A margin call does not read a prospectus. It sells whatever is liquid, regardless of what that position was actually for. Tuesday's settled tape has the fingerprints of exactly that kind of selling — Alcoa and Cleveland-Cliffs down double digits on no aluminum or iron-ore news, a South African platinum miner down alongside Corning glass, a jet-lessor-adjacent airline short moving harder than the airline itself. None of that is a demand story. It is a liquidity story wearing a demand story's clothes.

I. The Tape at the Settled Close

Two-source spot-checked (yfinance batch + Yahoo v8 chart) on the nine most volatile legs; broad indices and remaining book legs single-source (yfinance) pending full reconciliation. VIX actually fell on the session despite the semiconductor damage — the stress was narrow, not systemic; see Section III.

II. Tranche 11 at the Close — Dell Did More Work Than Anything Else in the Book

Corning finished down 12.1%, worse than the semiconductor index itself and worse than every name in the book except the small-industrial complex covered in Section III. But Dell fell 8.15% on the same session — and Dell is a short leg, not a long one. That single move turned Tranche 11's day around.

Tranche 11 finished the day essentially flat — a $490 gain against a book that lost nearly $60,000 elsewhere. That is not luck; it is the tranche doing what a five-pair, GNRC-concentrated structure is supposed to do when one short leg breaks hard in the desk's favor. P46 alone offset most of P42's damage. The tranche's real story this week is no longer Corning — it is that GNRC, carried on three separate pairs, has become the fulcrum for the entire tranche's daily variance.

III. Where the Real Damage Landed — and Why It Wasn't the Chips

The single worst pair in the book Tuesday was not a semiconductor pair. It was P16 (Alcoa long / Boeing short), down $4,208.76 on the day and $38,426.64 since inception — the book's largest loser, period. Alcoa has no meaningful semiconductor exposure. Neither does P29 (Century Aluminum / Boeing), down $5,735.54 on the day. Neither does P17 (Sibanye-Stillwater / Honda), down $6,572.10, or P24 (Alphabet / JetBlue), down $8,749.40, or P25 (Cleveland-Cliffs / Nucor), down $10,451.10 — the single worst day-change in the book.

Five of nine tranches finished the session net negative, and the pattern across the worst of them is not thematic, it's structural: small-capitalization, thinly-traded, or high-beta names that get sold first and asked questions later once a margin call goes out against a semiconductor-heavy book elsewhere. Cleveland-Cliffs and Alcoa are exactly that kind of name — liquid enough to raise cash fast, small enough that a forced seller moves the price disproportionately. That is the mechanical reading of Cramer's quote in the observances section above: Tuesday's selling did not discriminate between the companies that caused the stress and the companies that merely happened to be liquid.

IV. The June 30 Harvest, Marked at the Close

Combined, the two harvests preserve $226,912.37 of value relative to holding — essentially unchanged from the intraday reading, because both names finished the settled session close to where they traded mid-afternoon. The pattern that has held all week continues to hold: capital taken off the table in June is now materially outperforming the closely related structures still on the book.

V. The Pair Book — Operation Epic Fury at the July 28 Settled Close

Tranche attribution

Open flags carried forward, plus one new

1. New: the small-industrial short leg is now the book's single largest source of unforced risk. P16, P25, P29, P17, and P24 together account for over $124,000 of the book's active drawdown, on names with no direct semiconductor exposure. This is a liquidity-driven, not thesis-driven, loss cluster and deserves its own review before Wednesday's FOMC/earnings session compounds it.
2. GNRC now the fulcrum of Tranche 11, carried across P44, P45, and P46 (1,507 combined shares). Tuesday showed both sides of that concentration: it barely moved on its own account, while its short-leg counterparties (DELL, NVDA, AMZN) did nearly all the work.
3. Corning is now the worst-performing long leg in the book on two separate pairs (P21, P42), down over $48,000 combined since inception. No name carries more concentrated single-stock risk.
4. Jet crack spread now actively marked per the July 27 protocol; not re-pulled for this settled-close build. Carry forward for the next edition.
5. Full 62-ticker two-source reconciliation is outstanding. This build spot-checked the nine most volatile legs only. Complete before this file reaches Drupal.

The Synthesis

Tuesday cost the book nearly $60,000, and less than a fifth of that came from the semiconductor names that started the story Monday. The rest came from small, liquid, thematically unrelated positions that happened to be easy to sell when something bigger needed selling. That is a different risk than the one Monday's edition described, and it is worth naming plainly: a demand-verification problem in one sector became a liquidity event across five. Wednesday brings the FOMC and two megacap earnings reports in the same session — the highest-density catalyst day of the month, arriving directly on top of a book already absorbed a rough one.

The June 30 harvest remains the clearest evidence in this book that discipline exercised before a test is worth more than conviction exercised during one. Tuesday made that trade roughly $227,000 better than holding, for the second straight session.