Selling cash-secured puts on NVDA

Selling a put transfers the downside between $190 and zero to you, and you are paid $5.25 per share for taking it. "Cash-secured" simply means you hold the $19,000 required to buy the shares instead of leaning on margin. Same position, honest denominator.

At August 28, 2026: above $190 the put expires worthless and you keep $525 — that is the maximum this trade can make, $525. Below it you're assigned 100 shares at $190, with an effective cost basis of $184.75 once the credit is applied. That is 8.0% below where NVDA trades today.

The engine puts the probability of keeping the full credit at 74% on NVDA at $200.75 with 46% ATM implied vol on the Aug 28 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 $18,475 — 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 5.4% 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 Aug 28 strikes are $5 apart near the money (2.49% 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. 114k contracts of open interest on Aug 28 is deep enough that multi-leg orders fill near mid at retail size. 24 strikes on that expiry — 50% 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 inverted: the 25-delta CALL implies 2.1% more vol than the put. That is the market pricing upside risk above downside risk — a squeeze, a takeover rumour, or a crowded short. Selling calls into an inverted skew pays better than usual and is riskier than usual for exactly the same reason. The term structure is backwardated — Aug 28 implies 2.9% MORE vol than the following month. That is the signature of a dated event inside the front month, and it is the strongest argument for picking the expiry that sits behind it.

At 46% ATM implied vol, the Aug 28 options are pricing a one-standard-deviation move of $25.16 over 27 days — roughly −12.5% to +12.5%, or $175.59 to $225.91. A short-premium structure here is a bet that 12.5% over 27 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.