How crazy am I to believe that this system can work live?

Discussion in 'Automated Trading' started by trader221, Oct 27, 2023.

  1. %%
    Good points.
    IT could work, not that this is a typical OCT, its not for NQ or ES. It could work anyway.
    I doubt if longer term, largest lose trade will be $2k+ largest win trade will be $2k, long term.
    Possible but not likely.
    But using a good trend filter could improve it.
    I would not want an average win of $80+ average loss of $120. But that's just me;
    nice green equity OCT graph:caution:
     
    #21     Oct 27, 2023
  2. zghorner

    zghorner

    to be diligent historical data would be best. If winging the value just depends on the instrument.

    This is a big reason why I only trade a single liquid instrument during its most liquid hours only...helps bridge the gap between fantasy and reality.

    lol, i posted this in my recent thread about backtesting. it is so true it hurts.

    upload_2023-10-27_12-11-4.png
     
    #22     Oct 27, 2023
    Occam, piezoe, taowave and 6 others like this.
  3. hilmy83

    hilmy83

    At least backtest it from 2020 till now. I think it gives pretty wide range of market conditions in that timeline.

    If it's still profitable, run with it live.

    Anyway why are you even doing funding account crap? You short on funds?
     
    #23     Oct 27, 2023
  4. smallfil

    smallfil

    How can you account for something you have no control over? The market makers set the stock prices not you.
     
    #24     Oct 27, 2023
    trader221 and murray t turtle like this.
  5. SunTrader

    SunTrader

    On what time frame is this tested on?

    Tick-by-tick is the only way I would trust any backward or forward results.
     
    #25     Oct 27, 2023
    trader221 and murray t turtle like this.
  6. Didn't read other posts so probably others said the same, but you need to backtest for like 10 years to have any understanding of expected profit and risk.

    There's a misunderstanding around sample size when it comes to short time frames. The diffferent personality of markets has to be tested, as much for scalping as anything else. Don't let the large amount of trades lead you to a false sense of security around profitability and risk.
     
    #26     Oct 27, 2023
  7. %%
    Good read.
    Even more so, when many may not even call a number of trades in OCT ''a large number '' of trades. Even though that could be, for one OCT.
    Some things need to be repeated .
    It could work.
    [Another personal point, I like to assume max slippage + seldom disappointed there LOL:D:D] But the way I trade, slippage+ exchange fees aren't my main concern @ all .
     
    Last edited: Oct 27, 2023
    #27     Oct 27, 2023
    trader221 likes this.
  8. zghorner

    zghorner

    - Would recent data be exponentially more import?
    - How likely is it that systems are being modified unfavorably to suit conditions not relevant to today?
    - What is the expected life of an automated system before it requires tweaking?

    Obviously performance needs to be constantly monitored to spot weaknesses...and the system needs a real chance to prove itself without constant tweaking but It makes sense to me at least that recent data should be valued higher than data from 2020.
     
    #28     Oct 27, 2023
  9. kmiklas

    kmiklas

    What platform are you using to backtest?
     
    #29     Oct 27, 2023
  10. its a forward testing!
    Nninjatrader 8
     
    #30     Oct 27, 2023
    kmiklas likes this.