Can a consistently losing strategy be profitable if I do the opposite of it?

Discussion in 'Strategy Building' started by Amatrue, Oct 25, 2020.

  1. Amatrue,

    You high resolution with tick data for your back testing. Try it for 6 months of data and see what you get.
     
    #11     Oct 25, 2020
    Amatrue likes this.
  2. MrRenev

    MrRenev

    It's not necessarily fake. There are some consistently winning strategies on 5 minute charts and it is as smooth as his chart since you are taking alot of small trades.
    I can tell you I know for a fact there are strategies that will win like this, yes between 2009 and 2020.
    It is also highly likely they will keep working as no one is trading those, especially not on such small TFS.
     
    #12     Oct 25, 2020
  3. Amatrue

    Amatrue

    How would I go about using tick data to backtest if it changes the parameters of my indicators which is based off of the 5 minute timeframe?
     
    #13     Oct 26, 2020
    SimpleMeLike likes this.
  4. heispark

    heispark

    No.
     
    #14     Oct 26, 2020
  5. maxinger

    maxinger

    Just partner with another trader and make
    some agreement.

    When you buy, he will sell immediately.
    When you sell, he will buy immediately.

    net profit / loss to be shared between both of you.
     
    #15     Oct 26, 2020
  6. You need to use NT8, not NT7. I used NT7 5 minute bars and they give incorrect back testing results because the back testing software execute the code per bar.

    In NT8, Strategy Analyzer there is an option for Fill Resolution. Choose 1 minute, so you get better resolution in your back testing and then back testing on 5 minute bar and see if you see a difference. Every 1 minute, your code will execute. Also, you can use tick replay, every tick your code will execute, I think. I do not use NT anymore. I am going off memory.

    Also, check about 20 trades and see if it make sense to you.

    Your results look an error somewhere.
     
    #16     Oct 26, 2020
    Amatrue likes this.
  7. Amatrue

    Amatrue

    Wouldnt that result in loss in comissions :confused: or was that the joke :D
     
    #17     Oct 26, 2020
    murray t turtle likes this.
  8. Amatrue

    Amatrue

    care to elaborate?
     
    #18     Oct 26, 2020
  9. %%
    IT COULD work;
    with no commissions+ perfect executions.
    Is it likely to work??? It Depends.
    Even if it was a joke i usually look for truth in a joke sometimes. I like a PSAR[parabolic stop + reverse] some times anyway.:D:D:D:D:D:D:D
     
    #19     Oct 26, 2020
  10. heispark

    heispark