backtesting is stupid

Discussion in 'Strategy Building' started by jonp, Jul 15, 2010.

  1. Remember when Dr. Deco wrote this on page 6?

    We are observing the Dunning-Kruger effect at work.
    http://www.elitetrader.com/vb/showthread.php?s=&postid=2898687#post2898687

    Well here's what that means. I suggest you think about it.

    The Dunning–Kruger effect is a cognitive bias in which an unskilled person makes poor decisions and reaches erroneous conclusions, but their incompetence denies them the metacognitive ability to realize their mistakes.
    http://en.wikipedia.org/wiki/Dunning–Kruger_effect
     
    #41     Jul 16, 2010
  2. Blofas, that was most elegantly put. Truth can be expressed so much more succinctly than bullshit. I am moved to offer the OP my out-of-print copy of Babcock for a mere grand, US.
     
    #42     Jul 16, 2010
  3. 18, thank you for addressing me properly. I don't get much respect hereabouts for my divinity degree.
     
    #43     Jul 16, 2010
  4. jonp

    jonp

    If backtesting did consistently produce market beating returns,everyone would be billionaires, just compounding their money day after day. the truth is its all about fundamental analysis/general market confidence and perception.backtest lotto results and try to find a pattern...buy a trading strategy book, which in itself is an oxymoron for if the strategy was truly profitable it would never be published.when people make money they thank their amazing psychic indicators, when they lose they blame themselves for not following through with the plan.
     
    #44     Jul 16, 2010
  5. Will someone please answer this post the way I would? I am too busy reoptimizing a consistently profitable strategy to reply discourteously.
     
    #45     Jul 16, 2010
  6. I'll give it a shot.

    jonp:

    noobs who don't know enough => bad results

    people who blindly apply the 'optimize' button, and 99% of people who write books => bad backtesting => bad results

    people who have studied and understand robust, reliable methods => good backtesting => good results
     
    #46     Jul 16, 2010
  7. DT-waw

    DT-waw

    #47     Jul 16, 2010
  8. Truth to tell, DT, if you send a fool down the backtesting path, he will be a dumber trader than if he traded intuitively. So not to worry in this case.
     
    #48     Jul 16, 2010
  9. Backtesting doesn't "consistently produce market beating returns." WHEN DONE CORRECTLY, what it does is help one identify what might work going forward, out of tons of garbage that probably won't.

    But those who know how to backtest properly are as rare as those who can trade profitably so most of the criticism comes from incompetents. Just because you can't pitch a 100 MPH fastball doesn't mean it can't be done.
     
    #49     Jul 16, 2010
  10. YW. Speaking of divinity, did you see my revised Jack/Moses photoshop (with glider head) before they deleted it in the "The Documents by Jack Hershey" thread?
     
    #50     Jul 16, 2010