Grid orders with one combined stop loss and limit, can it be done?

• 09/29/2016 at 8:04 PM #13930
  cfta

  Hi traders,

  I have spent the last few days playing around with Mike’s Trend Chaser and made a couple of changes, first I removed the Stochastic since it fairly often doesn’t enter the trigger zone even if the price makes a nice move, instead it stradles the oversold or overbought area and we miss out when it doesn’t retrace properly.

  I replaced it with CCI which is more direct with condition to enter when the indicator crosses the zero line. As an additional filter to avoid small ranging moves I added a single ADX to only enter when the price is trending, the value should be higher than 23 and the current bar should be higher than the previous one.

  The below code is intended to be run on M1 TF with and the Coppock Curve and MACD values set to equal M15, the CCI to equal M1 and the ADX matches M1.

  /// Definition of code parameters defparam preloadbars =10000 DEFPARAM CumulateOrders = true // Cumulating positions deactivated COI= WEIGHTEDAVERAGE[210](ROC[165] +ROC[150]) MAC= MACDline[180,390,135] CCC= CCI [60] AAA= ADX [14] once RRreached = 0 accountbalance = 10000 //account balance in money at strategy start riskpercent = 2 //whole account risk in percent% amount = 1 //lot amount to open each trade rr = 2 //risk reward ratio (set to 0 disable this function) sd = 0.25 //standard deviation of MA floating profit //dynamic step grid minSTEP = 10 //minimal step of the grid maxSTEP = 50 //maximal step of the grid ATRcurrentPeriod = 60 //recent volatility 'instant' period ATRhistoPeriod = 600 //historical volatility period ATR = averagetruerange[ATRcurrentPeriod] histoATR= highest[ATRhistoPeriod](ATR) resultMAX = MAX(minSTEP*pipsize,histoATR - ATR) resultMIN = MIN(resultMAX,maxSTEP*pipsize) gridstep = (resultMIN) // Conditions to enter long positions c1 = (COI > COI[1]) c2 = (COI > 0) c3 = (MAC > MAC[1]) c4 = (MAC > 0) c5 = (CCC CROSSES OVER 0) c6 = (AAA > AAA[1]) c7 = (AAA > 23) // Conditions to enter short positions c11 = (COI < COI[1]) c12 = (COI < 0) c13 = (MAC < MAC[1]) c14 = (MAC < 0) c15 = (CCC CROSSES UNDER 0) c16 = (AAA > AAA[1]) c17 = (AAA > 23) //first trade whatever condition if NOT ONMARKET AND HIGHEST[7](c1)=1 AND HIGHEST[7](c2)=1 AND HIGHEST[7](c3)=1 AND HIGHEST[7](c4)=1 AND HIGHEST[7](c5)=1 AND HIGHEST[7](c6)=1 AND HIGHEST[7](c7)=1 then //close>close[1] BUY amount LOT AT MARKET endif if NOT ONMARKET AND HIGHEST[7](c11)=1 AND HIGHEST[7](c12)=1 AND HIGHEST[7](c13)=1 AND HIGHEST[7](c14)=1 AND HIGHEST[7](c15)=1 AND HIGHEST[7](c16)=1 AND HIGHEST[7](c17)=1 then //close<close[1] SELLSHORT amount LOT AT MARKET endif // case BUY - add orders on the same trend if longonmarket and close-tradeprice(1)>=gridstep*pipsize then BUY amount LOT AT MARKET endif // case SELL - add orders on the same trend if shortonmarket and tradeprice(1)-close>=gridstep*pipsize then SELLSHORT amount LOT AT MARKET endif //money management liveaccountbalance = accountbalance+strategyprofit moneyrisk = (liveaccountbalance*(riskpercent/100)) if onmarket then onepointvaluebasket = pointvalue*countofposition mindistancetoclose =(moneyrisk/onepointvaluebasket)*pipsize endif //floating profit floatingprofit = (((close-positionprice)*pointvalue)*countofposition)/pipsize //actual trade gains MAfloatingprofit = average[40](floatingprofit) BBfloatingprofit = MAfloatingprofit - std[40](MAfloatingprofit)*sd //floating profit risk reward check if rr>0 and floatingprofit>moneyrisk*rr then RRreached=1 endif //GRAPH floatingprofit as "float" //GRAPH RRreached as "rr" //GRAPH floatingprofit as "floating profit" //GRAPH BBfloatingprofit as "BB Floating Profit" GRAPH positionprice-mindistancetoclose //GRAPH moneyrisk //GRAPH onepointvaluebasket //GRAPH gridstep //stoploss trigger when risk reward ratio is not met already if onmarket and RRreached=0 then SELL AT positionprice-mindistancetoclose STOP EXITSHORT AT positionprice-mindistancetoclose STOP endif //stoploss trigger when risk reward ratio has been reached if onmarket and RRreached=1 then if floatingprofit crosses under BBfloatingprofit then SELL AT MARKET EXITSHORT AT MARKET endif endif //resetting the risk reward reached variable if not onmarket then RRreached = 0 endif

