experienced will chip in with an answer to the positionsize problem.

  

  Vonasi

  Moderator

  Topics: 192

  Replies: 5088

  Been thanked: 1811 times
  10/07/2017 at 4:10 PM #48560
  Ronny

  I changed the maxpositionsize to 5 (5000 was a stresstest, I could read earlier in this thread).

  I changed

  set target pprofit TakeProfit

  1	set target pprofit TakeProfit

  to

  set target profit TakeProfit

  1	set target profit TakeProfit

  This is the tick-by-tick results.

   

  

  Ronny

  Participant

  Topics: 6

  Replies: 22

  Been thanked: 4 times
  10/07/2017 at 5:32 PM #48568
  ozz87

  Ronny: Cool. Will test tomorrow.

  

  ozz87

  Participant

  Topics: 0

  Replies: 168

  Been thanked: 127 times
  10/07/2017 at 5:35 PM #48569
  Vonasi

  Thanks for that Ronny. It seems that the ATR based Stoploss and TakeProfit although profitable may not necessarily be an improvement on fixed SL and TP levels. Early on there is a big drawdown which is what I always fear with short period ATR based decisions. A big candle can leave you vulnerable with a big stoploss calculated from it. Pity really as markets are changing animals and ATR always seems to be a good way to stay in tune with the market but unfortunately ATR can be too spiky or too laggy depending how fast or slow you set it – just like any MA.

  Out of interest how did you decide on the fixed levels for SL and TP that you chose – was it an optimization backtest?

  

  Vonasi

  Moderator

  Topics: 192

  Replies: 5088

  Been thanked: 1811 times
  10/08/2017 at 7:35 AM #48585
  rejo007

  @ronny

  in resume, you have this code?

  because on backtest, they are a problem

   

  //EURUSD(DFB) – IG MARKET
  // TIME FRAME 1H
  // PROBACKTEST TICK by TICK
  // SPREAD 0.9 PIP
  // ALE – KASPER – VONASI

  DEFPARAM CumulateOrders = false

  CP = 101 //Fractal Period
  RSINum = 2
  Ave = 7 //AverageTrueRange Period
  TGLMult = 0.6 //ATR multipier for TGL
  STPMult = 1.8 //ATR multiplier for StopLoss
  AddOn = 1.1 //Added to STPMult for TakeProfit
  RSIHighLevel = 80

  RSILowLevel = 100 – RSIHighLevel

  //KASPER CODE OF REINVESTMENT
  Reinvest=1
  if reinvest then
  Capital = 5000
  Risk = 1//0.1//in % pr position
  StopLoss = AverageTrueRange[Ave] * STPMult
  TakeProfit = AverageTrueRange[Ave] * STPMult + AddOn

  REM Calculate contracts
  equity = Capital + StrategyProfit
  maxrisk = (equity*(Risk/100))
  MAXpositionsize=5
  MINpositionsize=1
  Positionsize= MAX(MINpositionsize,MIN(MAXpositionsize,abs(((maxrisk/StopLoss)))))
  else
  Positionsize=1
  StopLoss = AverageTrueRange[Ave] * STPMult
  TakeProfit = AverageTrueRange[Ave] * STPMult + AddOn
  Endif

  ///BILL WILLIAM FRACTAL INDICATOR
  if Close[cp] >= highest[2*cp+1](Close) then
  LH = 1
  else
  LH = 0
  endif
  if Close[cp] <= lowest[2*cp+1](Close) then
  LL = -1
  else
  LL = 0
  endif
  if LH = 1 then
  HIL = Close[cp]
  endif
  if LL = -1 then
  LOL = Close[cp]
  endif

  //CumulativeRSI2
  RSI2 = (SUMMATION[RSINum](RSI[RSINum](Close)))/RSINum
  RSILow = RSI2 < RSILowLevel
  RSIHigh = RSI2 > RSIHighLevel

  //LONG and SHORT CONDITIONS
  if (time >=100000 and time < 230000) then
  C1 = (close CROSSES OVER HIL)
  D1 = (close CROSSES UNDER LOL)

  IF c1 and NOT ShortOnMarket and RSIHigh THEN
  PositionMultiple = (RSI2/100) + 1//Increase PositionSize depending on CumRSI2 level
  PositionSize = (PositionSize/((RSIHighLevel/100)+1)) * PositionMultiple
  PositionSize = Round(PositionSize * 100)
  PositionSize = PositionSize / 100
  BUY positionsize CONTRACT AT MARKET
  ENDIF

