When a model becomes extinct

Discussion in 'Automated Trading' started by Zortauk, Apr 14, 2021.

  1. Zortauk

    Zortauk

    I have a question(worry) that i cant find an answer for. I will have to live through I guess, but meanwhile I wanted to ask if other traders had any similar experience and what happened. I am an independent trader( daytrader) with 20 plus years experience in stock trading. I wrote an algo..maybe 15 years ago..and has been consistently profitable. The results were not great, but helped me pay the bills. After the Pandemic hit, something happened..it stopped being profitable. I understand that we are living in extraordinairy times, but i am wondering if there are any other traders out there..who have models with consistent positive returns for a long time..and then stopped working. WHen do you decide..the model is done..can it come back? Because it had such a long run..i am not willing to give up on it yet, but one has to pull the plug at one point.. ANy thoughts, experiences are welcome..
     
  2. jys78

    jys78

    "i am wondering if there are any other traders out there..who have models with consistent positive returns for a long time..and then stopped working"

    The answer is a resounding yes.
     
  3. SabreMan

    SabreMan

    If you know the premise of your edge.

    You should be able to decide if the edge is still valid.
     
  4. Zortauk

    Zortauk

    But the question is after how long of a time..one would give up on the model after it had such a long time of consistency. Outsiders say that it is normal that it is not performing because it is the PANDEMIC, but i dont really understand how and why pandemic has changed trading. MY model is purely based on price action not looking at volume or any other indicators.
     
    murray t turtle likes this.
  5. Zortauk

    Zortauk

    To be honest, since it is based on purely price action. I do not understand why it was making money and i do not understand why it stopped either. There was an edge and after the pandemic the edge seemed to dissapear( in fact last summer it was decent as well). I believe last summer, there were talks of returning to some kind of normalcy(maybe thats why..)

    I understand in the sense that it hits my stops more quickly then it hits my target.
     
  6. SabreMan

    SabreMan

    What was the monthly win % and previous longest monthly losing streak, prior to 2020.
     
  7. Learn "Price TA".... then you won't have to deal with an approach that "stops working". Price TA always works in all markets and all time frames.

    KISS, baby. As always!

    :)
     
  8. Zortauk

    Zortauk

    I run the model on stocks that reported earnings. So i ran it every quarter for 4-6 weeks. I had only one small losing quarter..and i believe there was a hurricane or sth NY area( cant remember that well). Other than that..i did not have any negative quarters until pandemic. Out of stocks, i traded..my win/lose ratio was probably 50-50, but my gains outweighed losers.

    One theory i have for this is that since pandemic..there are all these no-name stocks making extraordinary moves..so that..the money maybe going into these names, not to stocks reporting earnings. IF that is the case, when will earnings report have the old effect again? If that is not the case, i have no explanation.
     
  9. I would look at the return contribution. If you're on IB run a portfolio analyst report and I can help you understand it.

    Most often the thing that goes wrong with models is that there is 1) some risk that you have an edge on and are harvesting 2) but your trade is not "clean" which means you are exposed to other factors 3) over time, you can become unknowingly overweight that undesirable factors...
     
  10. I guess it will return to work next quarter.
     
    #10     Apr 14, 2021
    murray t turtle likes this.