A calendar spread sells a near-dated option and buys a longer-dated one at the same strike, paying a net debit. It profits when the near option decays faster than the far one loses value, which makes it a trade on the volatility term structure: richest when near-dated volatility is expensive relative to far-dated.
Differential decay. The short near-dated leg bleeds time value at the steep end of the decay curve while the long far-dated leg sits on the shallow end, and the position collects the difference as long as the stock stays near the strike. Its risks are the mirror image: a large move in either direction hurts (both legs go far from the money and the differential vanishes), and a collapse in far-dated implied volatility hurts the long leg specifically. Maximum loss is the debit paid, which keeps the structure in the defined-risk family.
An earnings announcement inflates the expiration that spans it far more than later expirations: the event's uncertainty lives almost entirely in the front expiry. Selling that inflated front expiration and buying a calmer later one, at the money, harvests the collapse of event premium (the IV crush) the morning after the print while the long leg barely moves. The screen that keeps this honest checks that the event premium is genuinely rich first, comparing the implied announcement move against the stock's own history of realized earnings moves, and passes only when the market is paying above that history. Double calendars, one call calendar above and one put calendar below, widen the profitable band for stocks that drift on the print.
In steep contango away from any event: when far-dated volatility prices well above near-dated as a resting state, the calendar's long leg is the expensive one, and the structure fights its own entry price. The favorable configuration is the opposite, near-dated volatility elevated relative to far, whether from an event, a news cycle, or short-term stress, because then the spread sells the expensive end. Reading the curve before choosing the structure is the whole game; the same curve that sizes index positions also picks between condors and calendars.