  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

  /// Definition of code parameters defparam preloadbars =10000 DEFPARAM CumulateOrders = true // Cumulating positions deactivated COI= WEIGHTEDAVERAGE[210](ROC[165] +ROC[150]) MAC= MACDline[180,390,135] CCC= CCI [60] AAA= ADX [14] once RRreached = 0 accountbalance = 10000 //account balance in money at strategy start riskpercent = 2 //whole account risk in percent% amount = 1 //lot amount to open each trade rr = 2 //risk reward ratio (set to 0 disable this function) sd = 0.25 //standard deviation of MA floating profit     //dynamic step grid minSTEP = 10 //minimal step of the grid maxSTEP = 50 //maximal step of the grid ATRcurrentPeriod = 60 //recent volatility 'instant' period ATRhistoPeriod = 600 //historical volatility period   ATR = averagetruerange[ATRcurrentPeriod] histoATR= highest[ATRhistoPeriod](ATR)   resultMAX = MAX(minSTEP*pipsize,histoATR - ATR) resultMIN = MIN(resultMAX,maxSTEP*pipsize)   gridstep = (resultMIN)   // Conditions to enter long positions c1 = (COI > COI[1]) c2 = (COI > 0) c3 = (MAC > MAC[1]) c4 = (MAC > 0) c5 = (CCC CROSSES OVER 0) c6 = (AAA > AAA[1]) c7 = (AAA > 23)   // Conditions to enter short positions c11 = (COI < COI[1]) c12 = (COI < 0) c13 = (MAC < MAC[1]) c14 = (MAC < 0) c15 = (CCC CROSSES UNDER 0) c16 = (AAA > AAA[1]) c17 = (AAA > 23)   //first trade whatever condition if NOT ONMARKET AND HIGHEST[7](c1)=1 AND HIGHEST[7](c2)=1 AND HIGHEST[7](c3)=1 AND HIGHEST[7](c4)=1 AND HIGHEST[7](c5)=1 AND HIGHEST[7](c6)=1 AND HIGHEST[7](c7)=1 then  //close>close[1] BUY amount LOT AT MARKET endif   if NOT ONMARKET AND HIGHEST[7](c11)=1 AND HIGHEST[7](c12)=1 AND HIGHEST[7](c13)=1 AND HIGHEST[7](c14)=1 AND HIGHEST[7](c15)=1 AND HIGHEST[7](c16)=1 AND HIGHEST[7](c17)=1 then  //close<close[1] SELLSHORT amount LOT AT MARKET endif   // case BUY - add orders on the same trend if longonmarket and close-tradeprice(1)>=gridstep*pipsize then BUY amount LOT AT MARKET endif   // case SELL - add orders on the same trend if shortonmarket and tradeprice(1)-close>=gridstep*pipsize then SELLSHORT amount LOT AT MARKET endif   //money management liveaccountbalance = accountbalance+strategyprofit moneyrisk = (liveaccountbalance*(riskpercent/100)) if onmarket then onepointvaluebasket = pointvalue*countofposition mindistancetoclose =(moneyrisk/onepointvaluebasket)*pipsize endif   //floating profit floatingprofit = (((close-positionprice)*pointvalue)*countofposition)/pipsize   //actual trade gains MAfloatingprofit = average[40](floatingprofit) BBfloatingprofit = MAfloatingprofit - std[40](MAfloatingprofit)*sd   //floating profit risk reward check if rr>0 and floatingprofit>moneyrisk*rr then RRreached=1 endif   //GRAPH floatingprofit as "float" //GRAPH RRreached as "rr" //GRAPH floatingprofit as "floating profit" //GRAPH BBfloatingprofit as "BB Floating Profit" GRAPH positionprice-mindistancetoclose //GRAPH moneyrisk //GRAPH onepointvaluebasket //GRAPH gridstep   //stoploss trigger when risk reward ratio is not met already if onmarket and RRreached=0 then SELL AT positionprice-mindistancetoclose STOP EXITSHORT AT positionprice-mindistancetoclose STOP endif   //stoploss trigger when risk reward ratio has been reached if onmarket and RRreached=1 then if floatingprofit crosses under BBfloatingprofit then SELL AT MARKET EXITSHORT AT MARKET endif endif   //resetting the risk reward reached variable if not onmarket then RRreached = 0 endif

  The backtest for this evening’s move on the DAX looks promising 🙂 However I have not done much testing yet so I’m not sure if the code works as intended…

  Nicolas, what is the best way to add more ADX values in the code which equals higher timeframes? Does it make sense to insert a new line on line 8;

  AA2 = ADX [70]

  and on line 38 and 39;

  c8 = (AA2 > AA2[1])
  c9 = (AA2 > 23)

  In order to have a conidtion for the M5 (14) ADX to be above 23 and rising?

