Reverse engineering someone else's edge by analyzing open/close trades history?

Discussion in 'Trading' started by heispark, Nov 6, 2019.

  1. 2rosy

    2rosy

    as someone else already alluded to, you would need the entire history of all data sets. A strategy could be looking at a multiple of factors when generating a trade
     
    #31     Nov 19, 2019
  2. TommyR

    TommyR

    not to mention the reverse engineered strategy may not have the emotional ability to deal with future trades
     
    #32     Nov 21, 2019
  3. Handle123

    Handle123

    When I get real bored, for fun I will do it, too many create complicated when they should develop easier methods. Having more than one oscillator to confirm in long run seldom helps. Only can do so many ways to skin a cat or deer. It is easier to do with exact times than arrows on a chart.

    And those who do automation or HFTs never post such things.
     
    #33     Nov 21, 2019
  4. Sprout

    Sprout


    Most likely not and depends on the structure of the methodology.

    Unless it’s a simplistic and singular strategy, attempts at rev-eng might at the most yield ideas to explore vs exposing the full method.

    Since many trader’s orientation to the market is based in fear and greed, any public posts would reasonably have some type of obscurification included.
     
    Last edited: Nov 21, 2019
    #34     Nov 21, 2019
  5. speedo

    speedo

    Better to spend that effort to create your own strategy (ies), that way you fully understand it and know when and how to adjust when market winds shift. You are not the one you are trying to reverse engineer and will see thing differently because you ARE different.
     
    #35     Nov 21, 2019
  6. TommyR

    TommyR

    the complexity of your rules may grow with the number of trades if your learning methods charge it nothing for being inefficient. very sad.
     
    #36     Nov 23, 2019
  7. Zodiac4u

    Zodiac4u

    Many posters and lurkers on these journals are always looking for a free lunch so don't kid yourself. It takes creativity, out of the box thinking and lots of hours of hard work to come up with good ideas. If a person has enough know how, than yes you can easily do it, i know i can. But for me, I've explored 1000's of combinations, and with my present knowledge base, it's just not worth my time to try someones else's crap. Systems are personality based and that's what makes systems unique.
     
    #37     Jan 1, 2020
    jl1575 and themickey like this.
  8. Zodiac4u

    Zodiac4u

    Yes for sure but he got some from news paper articles, he would fade the company that went public long or short, so if Heinold commodities was long beans, he'd try to fry them.
     
    #38     Jan 1, 2020
  9. Snuskpelle

    Snuskpelle

    That's a very good point.

    A parallel - I can technically reverse engineer other people's code by looking at the machine code in a disassembler, but that doesn't mean it's necessarily fun or even a useful way of spending my time. Writing my own shit is 50x faster or even more when there are no debug symbols.
     
    #39     Jan 2, 2020
    Zodiac4u likes this.
  10. themickey

    themickey

    I could trade a certain simple way, but for you to reverse engineer would be most difficult, because you haven't the experience to know the method.
    For example, you would most likely think I'm using an indicator which you are familiar with, but I'm not using an indicator, so how are you going to work that out?

    Besides, buy and sell signals are only a small part, stock selection is another part, if you don't know my stocks selection, that's another issue to contend with.
     
    #40     Jan 2, 2020
    Zodiac4u likes this.