Module 5 of 12 · Selling premium

The volatility risk premium

The most direct way to be paid the premium is to sell a straddle: a call and a put at the money, same expiry. You collect both premiums. At expiry you pay out the distance the ETF has moved from the strike, in either direction. If that distance is smaller than what you collected, which is what a premium means on average, you keep the difference.

The lab sells a 30-day straddle on the synthetic ETF every cycle, unhedged, and holds it to settlement, sized so the book runs a 20% annual risk target. It reports what a professional would want to know before the Sharpe ratio: the win rate, the average win against the average loss, the skew of the monthly returns, and the three worst cycles.

The shape is the lesson. Most cycles win a little. A few lose a lot. The equity curve climbs in a nearly straight line, then falls off a shelf, then climbs again. No amount of backtesting changes that shape; it is what being short the tail looks like.

🎛 Short straddle lab

3 pts
4
2%
21

Sharpe ratio

0.16

Monthly skew

-1.08

Winning cycles

61%

avg win 9.9% / avg loss -13.7%

Worst cycle

-76.6%

then -76.5%, -44.4%

Synthetic ETF, 20 years. Each cycle sells an at-the-money straddle 30 days from expiry and holds it, unhedged, to settlement; size is volatility-targeted at 20% a year. Statistics are per cycle (roughly monthly). The histogram is the whole story of premium selling: a tall stack of small wins and a thin tail of large losses.

A high win rate is the symptom, not the edge

Set the premium to zero. The win rate stays near 60% and the average win still looks healthy, yet the Sharpe ratio is zero or negative: the losses are fewer but larger, and they exactly eat the wins. Premium selling always produces a comfortable-looking win rate, with or without an edge, because the payoff is capped on one side and open on the other. Judge it by Sharpe, skew and the worst cycle together, never by the win rate.

Then set the premium to 3 points and the jumps to 12 per thousand days. The Sharpe ratio is positive, but the worst cycle is several times the typical monthly gain. Whether that is acceptable is a sizing question, which Modules 10 and 12 take up: the same strategy at a 10% risk target has half the drawdown and half the income.

Things to try

  • • Compare the premium at 1 point and at 5 points: the win rate barely changes, the Sharpe ratio moves a lot.
  • • Raise the bid-ask to 8% of premium and find the premium level at which the strategy stops paying.
  • • With the premium at 3, try several seeds and note how much the 20-year Sharpe ratio varies on its own.

Original interactive lesson. Labs run on a seeded synthetic ETF in your browser, built to illustrate mechanics, not to predict any real market. Option prices use a textbook model with zero interest rates. Educational analysis, not investment advice.

How are we doing?