Module 6 of 12 · Selling premium

Delta hedging a short straddle

An unhedged short straddle has two risks tangled together: the ETF going somewhere, and the ETF moving around a lot. The premium only pays for the second. Delta hedging removes the first. At every close, hold enough ETF shares to offset the straddle's delta, so a small move up or down leaves the book flat. What is left is the gamma-versus-theta trade from Module 3: you lose on the size of the moves and earn on the passage of time.

Hedging does not add return. The same premium is collected either way. What it does is strip out the directional noise, so the premium shows up at a higher Sharpe ratio and a smaller worst case. The price is the hedge itself: a trade in the ETF every day, each paying a spread, and a book that needs attention at every close rather than once a month.

The lab runs the same straddles hedged and unhedged on the same synthetic path and decomposes the hedged book into premium collected, settlement paid, hedge trading P&L and costs, per year.

🎛 Hedge simulator

3 pts
1 bp
4
21

Sharpe, hedged

0.23

Sharpe, unhedged

0.16

Worst cycle, hedged

-37.6%

Worst cycle, unhedged

-76.6%

Premium collected / yr

$226,789

Settlement paid / yr

$213,969

Hedge P&L / yr

−$161

Costs / yr

$6,254

Both books sell the same straddles at the same size ($100,000 capital, 20% risk target). The hedged book trades the ETF at every close so its delta is zero, paying the hedge cost on every share traded. Per-year dollar figures are for the hedged book: what it collected in premium, what it paid at settlement, what the hedge trades made or lost, and what it paid in costs. The hedge does not change the premium; it removes the directional noise around it, so the same edge shows up at a higher Sharpe ratio, with a thinner but still fat left tail.

The hedge turns a bet on the ETF into a bet on volatility

The unhedged line is lumpier and its worst cycle is worse, but notice that over 20 years the two end up in roughly the same place. That is the decomposition telling you where the money comes from: premium minus settlement is the premium you were paid for carrying the tail, and the hedge P&L hovers near zero. Hedging changes the path, not the destination. For a small account the choice is practical: can you trade the ETF every day at a basis point, or not?

Push the hedge cost to 10 basis points and the hedged book falls behind. For the big ETFs the real cost is a basis point or less; for a thin one it is several. The research for this track found the hedged version viable only on the eight most liquid funds.

Things to try

  • • Set the premium to 0: both books lose money, the hedged one more slowly and more evenly.
  • • Compare the worst cycle hedged and unhedged with 12 jumps per thousand days; the hedge halves it but cannot remove it.
  • • Find the hedge cost at which hedging stops improving the Sharpe ratio.

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?