  1 user thanked author for this post.

  Wing

  

  cfta

  Participant

  Topics: 2

  Replies: 110

  Been thanked: 62 times
  09/30/2016 at 12:14 AM #13947
  MikeGC

  Excellent work, cfta!  A most impressive integration and result.  Many thanks for the alterations you have suggested to Trend Chaser.  I am now looking forward to testing them.

  

  MikeGC

  Participant

  Topics: 4

  Replies: 30

  Been thanked: 6 times
  09/30/2016 at 7:22 AM #13953
  Nicolas

  @cfta

  Yes, just like you did for the other indicators periods, you can try to play with multiplication to set a slower ADX on the 1 minute timeframe. I saw customers strategies doing well by calculating those higher TF periods just like this.

  Until we do not have the MTF support, that’s how we have to do.

  So the purpose of your actual development is to create a fully automated strategy with the money management concept we all have developed here?

  About the lines 49 and 53 and their conditions testing (with highest ..), I still don’t have understand well what Mike had in mind when he wrote this 🙂 Maybe you could explain it Mike?

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20828

  Been thanked: 4742 times
  09/30/2016 at 7:51 AM #13960
  MikeGC

  @Nicolas and @cfta,

  Extraordinarily, this coding was an amateurish attempt to the define the highest low Coppock beneath zero as the entry point for a long trade (in company with the other indicators).  I was advised by a number of contributors that it didn’t make any sense so I tried a number of different variations to achieve this, but none back tested as well as this peculiar code.  It is particularly strange that it also works well for short trades.

  I would appreciate it, Nicolas, if you could explain what is going on with this freak of a mistake.

  

  MikeGC

  Participant

  Topics: 4

  Replies: 30

  Been thanked: 6 times
  09/30/2016 at 12:50 PM #13974
  Nicolas

  @Mike

  In these 2 lines you are testing if your conditions (from c1 to c7 for a long order, for example)  were true (set to “1”) in the last 7 periods. Because it is surely the case of any of these 7 conditions in the last 7 bars, that’s the reason why it does not have much sense 🙂

   

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20828

  Been thanked: 4742 times
  09/30/2016 at 6:48 PM #13988
  sylvess

  c1 to c7 evaluate to true (1) or false (0). It is testing for at least a “true” ( =1 ) condition for c1 to c7 in any of the last 7 period to go long. If you increase the number of period, you will end up with more trades, less if you reduce. This together with grid is really extraordinary strategy indeed.

  

  sylvess

  Participant

  Topics: 1

  Replies: 8

  Been thanked: 3 times
  09/30/2016 at 8:58 PM #13993
  Nicolas

  @sylvess

  That’s right but is this a good idea? Since that is not what Mike was thinking about to initiate the first trade of the cycle. Seems like randomness to me. No offense here to Mike 😉

  On an other hand, testing all conditions met in the same candlestick instead of in the last 7 ones would mean poor trades quantity.

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20828

  Been thanked: 4742 times
  10/01/2016 at 7:16 AM #14000
  sylvess

  @nicolas

  Think not entirely random as we are running on M1 tf – 7 conditions to be true within 7 minutes. Running on higher tf produce poor result as expected. Tested live on demo account yesterday on Dax and Hang Seng, 2 entries each with 1 win and 1 loss but net profitable. From the test, can see that Grid order is definitely a winning strategy for high profit low loss. Time permit, will test using Supertrend for entry.

   

  

  sylvess

  Participant

  Topics: 1

  Replies: 8

  Been thanked: 3 times
  10/01/2016 at 10:46 AM #14007
  Nicolas

  @sylvess

  Yes you are right, if we can catch the beginning of a higher timeframe trend while possibly put orders on market on small TF with tight stoploss and then let me them run while adding more orders with the grid, it could possibly be a high profit strategy.

