long/short straddle strategy

Discussion in 'Options' started by qlai, Mar 5, 2023.

  1. qlai

    qlai

  2. qlai

    qlai

    Oops, I provided wrong link.


    We implement the strategy as follows.
    1. We collect all options data for SP 500 stocks from the OptionMetrics database.
    2. On the day following the standard option expiration date (usually the day after the third Friday of each month), we calculate the slope of Implied Volatility Term Structure (IVTS) for each stock.
    3. We then rank these stocks by IVTS and group them into 10 deciles. Group 1 includes stocks with the most upward-sloping term structure (positive slope) and group 10 includes those with the most inverted downward sloping (negative slope).
    4. For stocks within group 1, we buy equal dollar amount of one-month ATM strad- dles and hold them until expiration. For stocks within group 10, we sell short an equal amount of one-month ATM straddles and hold them until expiration.
    5. We repeat the process on the day following the expiration day.

    Attached the article
     
    tooriginal likes this.
  3. ffs1001

    ffs1001

    Sounds like a nice topic for an academic paper or a "study", but.....er......does it make money?
     
    qlai likes this.
  4. qlai

    qlai

    I like the simplicity of it. You are basically shorting high IV and buying low IV, as predicted by the market itself.
    I am guessing they would be opening short straddle into every earnings release.

    @Matt_ORATS would it be relatively easy to run this back test and compare results to what they posted?
     
  5. newwurldmn

    newwurldmn

    I have read the opposite strategy is more likely to work. High vol tends to outperform and low vol tends to underperform.
     
  6. qlai

    qlai

    But this strategy looks at the term skew, not just current IV rank. Are you saying your strategy was taking skew into account as well?
     
  7. newwurldmn

    newwurldmn

    The study I read used some form of IV rank.

    term structure is just a way of normalizing. For a strategy like this to work there would have to be some anomaly related to the flows (like people tend to overbuy the inverted term structures). That has not been my experience. Term structures are inverted for a reason and they are upward sloping for a reason (especially if that upward slope is 1st decile).
     
    qlai likes this.