Announcement

Collapse
No announcement yet.

Partner 728x90

Collapse

Backtesting and config

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Backtesting and config

    Hi,

    I'm pretty new to ninja trader and have a question concerning configuration and parameters when backtesting or running a strategie on realtime data.

    My setup is: IB and Ninjatrader. For historical data I use NinjaTrader servers, like it was written somewhere.


    Now I've put together a strategie for forex trading (EURUSD).
    When I run a backtest, it works out pretty well. It does about 40% if I run it for last year, the same if I run it for 2006.

    Now, when I try in real time, it just seams to be a whole different strategie it does not work at all. It just keeps loosing money.

    Has anyone an idea what could be wrong with my backtesting or my realtime settings?

    My settings when BACKTESTING are:

    Data series:
    - Type: Minute
    - Value: 1

    Fill type: default

    No comission

    Min. bars required: 20

    No slipage

    Exit on close

    default quantity: 1

    gtc

    My data source settings for BACKTESTING:

    Tick size: 0,0001

    Point value: 100000

    Currency: UsDollar

    Margin: 0


    My data source settings for REALTIME (IB):

    Tick size: 0,00005

    Point value: 1

    Currency: UsDollar

    Margin: 0


    -------------------

    When I run the strategy, I use the same settings as when backtesting.
    My realtime account uses a lot size of 1 lot = 10000 , so I use 10000 for backtesting value and only use 1 for realtime value.

    I think that it might have something to do with the tick size. Also, when I use "Type: Tick" instead of "Minute" in Backtesting, the Backtesting does not work out very well anymore.

    Of course the problem could also be slipage or comission, but entries and stops just seem to odd...


    Thanks for your help,

    Dennis

    #2
    Please review the information on the following link to see if there is anything of value.

    RayNinjaTrader Customer Service

    Comment


      #3
      Hey Ray,

      thanks for the very quick response.
      For sure that's good stuff to know.


      Is my assumption right, that if I'm using for example tick values for
      stop loss orders

      then if for my historical data the tick size is 0,0001

      and for my realtime data the tick size is 0,00005

      this would just mess up my whole strategie, because it would divide my stops into halve of them?


      Thanks,

      Dennis

      Comment


        #4
        There is *no* need to change the instrument's ticksize in case you are doing this. All you need to do is have sub-pip enabled under Tools > Options > Misc.

        See IB Connection Guide Information - http://www.ninjatrader.com/webnew/su...brokers.htm#FX

        This should keep backtest and real-time consistent with an internal ticksize of 1/10 of a pip. This means if you want a 10 pip stop loss, you specify 100. Please double check this in backtest to confirm.
        RayNinjaTrader Customer Service

        Comment


          #5
          Hello Ray,

          thanks for the hints. I now activated the sup-pip thing.

          What I'd like to do, is to just set the backtesting data to fit perfekt to my broker and then understand with what amount I'm trading, because that's kind of confusing to me.

          As historical data feed I'm using the NinjaTrader Servers,
          now I changed my settings (original settings in brackets)

          Tick size: 0,0001 (0,00001)
          Point value: 100'000'000 (100'000)

          As real time data feed I use IB, I changed this to the following settings (original settings in brackets)

          Tick size: 0,0001 (was: 0,00005)
          Point value: 1 (was: 1)

          My two questions:

          1. Is it okey to change the tick size like I did, when I backtest and realtime with these changed settings, or does this bring up any issues?

          2.

          a) Now, if I'm trading minilots of 10'000,
          would these settings give me the chance to put in the exact number of minilots (in default quantity field of backtest) I'm trading, when backtesting? (instead of the real $ amount of money)





          Thanks,

          Dennis

          Comment


            #6
            1) Never change the ticksize, there is no need to do this.

            2) Please see if this reference helps - http://www.ninjatrader-support.com/H...trategies.html
            RayNinjaTrader Customer Service

            Comment


              #7
              Hello Ray,

              sorry for continuing this.
              Switching around the tick sizes and point values does change my results a lot, same thing with the sub-pip checked.

              So could you please point me to the correct values (tick and point values)
              when sub-pip is checked for historical Ninja Data and for IB?
              That would be a great help, becaus this seems realy important for the results and I can't find a lot about it in the documentation.D
              Depending on these values my strategies either work or not, but I don't know the correct valus.

              Thanks and have a nice weekend,

              Dennis

              Comment


                #8
                Make sure you are on the latest NT 6.5 available from our website, its 6.5.1000.1.

                - Tools > Options > Misc > Reset Instruments
                - Enable sub pip
                - Press OK

                This is all you have to do. Don't touch any instrument settings for currencies. Its all set up.
                RayNinjaTrader Customer Service

                Comment


                  #9
                  Hello Ray,

                  thanks for your advice.
                  I did like you recommended.

                  One final question concerning the sub-pipe impact on the performance summary.

                  I suppose, that trading with the quantity of 1 (I know it's real currency and not one lot) and the performance summary showing an average trade of 0,00006 this would be in real world 0,0006 (and equal 6 pip), is that correct?

                  Thanks for your great help,

                  Dennis

                  Comment


                    #10
                    I believe that would be 6/10 of a pip.
                    RayNinjaTrader Customer Service

                    Comment


                      #11
                      Hello Ray,

                      thanks for all the answers, 6/10th seem to be correct, when comparing backtests with and without sub-pip.

                      Dennis

                      Comment

                      Latest Posts

                      Collapse

                      Topics Statistics Last Post
                      Started by Geovanny Suaza, 02-11-2026, 06:32 PM
                      0 responses
                      651 views
                      0 likes
                      Last Post Geovanny Suaza  
                      Started by Geovanny Suaza, 02-11-2026, 05:51 PM
                      0 responses
                      370 views
                      1 like
                      Last Post Geovanny Suaza  
                      Started by Mindset, 02-09-2026, 11:44 AM
                      0 responses
                      109 views
                      0 likes
                      Last Post Mindset
                      by Mindset
                       
                      Started by Geovanny Suaza, 02-02-2026, 12:30 PM
                      0 responses
                      574 views
                      1 like
                      Last Post Geovanny Suaza  
                      Started by RFrosty, 01-28-2026, 06:49 PM
                      0 responses
                      577 views
                      1 like
                      Last Post RFrosty
                      by RFrosty
                       
                      Working...
                      X