Selling cash-secured puts on NVDA

Selling a put transfers the downside between $215 and zero to you, and you are paid $4.22 per share for taking it. "Cash-secured" simply means you hold the $21,500 required to buy the shares instead of leaning on margin. Same position, honest denominator.

At October 30, 2026: above $215 the put expires worthless and you keep $422 — that is the maximum this trade can make, $422. Below it you're assigned 100 shares at $215, with an effective cost basis of $210.78 once the credit is applied. That is 6.3% below where NVDA trades today.

The engine puts the probability of keeping the full credit at 74% on NVDA at $225.07 with 32% ATM implied vol on the Oct 30 expiry. High win rate, capped payoff, uncapped-to-zero loss — the risk profile of every short-premium trade. It is not free money; it is a rent cheque for underwriting someone else's insurance.

Max loss is $21,078 — the strike, less the credit, times 100, if NVDA goes to zero. That number is not theoretical on single names; it is simply unlikely. Size the position against that figure, not against the premium.

The real-world failure mode is a gap, and NVDA has the catalysts for one: earnings (an event unto itself), AI-capex headlines, and semi-cycle sympathy moves. A put sold 4.5% out of the money offers no protection at all against a move twice that size overnight.

The structural problem with short premium is not the loss rate, it is the loss SIZE. A long run of small wins funded by an occasional large loss looks like skill on a monthly statement and like variance on a five-year one.

The premium is enormous and so is the reason for it. Selling puts on NVDA outside an earnings window is a defensible short-vol trade; selling them through one is underwriting a distribution whose tails you have watched print in real time. If you are running the wheel here, the assignment is not the hypothetical — it is the base case at least once a year, and your basis needs to survive it.

NVDA's Oct 30 strikes are $5 apart near the money (2.22% of spot). That is a coarse ladder: one rung is a large fraction of the implied move, so precision on the short strike is an illusion. 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. Neither side of the chain is being singled out, which is the condition under which a symmetric structure like a condor is actually symmetric. 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. A short-premium structure here is a bet that 9.6% over 33 days is more than NVDA will actually use. That is the thesis, stated honestly.

The NVDA-specific failure mode: Sizing a short-premium NVDA position off the credit rather than off the gap. The credit is large because the gap is large.