  IF D1 and NOT LongOnMarket and RSILow THEN
  PositionMultiple = ((100-RSI2)/100) + 1//Increase PositionSize depending on CumRSI2 level
  PositionSize = (PositionSize/((RSIHighLevel/100)+1)) * PositionMultiple
  PositionSize = Round(PositionSize * 100)
  PositionSize = PositionSize / 100
  SELLSHORT positionsize CONTRACT AT MARKET
  ENDIF
  ENDIF

  //TRAILING STOP
  TGL = AverageTrueRange[Ave] * TGLMult
  TGS = TGL
  if not onmarket then
  MAXPRICE = 0
  MINPRICE = close
  PriceExit = 0
  ENDIF
  if longonmarket then
  MAXPRICE = MAX(MAXPRICE,close)
  if MAXPRICE-tradeprice(1)>=TGL*pointsize then
  PriceExit = MAXPRICE-TGL*pointsize
  ENDIF
  ENDIF
  if shortonmarket then
  MINPRICE = MIN(MINPRICE,close)
  if tradeprice(1)-MINPRICE>=TGS*pointsize then
  PriceExit = MINPRICE+TGS*pointsize
  ENDIF
  ENDIF
  if onmarket and PriceExit>0 then
  EXITSHORT AT PriceExit STOP
  SELL AT PriceExit STOP
  ENDIF

  set target profit TakeProfit
  set stop ploss stoploss

  

  rejo007

  Participant

  Topics: 3

  Replies: 97

  Been thanked: 10 times
  10/08/2017 at 9:50 AM #48607
  Ronny

  @vonasi

  I kind of agree with the ATR-way of setting TP and SL. But I feel I lose control with the risk managment, and ususally prefer to use the fixed ones. The values I used in the backtest is from the top of my head at the time. I have now run a WF-test which returned a takeprofit of 50 and stoploss at 75… Every trader has his own risk profile 🙂 The results from the testing of this strategy seems to return too good of values, it’s easy to think overfit. But as I said, the results of live trading a similar strategy returns good results (but not in significant numbers yet).

  
  @rejo007

  Yes, there’s something wrong with the reinvest code. I don’t know what.

   

  

  Ronny

  Participant

  Topics: 6

  Replies: 22

  Been thanked: 4 times
  10/08/2017 at 4:00 PM #48630
  Vonasi

  Thanks for doing that test Ronny. Interesting that your off the top of your head numbers had higher TP than SL but the WF test gave higher SL  than TP. What was the win rate and returns like compared to your numbers? I prefer at least 1:1 Risk to Reward as even by tossing a coin I can be right 50% of the time – although give me a strategy with an almost certain 100% win rate and I’ll obviously take a worse R/R ratio!

  Is the strategy that you say you are testing live ALE’s original strategy or another one that you have – if another one would you be able to share it for comparison purposes?

  

  Vonasi

  Moderator

  Topics: 192

  Replies: 5088

  Been thanked: 1811 times
  10/08/2017 at 4:15 PM #48631
  Ronny

  The higher the stoploss is the higher win percentage you will have, as you say. And often much better results. The problem is the DD that will occur, that can wipe us out. I agree with you that RR of 1:1 is minimal. For that reason I’m a bit hesitant to let the optimalization “decide” this values. I’ll try with some values I’ll like to live with, and run OOS without any optimalization. For the TP and SL.

  Here’s the code running live since May:

