Discussion re Pure Renko strategy

• 03/10/2020 at 3:51 PM #121721
  nonetheless

  Break even also helps:

  //Break even startBreakeven = 50 PointsToKeep = 5 once breakeaven = 1//1 on - 0 off //reset the breakevenLevel when no trade are on market if breakeaven>0 then IF NOT ONMARKET THEN breakevenLevel=0 ENDIF // --- BUY SIDE --- //test if the price have moved favourably of "startBreakeven" points already IF LONGONMARKET AND close-tradeprice(1)>=startBreakeven*pipsize THEN //calculate the breakevenLevel breakevenLevel = tradeprice(1)+PointsToKeep*pipsize ENDIF //place the new stop orders on market at breakevenLevel IF breakevenLevel>0 THEN SELL AT breakevenLevel STOP ENDIF // --- end of BUY SIDE --- IF SHORTONMARKET AND tradeprice(1)-close>startBreakeven*pipsize THEN //calculate the breakevenLevel breakevenLevel = tradeprice(1)-PointsToKeep*pipsize ENDIF //place the new stop orders on market at breakevenLevel IF breakevenLevel>0 THEN EXITSHORT AT breakevenLevel STOP ENDIF 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

  //Break even startBreakeven = 50 PointsToKeep = 5 once breakeaven = 1//1 on - 0 off //reset the breakevenLevel when no trade are on market if breakeaven>0 then IF NOT ONMARKET THEN breakevenLevel=0 ENDIF // --- BUY SIDE --- //test if the price have moved favourably of "startBreakeven" points already IF LONGONMARKET AND close-tradeprice(1)>=startBreakeven*pipsize THEN //calculate the breakevenLevel breakevenLevel = tradeprice(1)+PointsToKeep*pipsize ENDIF //place the new stop orders on market at breakevenLevel IF breakevenLevel>0 THEN SELL AT breakevenLevel STOP ENDIF // --- end of BUY SIDE --- IF SHORTONMARKET AND tradeprice(1)-close>startBreakeven*pipsize THEN //calculate the breakevenLevel breakevenLevel = tradeprice(1)-PointsToKeep*pipsize ENDIF //place the new stop orders on market at breakevenLevel IF breakevenLevel>0 THEN EXITSHORT AT breakevenLevel STOP ENDIF endif

   

  

  nonetheless

  Participant

  Topics: 83

  Replies: 1237

  Been thanked: 978 times
  03/10/2020 at 4:15 PM #121727
  Paul

  Maybe it could add some benefit on the 1 sec too.

   

  

  Paul

  Participant

  Topics: 40

  Replies: 812

  Been thanked: 660 times
  03/10/2020 at 4:32 PM #121729
  Paul

  1s update, lucky day so far

  

  Paul

  Participant

  Topics: 40

  Replies: 812

  Been thanked: 660 times
  03/10/2020 at 4:37 PM #121732
  nonetheless

  @Paul — your 1 second code is amazing, 2 good wins in the 20 minutes since I loaded it! I like how responsive it is, locks in any profit so quickly. Nice one!

  All these Renkos are so hard working and they get out of losing positions so fast – blink and they change direction. I like that!

   

  1 user thanked author for this post.

  Paul

  

  nonetheless

  Participant

  Topics: 83

  Replies: 1237

  Been thanked: 978 times
  03/10/2020 at 4:47 PM #121733
  Paul

  @nonetheless Yeah me too! I’am losing faith a bit of systems working 10-20 years back, how about this! 🙂

  Never tried anything on this timeframe and most likely it got lucky because of the dayrange.

  But you see the difference between the backtest & demo?  backtest long at 1642 and 1645, but not in the demo. I don’t understand it.

   

  

  Paul

  Participant

  Topics: 40

  Replies: 812

  Been thanked: 660 times
  03/10/2020 at 4:54 PM #121734
  GraHal

  nonetheless wrote:

  blink and they change direction. I like that!

  Me too!

  Since lately … blink on the DJi and if you are going in the wrong direction then you can be 100’s in loss!!!

  Paul wrote:

  backtest long at 1642 and 1645, but not in the demo.

