Module 11 of 12 · Running it for real

Expiry, assignment and the roll

ETF options are American: the holder can exercise any day. For a seller that means assignment can arrive early, almost always on a short call just before the ETF goes ex-dividend, when the call's remaining time value is worth less than the dividend. The other hazard is pin risk: the ETF closes at expiry within pennies of the strike and you do not know until the next morning whether you have been assigned shares you must hedge.

Neither is a reason not to sell; both are reasons to know the dividend calendar, to close or roll a short option in its last day when it sits near the strike, and to keep a margin cushion for the shares that may appear. Rolling means closing the expiring option and opening the next cycle, and every roll pays a bid-ask spread.

The lab runs the hedged straddle strategy at four expiry cycles, 14 to 90 days, on the same synthetic ETF at the same risk target, and reports rolls per year, costs, Sharpe ratio, worst cycle and drawdown for each.

🎛 Expiry cycle lab

2%
3 pts
21
cyclerolls / yrSharpereturn / yrworst cyclecosts / yrskew
14 days25.00.4914.0%-69.6%9.2%-2.82
30 days11.90.236.5%-37.6%6.3%-1.52
60 days6.1-0.03-1.1%-39.4%4.5%-0.99
90 days4.0-0.18-8.7%-91.0%3.5%-1.10

14-day max drawdown

117.9%

30-day max drawdown

68.7%

60-day max drawdown

96.0%

90-day max drawdown

286.6%

Hedged short straddles on the synthetic ETF at four expiry cycles, all at the same 20% risk target. Shorter cycles collect premium more often and spend more time in the last week before expiry, where gamma is largest and where the bid-ask is paid most often. Longer cycles trade less and carry more vega. Rolls per year and costs per year are the practical difference; with real ETF options the shorter cycles also run into assignment and pin risk more often.

Shorter cycles pay more often and bleed more often

Fourteen-day cycles roll 18 times a year. Each roll pays the spread, each cycle spends a larger share of its life in the final week where gamma is highest, and the resulting P&L is the choppiest. Ninety-day cycles roll four times, collect more premium per trade, carry more vega and respond more to a change in implied vol than to a single day's move. Thirty days is the common compromise and it is what the published vol indices measure.

At a 2% spread the differences are modest; at 8% the short cycles lose outright. For a retail account trading at the quoted spread rather than inside it, that is the deciding number, and it is why the research for this track used 30-day cycles and measured spreads from live quotes rather than assuming them.

Things to try

  • • Set the spread to 0 and compare the four cycles: with free rolling the shortest cycle has the best Sharpe ratio.
  • • Set the spread to 8% and watch the ranking reverse.
  • • Compare the worst cycle at 14 and 90 days: a long cycle has more time to recover inside the cycle, but a larger single position when the shock arrives.

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?