Every structure in option strategies can be drawn as a kinked line on a payoff diagram, because every leg expires on the same date. A calendar spread cannot be drawn that way, and that single fact is the whole difficulty. Sell the one-month 100 call and buy the three-month 100 call, and on the day the front leg expires the back leg still has two months of life and therefore a value that depends on where the underlying is, what implied volatility is doing, and how much time is left. There is no static expiry payoff to memorise, only a valuation, and a structure that has to be valued rather than drawn is a structure most people size by feel. The remedy is to price it at the front expiry across a grid of underlying prices, which is what Worked Example 10 does.
A diagonal spread is the same trade with the strikes allowed to differ as well as the dates. Everything below applies to it, with the added complication that the position now has a directional view built into the strike gap.
Advertisement
What are the Greeks of a calendar spread?
The Greeks point in the opposite direction from the vertical spreads in option strategies. Both legs are long the same underlying and the same strike, so the direct exposures largely cancel and what is left is a bet on the shape of time. Take the illustrative book: short one 30-day 100 call at 24% implied, long one 90-day 100 call at 21%, with the underlying at 100 and r = 4%.
| Short 30-day 100 call | Long 90-day 100 call | Net | |
|---|---|---|---|
| Price | 2.9072 | 4.6470 | debit 1.7399 |
| Delta | 0.5328 | 0.5583 | +0.0256 |
| Gamma | 0.0578 | 0.0378 | −0.0199 |
| Vega (per 1.00) | 11.3988 | 19.5979 | +8.1992 (0.082 per point) |
| Theta (per day) | −0.0511 | −0.0285 | +0.0226 |
Read the last two rows together, because they are the trade. The position is long vega and long theta at the same time, which no single-expiry structure can be. It manages that because vega scales with √T and theta scales with 1/√T: the long back-month leg carries most of the volatility exposure while the short front-month leg carries most of the decay. The price of the combination is short gamma, which is the thing that hurts, and it hurts most on a large move in either direction, because a large move takes the front-month leg to intrinsic value while leaving the back-month leg's time value largely unrecovered.
When does a calendar spread work?
The term structure is what makes the trade attractive or not. A calendar spread is a position in the slope of the volatility term structure, not in its level. It is bought when front-month implied volatility is rich relative to back-month, which is the normal shape ahead of a dated event (an earnings release, a policy meeting, an election) that falls inside the front expiry and outside the back one. It is sold when the term structure is unusually steep upward and you expect it to flatten. Two conditions have to hold together for the long calendar to work, and confusing them is the standard error. The underlying has to stay near the strike, because the structure is short gamma. The back-month implied volatility has to hold, because the structure is long vega. Get the first right and the second wrong and the trade still loses. At the front expiry with the underlying pinned at 100, the position is worth 3.7223 against a 1.7399 debit if back-month volatility holds at 21%, and 3.0803 if it collapses to 17%. A four-point volatility fall takes 0.6420 out of a 1.9825 profit.
Why do earnings calendar spreads often fail?
The event-calendar trap is worth naming, because it is where most of the money is lost. The most common retail version of this trade is to sell the expiry that contains an earnings date and buy the one after it, on the theory that the front leg is rich. It usually is rich, and that is not enough. After the event the front-month volatility collapses as intended, but the back-month volatility usually falls too, because part of what it was pricing was the same event, and the residual is a short-gamma position held through the exact moment the underlying gaps. A calendar spread wants the event to be contained in the leg you sold and absent from the leg you bought, and on a single name with quarterly earnings that separation is rarely as clean as the pitch implies. DV1.05 carries the term-structure evidence in full.
What are the settlement and early-exercise risks?
Sizing brings its own complication: two dates means two sets of contract mechanics, and on Indian single-stock options it means two settlement events: the front leg physically settles at its own expiry while the back leg remains open, which is a delivery obligation and a margin ramp the payoff picture never shows. On US equity options it means early-exercise risk on the short leg around a dividend. Neither risk is priced in a payoff diagram, which is the deeper reason this structure sits in its own section.