SHORT EUR USD (m15) // PAC MAN STRATEGY

• 05/01/2018 at 4:09 PM #69577
  Balmora74

  This is a strategy who takes only short positions on EUR/USD

  TimeFrame M15 Test made with spread = 1 pip and n = 7 mini lot of $1

  Test, Improve and Enjoy it !!

  Happy to have your returns.

  Balmora

  PAC MAN SHORT STRATEGY / EUR USD (M15)

  // PAC MAN STRATEGY - PIP HUNTER // SHORT EUR USD (M15) // SPREAD = 1 PIP DEFPARAM CumulateOrders = false DEFPARAM Preloadbars = 4000 N = 7 //MINI LOT $1 Cv1 = (close < Average[100]) and (Average[50] < Average[50](close[1])) Cv2 = RSI[14](close) <= 28 Cv3 = STD[10](close) >= 10.93 * pipsize Cv4 = close <= BollingerDown[13](close[2]) OKSHORT = cv1 and cv2 and Cv3 and Cv4 IF OKSHORT then Sellshort n CONTRACT at market SET STOP pLOSS 52 SET TARGET PPROFIT 25 ENDIF

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

  // PAC MAN STRATEGY - PIP HUNTER // SHORT EUR USD (M15) // SPREAD = 1 PIP   DEFPARAM CumulateOrders = false DEFPARAM Preloadbars = 4000   N = 7   //MINI LOT $1   Cv1 = (close < Average[100]) and (Average[50] < Average[50](close[1])) Cv2 = RSI[14](close) <= 28 Cv3 = STD[10](close) >= 10.93 * pipsize Cv4 = close <= BollingerDown[13](close[2]) OKSHORT = cv1 and cv2 and Cv3 and Cv4   IF OKSHORT then Sellshort n CONTRACT at market SET STOP pLOSS 52 SET TARGET PPROFIT 25 ENDIF

   

  1 user thanked author for this post.

  Berra

  

  Balmora74

  Participant

  Topics: 5

  Replies: 97

  Been thanked: 77 times
  05/02/2018 at 1:28 PM #69645
  JR1976

  HI Balbora ,

  congrats for  the  code , seems simple and good …

  IF you  have time , try  to set up:     TP 11 and SL 52   in  200 kb test  ..

  Thanks

  

  JR1976

  Participant

  Topics: 16

  Replies: 88

  Been thanked: 8 times
  05/03/2018 at 6:16 AM #69678
  Balmora74

  Hi JR 1976,

  Results are worst with SL 52 / TP 11 / 200 kb backtest.

  Balmora

   

  

  Balmora74

  Participant

  Topics: 5

  Replies: 97

  Been thanked: 77 times
  05/03/2018 at 8:55 AM #69688
  Inertia

  Tweacked a bit…and reduced to 1 lot.

  This is what I got.

  Anyone can beat it ? Let’s GO !

  

  Inertia

  Participant

  Topics: 40

  Replies: 223

  Been thanked: 52 times
  05/03/2018 at 9:02 AM #69693
  Balmora74

  Thanks Inertia. Excellent !

  Do you use a trailing stop ?

  Can you share the code and modifications please ?

   

  

  Balmora74

  Participant

  Topics: 5

  Replies: 97

  Been thanked: 77 times
  05/03/2018 at 9:43 AM #69698
  Inertia

  There you go.

  Anyone would have a “not too bad” EUR/USD Long strategy to play with ??

  Pac Man

  // PAC MAN STRATEGY - PIP HUNTER // SHORT EUR USD (M15) // SPREAD = 1 PIP DEFPARAM CumulateOrders = False DEFPARAM Preloadbars = 6000 N = 1 //MINI LOT $1 Cv1 = (close < Average[100]) and (Average[50] < Average[50](close[1])) Cv2 = RSI[14](close) < 28 Cv3 = STD[10](close) >= 10.93 * pipsize Cv4 = close <= BollingerDown[13](close[2]) indicator1 = ExponentialAverage[5000](close) c1 = (close > indicator1) OKSHORT = cv1 and cv2 and cv3 and Cv4 and c1 IF OKSHORT then Sellshort n CONTRACT at market endif startBreakeven = 9 PointsToKeep = 1 IF SHORTONMARKET AND (tradeprice(1) - close) >= (startBreakeven * pipsize) THEN breakevenLevel = tradeprice(1) + (PointsToKeep * pipsize) IF breakevenLevel > 0 THEN EXITSHORT AT breakevenLevel STOP ENDIF ENDIF SET STOP pLOSS 48 SET TARGET ppROFIT 12

