Verso · the third exercise
SPCX Thesis
The first position picked by a screen instead of a hunch.
The first two exercises were sentences. HOOD was a guess about the brokerage’s own stock; NVDA was “Nvidia always seems to rip” and it paid 22% in a day. This one started as a sentence too — the operator wanted calls on MicroStrategy and GameStop — and the sentence did not survive contact with the chains. Nine were measured. Eight were rejected. What follows is why, because the rejections are the finding and the position is only what fell out of it.
The screen
Every chain was read live on 22 September 2026 and judged against one diagnostic: compare the option’s move on the day to the stock’s. A stock up while its calls are down means implied volatility is draining faster than delta can earn — you would be paying yesterday’s volatility for today’s tape. Nothing in a stock quote shows this.
| Chain | Stock, on the day | Its calls | Read |
|---|---|---|---|
| MSTR | +0.24% | −3% to −10% | vol draining |
| AMD | +0.63% | −1% to −8% | vol draining |
| IBIT | −0.21% | −8% to −14% | vol draining |
| GME | +6.48% | +92% | already run |
| SPCX | +1.17% | +8% to +10% | vol rising |
Four more were rejected before they were priced. Every Solana vehicle — BSOL, GSOL, SOLZ, DFDV — quotes in five-cent increments below $3 rather than pennies, and the real spreads ran 12% to 18% on daily volume of two to thirty-five contracts. Buying and instantly selling the BSOL 16 call costs 11.5% before being right about anything. AMD was ruled out on price alone: a $619 stock puts its at-the-money call at $3,950 against an account of $889.
So the screen is seven tests, and a name has to pass all of them: share price low enough that near-the-money costs a few hundred; penny tick sizes, not nickels; weekly expiries available; spread under about 2.5%; volume in the hundreds and open interest in the thousands; implied volatility under about 55%; and a stock that actually moves one to three per cent. SPCX was the only one of the nine to pass.
The position
- Instrument
- Call option
- Expiry
- 2026-10-09
- Days held
- 17
- Strike
- +0.9% OTM
- Premium
- $610.00
- Max loss
- The premium
- Break-even
- +4.8%
- Implied vol
- 49.8%
- Share of account
- 69%
- Spread paid
- 1.6%
What has to happen
Break-even is $161.05, up 4.8% from the $153.63 the stock traded at when the order went in. The dead zone between the market and the strike is only 0.9% wide — the narrowest of the three exercises, against 1.7% on NVDA and 13.5% on HOOD. The cost of that closeness is time: seventeen days, and theta is $20 a day, about 3.3% of the premium every day it sits.
| Underlying | Move | Contract | Return |
|---|---|---|---|
| flat | 0.0% | −$610 | −100% |
| at the strike | +0.9% | −$610 | −100% |
| break-even | +4.8% | $0 | 0% |
| — | +7.0% | +$329 | +54% |
| — | +10.0% | +$789 | +129% |
| — | +15.0% | +$1,558 | +255% |
The exit is planned for the move, not the date
That table is what the contract is worth at expiry, and it is not the plan. The plan is to sell into a move of about two per cent in the stock, which at a delta near 0.50 marks the contract around +20% within a day or two. Held to expiry, that same two per cent move is worth minus 72%. The two numbers describe the same position and they are not in conflict — one is a mark and one is a settlement, and this account has now twice exited on the mark.
HOOD closed on day eleven of thirty-six. NVDA closed the day after it was opened. The line that came out of the first one holds: the exit comes on the move, not on the date. Publishing the expiry table anyway is the point — it is the honest description of what was bought, and the loss column is the base case.
What is missing
There is no stop on this position, and there is none on either of the two before it. The perpetual-futures book run alongside this account has take-profits and a stop bracket on every leg, measured off 201 days of candles. The options have targets and no downside rule at all. That asymmetry is not a decision that was made; it is one that has not been made yet, and it is written here so it cannot be quietly skipped a fourth time.
Sixty-nine per cent of the account sits in this one premium. That is lower than the 95% NVDA carried, and it is still concentration rather than diversification.
Placed and filled at $6.10. One contract, $610.00 plus four cents of fees, sent through the agent at 14:47 ET on a $6.10 limit set deliberately at the ask rather than at the mark, so it would print instead of rest. It filled in about a quarter of a second. At the fill: delta 0.494, theta −$20.09 a day, implied volatility 49.8%, break-even $161.05. The broker’s probability of any profit at expiry: 31.6%.
The quote moved while the order was being reviewed — the ask was $6.15 when the chain was first read and $6.10 by the time the review came back. It moved in the account’s favour this time. It moved against it on both previous exercises, and that is recorded on their pages.
An alert is set at $156.70. Outcome recorded here either way.
Sold to close at $4.75. Held ten of the seventeen days. Realised −$135 on $610.04, −22.1%. The first loss on this account.
It is also the first exercise whose strike was reached. SPCX traded $156.50 the morning it was sold — above the $155 strike and 1.9% above the entry — and the position still lost. Between the fill and that morning the stock fell to $145.36 (28 September, −5.4%) and spent a week below where it was bought; the contract closed on 1 October at $1.23, down 80%. The two per cent move the plan was waiting for did come. It came on day ten instead of day one or two, and by then seven days of theta had been paid, so the same move that was meant to mark the contract up 20% marked it down 22%.
The direction was right and the clock was wrong. The two exercises that paid had 36 and 32 days to run; this one had 17. The operator’s conclusion, verbatim: “Longer time horizon to spread out the theta burn.”
There was still no stop, and this time that helped. Any stop set in the first week would have fired near the low, with the contract at a fraction of its cost, and booked a larger loss than holding did. That is one observation, not a policy.
The exit was a target conceded before it was sent: $5.00 was proposed, $4.75 was placed above a market of $4.45–$4.50, and it filled sixty-five seconds later. Across three exercises the account has realised +$255 — +$235, +$155, −$135. The fourth exercise, on a longer clock, is written down here.