difference between broker IG and backtest prorealtime

• 06/30/2016 at 8:47 PM #10056
  Pjotterd

  Hi all,

  I have a question. I have started a trading strategy and this is the code:

  // Conditions to enter long positions IF close>open then BUY 1 CONTRACT AT MARKET ENDIF // Conditions to enter short positions If close<open then SELLSHORT 1 CONTRACT AT MARKET ENDIF // Stops and targets SET STOP pTRAILING 5

  1 2 3 4 5 6 7 8 9 10 11 12

  // Conditions to enter long positions IF close>open then BUY 1 CONTRACT AT MARKET ENDIF   // Conditions to enter short positions If close<open then SELLSHORT 1 CONTRACT AT MARKET ENDIF   // Stops and targets SET STOP pTRAILING 5

   

  I have started the strategy with broker IG. Now, the strange thing is that there is a big difference between the results of broker IG and backtest prorealtime. Can somebody help me out why there is such a big difference? Attached you will find the results of broker IG and prorealtime.

  Thanks in advance

  

  Pjotterd

  Participant

  Topics: 22

  Replies: 33

  Been thanked: 2 times
  06/30/2016 at 9:06 PM #10061
  Nicolas

  Seems that you don’t test your strategy on the same timeframe? Trades are closed on round hour in the first screenshot and that’s not the case in the second one.

  Are you talking about prorealtime CFD?

  

  Nicolas

  Keymaster

  Topics: 83

  Replies: 20899

  Been thanked: 4866 times
  06/30/2016 at 9:21 PM #10064
  Pjotterd

  Yes I’m talking about prorealtime CFD. I have been testing on the same timeframe, that;s just how prorealtime shows backtest closings.

  

  Pjotterd

  Participant

  Topics: 22

  Replies: 33

  Been thanked: 2 times
  07/01/2016 at 9:14 AM #10068
  Nicolas

  I think you are talking about trading real-time in demo? isn’t it?

  If yes, backtest are more efficient because it only deals with condition once per bar, so it doesn’t know at what time your trailing stop occurred, and that’s of course not the case in real-time where stoploss are well executed at correct prices.

  

  Nicolas

  Keymaster

  Topics: 83

  Replies: 20899

  Been thanked: 4866 times
  07/01/2016 at 9:37 AM #10071
  Pjotterd

  Ahhhh, I see, thanks for answering my question, that explains it all.

  Case closed 😉

  

  Pjotterd

  Participant

  Topics: 22

  Replies: 33

  Been thanked: 2 times
  03/26/2020 at 12:36 AM #123334
  VN

  Hi Nicolas, just wanted to clarify to your response above are you saying that in backtest the trailing stop would be calculated at end of each bar but auto trading mode it would move in real time as the underlying price moves?

  

  VN

  Participant

  Topics: 5

  Replies: 51

  Been thanked: 0 times
  03/26/2020 at 10:02 AM #123359
  Nicolas

  That was indeed the case 4 years ago, before PRT add the tick-by-tick backtest engine. The “0 bar issue” is something we don’t talk anymore since years. Always make your backtests with the tick mode option checked.

  

  Nicolas

  Keymaster

  Topics: 83

  Replies: 20899

  Been thanked: 4866 times
• Author
  Posts