  Do you mean Backtest on Demo Account vs Forward Test / Live running on Demo Account?

  

  GraHal

  Participant

  Topics: 212

  Replies: 8789

  Been thanked: 2187 times
  03/10/2020 at 4:59 PM #121735
  nonetheless

  Paul wrote:

  the difference between the backtest & demo?

  Yeah, that is a weird one. Maybe different engines? maybe slippage? or the way the Renko blocks are calculated … who knows.

  

  nonetheless

  Participant

  Topics: 83

  Replies: 1237

  Been thanked: 978 times
  03/10/2020 at 5:05 PM #121736
  Paul

  @GraHal, all in demo account. Loaded strategy on the chart compared to live running in demo. It’s not always in sync. Price-difference (spread/slippage) is oke, but big time-difference or not trading is not right.

  

  Paul

  Participant

  Topics: 40

  Replies: 812

  Been thanked: 660 times
  03/10/2020 at 5:16 PM #121737
  GraHal

  Paul wrote:

  big time-difference or not trading is not right.

  I agree … doesn’t give us confidence.

  We should try this and that to see if we can get Backtest (BT) synchronised nearer to Forward Test (FT).

  We only have spread to adjust on the Platform, so we’d have to have several Systems on BT and FT at same time with 1, 2, 3, 4 … etc points in the code to simulate slippage??

  We may be surprised at the results?

   

   

  

  GraHal

  Participant

  Topics: 212

  Replies: 8789

  Been thanked: 2187 times
  03/10/2020 at 5:27 PM #121738
  Paul

  here’s an update

  Seems the trailing-stop I modified some time works good too (in the backtest)! Now you can set which one you want to use in top of the code.

  I’ve tested the breakeven. But for now it doesn’t seem too add much because of the quick trailingstop & small stoploss.

  Increased the nbbarlimit to 4 (minimum=1) It should make sure, if an order is not executed for any reason, like there was no data on that second, it’s repeated again. At least that’s the intention.

   

  1 user thanked author for this post.

  GraHal

  

  Paul

  Participant

  Topics: 40

  Replies: 812

  Been thanked: 660 times
  03/10/2020 at 5:30 PM #121740
  GraHal

  It could be that 1 second is not always long enough for the signal to travel from the PRT Server location to IG Server location?

  Maybe 2 seconds may give better like for like comparison between BT and FT?

  Do we even know if PRT Server is co-located with IG Server … or is one in France and one in London??

  Are BT Servers in different locations and / or  not as fast as FT Servers??

  Just a few thoughts anyway.

  

  GraHal

  Participant

  Topics: 212

  Replies: 8789

  Been thanked: 2187 times
  03/10/2020 at 5:36 PM #121741
  Paul

  Interesting thoughts! I dunno. But if they offer 1 sec, it should be reliable regardless. I haven’t tried it on a different short timeframe. It’s something that needs testing.

  

  Paul

  Participant

  Topics: 40

  Replies: 812

  Been thanked: 660 times
  03/10/2020 at 5:48 PM #121743
  Paul

  @GraHal it could be very well some kind lag then. Both systems should go onmarket at the same time but only one goes. I don’t have any rejections.

  

  Paul

  Participant

  Topics: 40

  Replies: 812

  Been thanked: 660 times
  03/10/2020 at 5:51 PM #121744
  nonetheless

  I just tried it v quickly – 10 sec, 5 sec, 2 sec all worse. 1 second is what works.

  2 users thanked author for this post.

  GraHal, Paul

  

  nonetheless

  Participant

  Topics: 83

  Replies: 1237

  Been thanked: 978 times
  03/10/2020 at 5:56 PM #121746
  GraHal

  They are good fun Paul, thank you so much!

  I am going to put 2 x v2p Systems on FT … they take so many trades that I am going to use 1 x v2p System for semi-manual trading … as in, I stop the System when I am away from the screen or if I think ‘I know better‘ than the System!  Then I see how results stack up after 1 week.

  Just an experiment … keep my hand in on manual trading! 🙂

  1 user thanked author for this post.

  Paul

  

  GraHal

  Participant

  Topics: 212

  Replies: 8789

  Been thanked: 2187 times
• Author
  Posts