Intelligent swing trading algorithm

Discussion in 'Options' started by digitalnomad, Mar 17, 2019.

  1. That's exactly the idea I had in mind, but I wouldn't even deal with historical option data, or backtested option trades. I'd just design the algo to run live trades based on underlying signals, a month or two on a sim account for stability checks, and then small capital commitment. Perhaps I can get my front end to send an email alert to a developer who can build an App to work the option logic from there. I have no intention of coding this up. My brain is fried.
     
    #11     Mar 18, 2019
  2. Say what? Last time we communicated, we discussed game/gambling theory, which by the way, still fascinates me :)
     
    #12     Mar 18, 2019
  3. tommcginnis

    tommcginnis

    And theta burn -- 100% suckage for that long investment.
     
    #13     Mar 18, 2019
  4. I was hoping the back month approach might alleviate some of that.
     
    #14     Mar 18, 2019
    tommcginnis likes this.
  5. qlai

    qlai

    Did you try paper trading it (forward testing it)? Wouldn't that answer your questions? Just curious.
     
    #15     Mar 18, 2019
    tommcginnis likes this.
  6. On the underlying yes, forward and back with positive expectancy. The option part is what needs to be built and live tested. For this swing system to be worth my time and capital, a leveraged vehicle is needed, enough that the performance can at least 1/2 match the returns on my intraday systems.
     
    #16     Mar 18, 2019
  7. qlai

    qlai

    Right, I am not understanding why you can't forward test the options part ... If it's a swing system, you should have plenty of signals you can take manually, no?
     
    #17     Mar 18, 2019
  8. It averages few trades per instrument, per month, so theoretically to get a significant enough sample size, would take years. I don't mind throwing some real money at it for a year or two. If I lose 1/2 my stake, consider it a research cost. If it works, than the system is monetized, instead of sitting on the shelf collecting dust.
     
    #18     Mar 18, 2019
    tommcginnis likes this.
  9. "The algo is also intended to exit the option position based on its underlying stop signal, so that means a CBOE market order (bend over baby)."
    I would use a limit order and add $0.10 to $0.20 or so to the mark price, so you don't get screwed by the MM yet still get filled (99% of the time).
     
    #19     Mar 23, 2019