  //------------------------------------------------------------------------- // Main code : PRC - FBS_EURUSD_60M_V2 //------------------------------------------------------------------------- //EURUSD(-) - IG MARKET // TIME FRAME 1H // PROBACKTEST TICK by TICK - 200.000 bars // SPREAD 0.6 PIP // ALE DEFPARAM CumulateOrders = false DC=20// period of Donchian Stop TGL=5//Trailing Stop pip distance TGS=5 TP=30 //SL IF LONGONMARKET THEN DC=20// period of Donchian Stop TGL=5//Trailing Stop pip distance TP=30//take profit //SL=50// stop loss ELSIF SHORTONMARKET THEN DC=20// period of Donchian Stop TGS=5 TP=30 //SL=50 ENDIF //KASPER CODE OF REINVESTMENT Reinvest=0 if reinvest then Capital = 10000 Risk = 1//0.1//in % pr position StopLoss = 40 REM Calculate contracts equity = Capital + StrategyProfit maxrisk = round(equity*(Risk/100)) MAXpositionsize=5000 MINpositionsize=1 Positionsize= MAX(MINpositionsize,MIN(MAXpositionsize,abs(round((maxrisk/StopLoss)))))//*Pointsize)))) else Positionsize=1 StopLoss = 40 Endif ///BILL WILLIAM FRACTAL INDICATOR MODIFIED CPL=80//period of superior fractal level CPS=100//period of inferior fractal level if close[CPL] >= highest[2*CPL+1](close) then LH = 1 else LH=0 endif if LH=1 then HIL = close[CPL] endif if close[CPS] <= lowest[2*CPS+1](close) then LL= -1 else LL=0 endif if LL = -1 then LOL=close[CPS] endif //LONG and SHORT CONDITIONS RL=0 // range long candles RS=0 // range short candles if (time >=100000 and time < 230000) then A1 = (close CROSSES OVER HIL) A2 = (close CROSSES UNDER LOL) FL= (CLOSE-OPEN)>=RL*POINTSIZE FS= (CLOSE-OPEN)<=-RS*POINTSIZE IF A1 and not shortonmarket AND FL THEN BUY positionsize CONTRACT AT MARKET ENDIF IF A2 and not longonmarket AND FS THEN SELLSHORT positionsize CONTRACT AT MARKET ENDIF ENDIF //TRAILING STOP if not onmarket then MAXPRICE = 0 MINPRICE = close PREZZOUSCITA = 0 ENDIF if longonmarket then MAXPRICE = MAX(MAXPRICE,close) if MAXPRICE-tradeprice(1)>=TGL*pointsize then PREZZOUSCITA = MAXPRICE-TGL*pointsize ENDIF ENDIF if shortonmarket then MINPRICE = MIN(MINPRICE,close) if tradeprice(1)-MINPRICE>=TGS*pointsize then PREZZOUSCITA = MINPRICE+TGS*pointsize ENDIF ENDIF if onmarket and PREZZOUSCITA>0 then EXITSHORT AT PREZZOUSCITA STOP SELL AT PREZZOUSCITA STOP ENDIF // DONCHIAN STOP f=Lowest[DC](low) e= Highest[DC](high) if longonmarket then laststop = f[1] endif if shortonmarket then laststop = e[1] endif if onmarket then sell at laststop stop exitshort at laststop stop endif SET STOP PLOSS StopLoss SET TARGET PPROFIT TP

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115

  //------------------------------------------------------------------------- // Main code : PRC - FBS_EURUSD_60M_V2 //------------------------------------------------------------------------- //EURUSD(-) - IG MARKET // TIME FRAME 1H // PROBACKTEST TICK by TICK - 200.000 bars // SPREAD 0.6 PIP // ALE   DEFPARAM CumulateOrders = false DC=20// period of Donchian Stop TGL=5//Trailing Stop pip distance TGS=5 TP=30 //SL IF LONGONMARKET THEN DC=20// period of Donchian Stop TGL=5//Trailing Stop pip distance TP=30//take profit //SL=50// stop loss ELSIF SHORTONMARKET THEN DC=20// period of Donchian Stop TGS=5 TP=30 //SL=50 ENDIF //KASPER CODE OF REINVESTMENT Reinvest=0 if reinvest then Capital = 10000 Risk = 1//0.1//in % pr position StopLoss = 40 REM Calculate contracts equity = Capital + StrategyProfit maxrisk = round(equity*(Risk/100)) MAXpositionsize=5000 MINpositionsize=1 Positionsize= MAX(MINpositionsize,MIN(MAXpositionsize,abs(round((maxrisk/StopLoss)))))//*Pointsize)))) else Positionsize=1 StopLoss = 40 Endif   ///BILL WILLIAM FRACTAL INDICATOR MODIFIED CPL=80//period of superior fractal level CPS=100//period of inferior fractal level if close[CPL] >= highest[2*CPL+1](close) then LH = 1 else LH=0 endif if LH=1 then HIL = close[CPL] endif if close[CPS] <= lowest[2*CPS+1](close)  then LL= -1 else LL=0 endif if LL  = -1 then LOL=close[CPS] endif //LONG and SHORT CONDITIONS RL=0 // range long candles RS=0 // range short candles   if (time >=100000 and time < 230000) then A1 = (close CROSSES OVER HIL) A2 = (close CROSSES UNDER LOL) FL= (CLOSE-OPEN)>=RL*POINTSIZE FS= (CLOSE-OPEN)<=-RS*POINTSIZE IF A1 and not shortonmarket AND FL THEN BUY positionsize CONTRACT AT MARKET ENDIF IF A2  and not longonmarket AND FS  THEN SELLSHORT positionsize CONTRACT AT MARKET ENDIF ENDIF //TRAILING STOP if not onmarket then MAXPRICE = 0 MINPRICE = close PREZZOUSCITA = 0 ENDIF if longonmarket then MAXPRICE = MAX(MAXPRICE,close) if MAXPRICE-tradeprice(1)>=TGL*pointsize then PREZZOUSCITA = MAXPRICE-TGL*pointsize ENDIF ENDIF if shortonmarket then MINPRICE = MIN(MINPRICE,close) if tradeprice(1)-MINPRICE>=TGS*pointsize then PREZZOUSCITA = MINPRICE+TGS*pointsize ENDIF ENDIF if onmarket and PREZZOUSCITA>0 then EXITSHORT AT PREZZOUSCITA STOP SELL AT PREZZOUSCITA STOP ENDIF // DONCHIAN STOP f=Lowest[DC](low) e= Highest[DC](high) if longonmarket  then laststop = f[1] endif if shortonmarket  then laststop = e[1] endif if onmarket then sell at laststop stop exitshort at laststop stop endif SET STOP PLOSS StopLoss SET TARGET PPROFIT TP

