A question about Sim Accounts

Discussion in 'Trading' started by Darc, Apr 19, 2023.

  1. Darc

    Darc

    If someone uses a 100% Rules/Mechanical System to trade a Sim Account and also realises that it's performance isn't indicative of reality due to slippage etc, that should be ok to test just the Rules right?

    And they then use very small Position sizes to then Trade a real Account, upon success of the Sim Account, that should be Ok to test a new Strategy right?

    Thanks.
     
    Sekiyo and murray t turtle like this.
  2. Should be OK.

    You always have to deal with spreads and slippage.... but if you're trading an issue that is highly liquid, that's not really a concern.... big stocks, large ETFs, stock index futures, etc.

    The greatest value to a SIM account is to reveal strategies that "don't work". So.. if you think you've got something that actually does work, trade it small with real $$$ until you're confident to trade bigger.

    And yes... your method should be "rules based".

    FWIW...

    KISS, baby! :)
     
    Last edited: Apr 19, 2023
    semperfrosty and Darc like this.
  3. M.W.

    M.W.

    If the performance is not fill sensitive then yes. That usually is only the case for non intraday strategies.

     
    Darc likes this.
  4. oshjdf

    oshjdf

    If you're trading top liquid instruments on intraday, then slippage rarely goes beyond 1 point unless you're trading big size which you will realize the limit one day. You don't even need to paper trade.
     
    murray t turtle and Darc like this.
  5. dholliday

    dholliday

    IBs paper trading account seems to be very realistic (at least for stocks). I have run the sim account and real account at the same time (many times). Very close to the same results, averages out to a wash.
    As an example; I had to wait for about 6 minutes this morning for one of my shorts to fill in a sim account.
    I wouldn't run a system without running on sim first, and then running sim and real in parallel.
    The results always average the same for me.
     
    rb7 and Darc like this.
  6. %%
    SOUNDs fine\
    strange how almost never does a real time % do much better than sim or backstest.
    Even allowing a bit for slippage.
     
    Darc likes this.