Buying NVDA calls: the math before the ticket

A long call is the right to buy 100 shares at $225 until October 30, 2026. You pay $895 for it and that debit is the entire risk — max loss $895, no margin calls, no assignment exposure.

The payoff below the strike is flat at −$895; above it, P/L rises one-for-one with the stock and turns positive at $233.95. Upside is unlimited, which is the whole appeal.

Every day you hold it, theta takes a slice. At 32% implied vol with 33 days left, that decay is modest now and vicious in the final fortnight — an ATM call loses roughly half its remaining extrinsic value in the last third of its life.

The engine's 34% probability of profit is the honest framing: long calls are low-probability, high-payoff. That is not a criticism — it is the shape you are buying — but it is the opposite of how most retail traders size them.

Being right and still losing is routine: NVDA can rise 2.0% and this call still expires worthless because the breakeven is $233.95.

Vol crush after earnings (an event unto itself) can take 20–40% of an ATM option's value overnight even with the stock flat. If you buy a call into the event, you are paying event-priced vol.

The ceiling on a spread is a real cost, not a theoretical one. It is paid exactly in the scenarios where your thesis worked best, which is when it hurts most to notice.

The one name on this list where a call spread's capped upside genuinely costs you money often enough to notice. NVDA's post-earnings moves have repeatedly cleared the short strike of any sensibly-priced spread, so if your thesis is a re-rating rather than a drift, the spread is the wrong instrument and you should pay for the tail.

NVDA's Oct 30 strikes are $5 apart near the money (2.22% of spot). Coarse enough that the strike you want frequently does not exist, and the nearest rung is a different trade. 65k contracts of open interest on Oct 30 is workable around the money and thin in the wings — width costs more here than the ladder suggests. 21 strikes on that expiry — 100% of the board — carry prints that agree with their own implied volatility and hold up across the ladder, and those are the strikes priced here. Enormous volume and open interest; complex structures fill near mid even in size, including through the print.

Skew is ordinary — the 25-delta put implies 2.5% more vol than the 25-delta call, about what an equity surface looks like when nothing unusual is being priced. Nothing on the surface argues strongly for one direction of structure over the other. The term structure is in contango: the back month implies 5.0% more vol than the front. Calm now, uncertainty later — which rewards selling the front month and makes the back month an expensive thing to own outright.

At 32% ATM implied vol, the Oct 30 options are pricing a one-standard-deviation move of $21.67 over 33 days — roughly −9.6% to +9.6%, or $203.4 to $246.74. That band is the free part of the move. Anything your structure needs beyond it is the part you have to be right about.

The mistake this name punishes hardest: Sizing a short-premium NVDA position off the credit rather than off the gap. The credit is large because the gap is large.