   

  

  Ronny

  Participant

  Topics: 6

  Replies: 22

  Been thanked: 4 times
  10/08/2017 at 5:33 PM #48642
  Barney

  Ronny

   

  Then you have run algon live for 4 months now.

  Has it gone as you expected?

  In that case you can increase the position size as it looks stable 🙂

  

  Barney

  Participant

  Topics: 9

  Replies: 47

  Been thanked: 7 times
  10/08/2017 at 7:59 PM #48662
  Ronny

  The results are in post 48533.

  

  Ronny

  Participant

  Topics: 6

  Replies: 22

  Been thanked: 4 times
  10/08/2017 at 8:51 PM #48665
  Vonasi

  Thanks for sharing that Ronny. It looks very similar to what ALE first posted on this thread except with different fractal look back periods for upper and lower fractals and also with a candle body size condition that is not used. My quick backtest of it shows that it performed better in the past than recently which makes me think it may be a little over fitted. Different periods for upper and lower fractals rings alarm bells for fitting. The fractal period is my major concern with this strategy as it is the most likely thing to be fitted to past market movements. Having said that a wide range of cp numbers result in profit of some sort or other so it does still seem to be a method worth progressing with.

  Can I ask where you got that strategy from? Is there an old thread that I missed that I could read?

  

  Vonasi

  Moderator

  Topics: 192

  Replies: 5088

  Been thanked: 1811 times
  10/08/2017 at 9:50 PM #48669
  ozz87

  Tried to optimize usd/chf. Also tried WF with…. average results. Should mention I’m new to using that. Anyway, uploading results if anyone wants to use it.

  

  ozz87

  Participant

  Topics: 0

  Replies: 168

  Been thanked: 127 times
  10/08/2017 at 11:24 PM #48677
  Despair

  Very good. Now you start using the WFA tool correctly. This makes it of course much more difficult to create a good strategy but it says so much more as a standard backtest.

  2 users thanked author for this post.

  Nicolas, ozz87

  

  Despair

  Blocked

  Topics: 97

  Replies: 1034

  Been thanked: 218 times
  10/13/2017 at 10:21 AM #49247
  Frytconcord

  Bonjour à tous, Quelqu’un at-il eu la dernière version du code FBS qui fonctionne avec IG PRT.

  there are many different version. which is the most optimized? Thank you for your answers

  

  Frytconcord

  Participant

  Topics: 1

  Replies: 10

  Been thanked: 0 times
  10/14/2017 at 10:23 AM #49348
  Frytconcord

  Here is a GBPUSD 15M LONG
  version of the strategy (100,000 bars, backtest tick by tick, 2 point spread)

  With the reinvestment version with the Kasper code, in my example the risk per position is 2%

  Can you tell me what you think?
  If somebody would have time to do it with the version 200.000 bars this would be well. thank you

  

  Frytconcord

  Participant

  Topics: 1

  Replies: 10

  Been thanked: 0 times
• Author
  Posts