  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

  // PAC MAN STRATEGY - PIP HUNTER // SHORT EUR USD (M15) // SPREAD = 1 PIP DEFPARAM CumulateOrders = False DEFPARAM Preloadbars = 6000 N = 1   //MINI LOT $1 Cv1 = (close < Average[100]) and (Average[50] < Average[50](close[1])) Cv2 = RSI[14](close) < 28 Cv3 = STD[10](close) >= 10.93 * pipsize Cv4 = close <= BollingerDown[13](close[2])   indicator1 = ExponentialAverage[5000](close) c1 = (close > indicator1)   OKSHORT = cv1 and cv2 and cv3 and Cv4 and c1 IF OKSHORT then Sellshort n CONTRACT at market endif   startBreakeven = 9 PointsToKeep   = 1   IF SHORTONMARKET AND (tradeprice(1) - close) >= (startBreakeven * pipsize) THEN breakevenLevel = tradeprice(1) + (PointsToKeep * pipsize)   IF breakevenLevel > 0 THEN EXITSHORT AT breakevenLevel STOP ENDIF ENDIF   SET STOP pLOSS 48 SET TARGET ppROFIT 12

   

   

  

  Inertia

  Participant

  Topics: 40

  Replies: 223

  Been thanked: 52 times
  05/03/2018 at 9:59 AM #69700
  Balmora74

  Thanks a lot.

  I see a vert strange thing on the equity curve.

  On the period from May 2014 to May 2015, the algorithm don’t take any position while we have a big and long short movement on Eur / Usd.

  Strange isn’t it ?

  

  Balmora74

  Participant

  Topics: 5

  Replies: 97

  Been thanked: 77 times
  05/03/2018 at 10:16 AM #69704
  Inertia

  Well, that is the EMA that act like a filter…Yeah, nothing is always perfect 🙁

  

  Inertia

  Participant

  Topics: 40

  Replies: 223

  Been thanked: 52 times
  05/03/2018 at 10:19 AM #69705
  jebus89

  indicator1 = ExponentialAverage[5000](close) c1 = (close > indicator1)

  1 2	indicator1 = ExponentialAverage[5000](close) c1 = (close > indicator1)

  This is what keeps it from taking trades during the big downwards movement in 2014-15

  

  jebus89

  Participant

  Topics: 118

  Replies: 586

  Been thanked: 238 times
  05/03/2018 at 10:23 AM #69706
  Inertia

  With and without filter.

  Which one you would prefer working with ?

  

  Inertia

  Participant

  Topics: 40

  Replies: 223

  Been thanked: 52 times
  05/03/2018 at 10:58 AM #69711
  bezieh

  I guess you optimized this code on 100% of your data, dont u think its curve fitted? How does a walk forward result look?

  

  bezieh

  Participant

  Topics: 11

  Replies: 25

  Been thanked: 1 time
  05/03/2018 at 11:07 AM #69712
  GraHal

  I’ve added the latest version to the Link Library, row 24 here Snippet Link Library

  More than a Snippet I know, but as I can follow all the code (without having to try and visualise what the eff is going on! 🙂 ) I felt it would be useful to others.

  

  GraHal

  Participant

  Topics: 204

  Replies: 8397

  Been thanked: 2039 times
  05/03/2018 at 11:17 AM #69714
  auvergnat

  Hello Balmora

  Can you explain your number choice of 10,93 for :

  Cv3 = STD[10](close) >= 10.93 * pipsize

  Thks.

  Tof

  

  auvergnat

  Participant

  Topics: 6

  Replies: 32

  Been thanked: 0 times
  05/03/2018 at 11:20 AM #69715
  Balmora74

  I choose 10.93 after a optimisation backtest with Walk Forward.

  

  Balmora74

  Participant

  Topics: 5

  Replies: 97

  Been thanked: 77 times
  05/03/2018 at 2:12 PM #69725
  jebus89

  Yea im also a bit skeptical, it seems like its been optimized on 100% of the data and is curvefitted. Would definitly run in demo in a few months before running it live.

  Also i removed a rule here and there to see what would happen and the results are strange.. also there is no exit. Thats a red flag for me. I feel like using Profit target + stop loss as the only exit is not good enough. Maybe thats just me tho 🙂 If u got a better exit with it, then ur probably going to see bigger profits. Im also negative to 90%~ winrate strategies because they usually (this included) require a stop loss 3-5x higher then the target. Meaning if winrate would start to drop your gonna start bleeding money very fast.

  I would still consider myself a rookie here tho, and im mainly a trend follower, meaning im biased as hell 🙂

  

  jebus89

  Participant

  Topics: 118

  Replies: 586

  Been thanked: 238 times
• Author
  Posts