Expected vs. Actual Win Rates When Selling Options [SPY Short Strangle Backtest]

In this episode we're going to use our new Trader's Toolbox and options backtesting software to test win rates when selling options.
Expected vs. Actual Win Rates When Selling Options [SPY Short Strangle Backtest]
Kirk Du Plessis
Jun 8, 2017

The truth is that many people have this huge "rub" with options trading and expected vs. actual win rates. They assume that if we place a new trade with an initial probability of success of 70% that the trade will win at exactly 70% over the long-term. As you'll see today, that's just not the case when you account for the implied volatility premium inherent in the options market. We'll look at two different SPY short strangle backtests.

Key Points from Today's Show:

SPY Short Strangle Backtest

  • Daily entries with 30 days until expiration.
  • No IV rank, no profit taking exit, and no stop loss exit.
  • Wanted to test expected versus actual win rates.
  • Sold options with 20 Delta on either side.
  • A 60% chance of success upon trade entry.

Results:

  • Win rate was 79.6%, which is way more than what the initial probability suggested.
  • Had a drawdown of 37%, which was all recovered within 54 days.
  • Created a smooth, stable equity curve after 2008.
  • Won 145% total return compared to the market, with an annual CAGR of 9.41% and a Sharpe ratio of 0.62.
  • The average P&L earned per trade was 37.33%.
  • Overall a very profitable trade, market neutral, always selling premium and ends up working out most of the time.

Adjustments:

  • Decreased days until expiration to 10 days.
  • Sold options at the 15 Delta, 15% probability of being in the money options on each side.
  • Expected to win at least 70%.

Results:

  • Saw a 125% return, and an annual CAGR of 8.48%.
  • Sharpe ratio had a huge increase to 7.6%, creating a large stabilization in the equity curve.
  • The win rate was 82%, which was more than the probability suggested.
  • Only had a max drawdown of 13% in the portfolio at any time.
  • Collected an average of $66 in premium.
  • Average P&L per individual trade was 40.87%.

Conclusion:

  • Although weekly strategies collect more premium they do not ultimately become more profitable than a monthly strategy.
  • Although gains were less, the stabilization of the equity curve was much smoother and created the opportunity for a much higher Sharpe ratio.
  • To create a more dynamic portfolio, you want the returns of the monthly contract, but also want the stability of the weekly contracts.

Trade smarter with automation