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20828

  Been thanked: 4742 times
  10/02/2016 at 5:34 PM #14066
  cfta

  Great to see the thread coming to life again, I appreciate your feedback 🙂

  @ Nicolas, no my primary goal is not to have a fully automatic strategy, rather I still watch the H4 and H1 charts waiting for a good setup and start the system to catch a good entry rather than monitoring the market to pick the starting point manually. Of course it can be used fully automated but I think it is too risky leaving it unsupervised for than a few hours given news and global events which cause irratic price movement. I also intend to split the new code in two version, one for long and another for short since sometimes I don’t want an entry in either direction.

  Regarding line 49 and 53 I didn’t really have any idea of their function, I thought it was a necessary piece of code to check if all conditions were met at the close of the most recent candle, I still have a lot to learn about coding. Perhaps we should remove those lines?

  Friday’s testing turned out good with a 5 % gain on DAX, about 1.5 going down in the morning and then 3.5 going up after lunch but stopped out before the late afternoon/evening move since I havn’t thought about increasing the period for the BB exit which would make sense on M1.

  

  cfta

  Participant

  Topics: 2

  Replies: 110

  Been thanked: 62 times
  10/02/2016 at 6:43 PM #14072
  Nicolas

  About the Friday move of the DAX, and the result made, unfortunately, no one can ever know in advance it would occur. Of course, anticipating economic news can be an advantage with  the strategy. The auto grid step is also something great for the strategy, particularly in that kind of highly directional market move.

  Recently, a customer came to me with an idea: putting pending orders on market, to anticipate breakout of a congestion zone… well, something very often asked, sounds like an ordinary job for me.. But, the main idea behind this simple strategy were to continuously put pending orders on the same price if the stoploss occurred (10 pips in this case). My customer just tell me, that it is preferably to loss something like .. 5 trades of 10 points and to let the market taking its trend, instead of just say .. “well ok, I’m done, lost too much”. The concept is always almost the same: cut the loss quickly and let the profit run, that’s what this guy do and it seems that he is doing well with his trading.

  The truth is ..  that there is no truth. Everyone can take advantage of the market, in the way it is the more suitable to him.

  So, maybe it can be fully automated, in a way suitable for someone in particular, but even not for everyone.

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20828

  Been thanked: 4742 times
  10/05/2016 at 7:22 PM #14342
  cfta

  Well about las Friday’s move of the DAX it could have been anticipated by those paying close enough attention to the market since it was a large drop on Thursday (when there is a single major move for a whole day, it often retraces to some extent the following day but that’s not the point) due to the Deutsche Bank fines which dragged down the entire index big time but on Friday the US department of justice gave in a bit to the pressure from the EU saying they were considering reducing the fine by about 70 % since the fine may have been exaggerated.

  Anyway the system is great for news trading, my favourite way is to catch a move and build some profit a day or 4-8 hours before the release in the direction I think the move is most likely to occur and reach BE before the release and either not lose anything or have the potential for a huge gain, besides that I usually wait a while after news for the market to settle.

  @ Nicolas &/ Mike, I’m testing and making progress but please remind me how I need to adjust the code for the system not to start a new cycle efter being stopped out, thanks 🙂

  

  cfta

  Participant

  Topics: 2

  Replies: 110

  Been thanked: 62 times
  10/06/2016 at 12:17 AM #14354
  MikeGC

  @Nicolas, thanks for your explanation of how this code works.  I have found in testing that it produces more quality trades, and am continuing to use it.

  However, ideally, I would like to be able to only select trades where the Coppock is making higher lows below zero for a long, and lower highs for a short above zero.  I have not been able to code this and would appreciate your help, please.

  

  MikeGC

  Participant

  Topics: 4

  Replies: 30

  Been thanked: 6 times
  10/06/2016 at 6:28 AM #14358
  Nicolas

  @cfta

  You must retrieve the condition I made at the beginning of the project about the strategyprofit test (the first trade don’t go if the strategyprofit is not equal to 0).

  
  @Mike

  So if you want to test a piercing of the highest high or lowest low of the Coppock curve, we must determine first the lookback period for these values to be stored.

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20828

  Been thanked: 4742 times
  10/06/2016 at 6:37 AM #14359
  MikeGC

  @Nicolas, I think we could make the look-back period a variable and optimise it for each instrument and time frame.

  The trick is to find the lower high (and its converse).  This is the thing I can’t code.  Thus if the Coppock makes a high (above zero), then how do I determine whether the next high is lower than the last one?  That is the point (with recent coincidence of the MACD and the stochastic), that I would like ProOrder to take a short.

  

  MikeGC

  Participant

  Topics: 4

  Replies: 30

  Been thanked: 6 times
• Author
  Posts