Minimum number of trades required for backtesting results to be trusted

Discussion in 'Strategy Building' started by helpme_please, Mar 2, 2021.

  1. Exactly
     
    #21     Mar 3, 2021
  2. Bekim

    Bekim

    For me the minimum I look at is 4 quarters
     
    #22     Mar 6, 2021
  3. CannonTrading_Ilan

    CannonTrading_Ilan Sponsor

    I try to backtest 15,000 bars.

    I will usually leave the last 2-6 months ( depending if I use short term or swing trading) and then see how my system did going "forward to the data sample I did not test/ optimize.
     
    #23     Mar 8, 2021
    murray t turtle likes this.
  4. %%
    THAT Title Maybe True ''....Trading Edge Maybe Scarce...'' BUT since Most Investors Make Big Bucks Buying every Month for 4o Years....... Depends on your goals..Sorry my sound does not work on the video, but his title maybe right.
    NOT a prediction + not suitable for all investors/traders.......................................................[For trading/ investing QQQ, look @ all the data, including over. 100% run up+80%/+ drawdown]
     
    #24     Mar 9, 2021
  5. Shmo007

    Shmo007

    To have a robust system, you should really backtest on a number of different timeframes: during periods of strong uptrends; strong downtrends; and periods of extended consolidation. This way you'll identify the strengths and more importantly the weaknesses of your system. You may even want to test the system across different asset classes like commodities, forex, stocks; index futures, and fix income to really see how robust the system is.
     
    #25     Mar 13, 2021
    CannonTrading_Ilan likes this.