The barrel steadied and the margin to refine it stayed wide. That is the whole thesis in one line: it is the crack spread, not the crude price, that the war keeps pricing, and the refiner pairs keep proving it.

Dashboard — July 16, 2026 Settled Close

All dashboard levels are July 16, 2026 New York settled closes from a single source (yfinance batch OHLC); the Yahoo v8 chart API did not confirm for this historical date, so this edition is single-source by exception. The distillate crack spread is front-month ULSD times forty-two, less front-month WTI, in dollars per barrel. Day changes are measured against the July 15 settled close.

I. The Margin Makes a New High While the Barrel Slips

Here is the single most important number in today’s tape, and it is not an index level. The distillate crack spread widened $4.11 to $90.34 a barrel — a fresh high for the escalation — even as West Texas Intermediate slipped $0.65 to $78.95. That divergence is the entire thesis of this book stated in one line: it is the margin to refine the barrel, not the price of the barrel, that the war keeps pricing, and on a day the crude drifted lower the margin went to a new high. A year ago a crack spread of $90 would have read as a data error. It is now the number that carries the refiner pairs.

The equity tape, by contrast, had its first genuinely soft session of the week. The Standard & Poor’s 500 fell 0.51 percent to 7,533.77 and the Nasdaq Composite fell 1.47 percent to 25,882, giving back the semiconductor firmness of the prior two days. The CBOE Volatility Index rose 1.06 to 16.73. This was not a break — it was a technology-led pullback in a market that had gone three sessions without one, and it fell on a day the book was positioned to absorb through its short-technology legs.

II. Rates Hold, and the Register Waits

The curve did little: the ten-year par yield rose two basis points to 4.57 percent, and gold fell $58 to $3,986 as the dollar held firm. A modestly higher long yield alongside falling gold and a softer equity tape is close to the vigilante posture — the long end declining to rally on an equity pullback — but the move is too small to lean on. The honest read is that the register is intact and dormant. It was built for the days when a supply shock hits and the long bond sells instead of bidding; today was a technology pullback with a firm crude complex and a two-basis-point drift, which is not that test.

What is not dormant is the crack spread, and that is where the attention belongs. If the margin holds near $90 into the back half of the month, the refiner-versus-megacap expression stops being a catalyst trade and starts being a regime. That is the question P40 and P41 are quietly posing.

III. The Pair Book

Twenty-six active pairs per the v6 canonical book of record. Day is the July 15 to July 16, 2026 settled move; QTD is measured from the June 30 settled close per the period-baseline rule; pairs marked † (P39, P40, P41) were opened during the quarter and their QTD is inception-to-date. Exact inception share counts throughout; zero notional approximation. Ranked by unrealized profit and loss.

The book lost −$15,420.08 on the session, its worst day of the week, and the loss was broad rather than concentrated — the mirror image of a market where the semiconductor complex pulled back and dragged the long technology legs of the pairs with it. The single largest drag was P17 (long Sibanye-Stillwater against short Honda), off $8,080 as both legs fell and the short could not keep pace; P18 (long GitLab against short Atlassian) cost another $6,765, and P24 (long Alphabet against short JetBlue) lost $5,706 as the low-cost carrier bounced. Against the tide, the refiner and integrated-energy pairs did exactly what the crack-spread high implies they should: P36 (long Chevron against short Broadcom) added $5,335, P41 (long Marathon against short Tesla) added $3,311, and P8 (long Apollo against short the global-banks fund) added $3,042 as the bank complex sold off. Quarter-to-date the book stands at +$76,398.43; active unrealized is −$115,465.97 against the frozen realized register of +$487,933.25, for an inception-to-date result of $372,467.28. A soft day for the open book; the register holds the line.

IV. Harvest Shadow — the Edge Stands at $334,836.65

The standing counterfactual: the ten pairs closed at the June 30 settled-close sweep, marked forward as though the harvest had never happened, exits frozen at their actual June 30 prints. Through the July 16 close the harvest decision is +$334,836.65 better than holding would have been, another weekly high — and today the mechanism is worth naming precisely. The edge widens on exactly the days the active book struggles, because the harvest was, in structural terms, a decision to be short the technology and memory complex at the top of a move. When that complex pulls back, as it did today, the counterfactual hold path deteriorates faster than the book does, and the gap between the frozen realized figure and the marked-forward hold value grows. The decay remains concentrated in Corning, which anchored three of the eight profit closures; P1 alone accounts for over $100,000 of the edge.

V. What to Watch

Whether the distillate margin holds into the weekend. Whether the refiner-versus-megacap expression is building into a regime rather than a one-week catalyst trade. And whether the metals-and-mining pairs stabilize after a soft stretch.