Brent's premium over WTI is $2.41 a barrel. On the matched spot marks of July 29 it was 5.84, and almost all of the compression is the American grade rising: WTI sits 3.48% above its $83.43 July 29 mark, Brent 0.59% below its $89.27 one.
By Monday morning in London, WTI marked $86.33 against $83.40 twenty-four hours earlier, up 3.51%. Brent went to $88.74 from $88.10 over the identical window, up 0.73%. That is 2.29 points out of the spread in a single session, and the spread had already been sitting at 4.70 the day before, inside the 5.00 level the bullish crude case was built on.
The grade cast as the follower has done all the work.
The spread was the argument, not a detail
Convex's house view, set on July 30, took oil to bullish on one piece of evidence. A Brent-led move had carried the spread through 5.00, and the desk read that as a waterborne supply-risk premium tied to the July 7 tanker strike near the Strait of Hormuz, an event the newsroom scored 7 and classed a major oil-shock signal. The expression was written as narrowly as the reasoning: the Brent leg and the spread, with flat-price WTI length ruled out.
Both of those legs have gone nowhere or backwards in a month when the ruled-out expression gained 3.5%.
The conditional was explicit. Brent extends above $89.27 and WTI follows toward a sustained $90, provided the spread held above 5.00 and Brent held its $81.62 July 21 marker, with no confirmed easing of Hormuz shipping risk. The first condition failed at 2.41. Brent has not extended; it sits 0.59% under $89.27. WTI at $86.33 is $3.67 short of the $90 it was meant to be dragged toward, and it arrived from the other direction.
None of that trips the written exit, and the distinction matters. Invalidation needs Brent below $81.62 with the spread back inside 5.00, or WTI below $70 for three sessions or more. The spread leg is met; its partner is not, because Brent at $88.74 is 8.7% above $81.62, and nothing on the tape is near $70. The position survives on its own terms while the reasoning underneath it has come apart.
Nobody rang a bell on the risk premium
This news feed stops on July 7, which limits what can be claimed here. Its most recent Middle East items are the tanker strike and the Khamenei report, the second scored 6 and classed a de-escalation signal, and the July 30 view already treated both as background rather than catalysts. Nothing here shows that Hormuz risk eased, and nothing here would show it if it had.
Price is the other witness, and it speaks in one place: the gap between the waterborne barrel and the landlocked one, which is where a seaborne premium lives if it exists at all. That gap has more than halved.
Volatility says something similar without raising its voice. VIX marked 15.22 on Monday, up 5.47% from 14.43, which reads as a wobble until it is set beside the 20.66 print of July 29 and the 18.65 marker of July 21. Equity vol is 26.3% below the first and 18.4% below the second. Europe is as quiet, VSTOXX at 16.3131 against 16.1317 on August 28, up 1.12% across three sessions. Crude rallying into subdued equity vol, with no fresh chokepoint headline, looks like a barrel story rather than a war-risk story.
One caution on that. The freshest credit spread available is high-yield OAS at 2.84% on July 28, so this says nothing about what credit has done since, and vol and credit already came apart once this summer, when the conditions index eased through July 24 while spreads widened to July 28.
The price index does not care which grade rallied
Whatever is lifting crude, a consumer price index takes the barrel and not the explanation.
June's CPI index fell 0.42% to 332.568, the outright decline in the level that pushed the regime call toward reflation. July printed 332.813, up 0.07% on the month, which leaves the index 0.35% below May's 333.979. The fall stopped and the level has barely turned. It arrived with crude far above where June's disinflation was recorded: $70.56 on the FRED series on June 30, $84.25 on July 27, and an $86.33 spot mark now, roughly 22% higher across two different sources. A barrel that much dearer has so far bought seven hundredths of a percent on the index.
The long-breakeven position that arithmetic supported cannot be scored at all. No fresh 10-year breakeven or real-yield observation has printed since July, so inflation compensation stands at 2.26% on July 29 and the 10-year real yield at 2.41% on July 28. The central variable of the top position in the book has gone unobserved for a month.
Nominal yields did move. The 10-year was 4.67% on August 27 against 4.61% on July 28, and the 2-year 4.20% against 4.26%. On that pair the bearish call on nominals is paying: 4.67% is above the 4.50% floor its thesis named, short of the 4.85% target, and nowhere near the 4.25% invalidation.
Decomposing the move is where the data stop cooperating. The published 10-year-minus-2-year spread reads 0.39 on August 28 while the August 27 legs imply 0.47, an 8bp disagreement across one session, and the same two sources disagreed by 10bp the other way in July, when the legs implied 0.35 on July 28 and the published series printed 0.45 on July 29. Read the legs and the curve steepened from both ends, which is the reflation shape. The published series says it flattened 6bp over the month, which is not. Neither is strong enough to establish that the front end has begun pricing a cut, or to rule it out.
The July 30 view named its resolution points: the Q2 GDP advance and the PCE releases that day, PPI on July 31, the conditions index on August 5, payrolls on August 7, CPI on August 12. Every one of those dates has passed. Not one of those prints appears in the data available here, so the growth question that was supposed to be settled a month ago is, on this evidence, still open.
The strongest case against
Gold is the objection. It marked $4,505.4 on Monday, down 0.54% on the day and up 9.11% from $4,129.2 on July 29. Equity vol fell 26.3% across that same stretch and the metal gained nearly a tenth anyway. If the fear premium has drained out of energy, something somewhere is still paying up for a hedge, and the July 30 view singled gold out as the one asset in the book that had improved on its own evidence rather than on a forecast.
Sample size is the second objection. Two observations support a net change and nothing beyond it, and one session of spread compression is not a trend. No inventory or freight series here can say whether Cushing or shipping economics is doing the work. Spreads that close from the wrong end can reopen the same way.
The demand read does not obviously explain a stalled Brent either. Copper marked 6.68 against 6.64 on August 29, up 0.60%, and the DAX was 26,370.92 against 26,569.99 over that same longer window, down 0.75%; European government yields edged down to 2.6476 from 2.6608. Small moves, wider windows than the crude comparison, and no S&P observation is available here, so the equity leg of the reflation call cannot be scored. The dollar is worse: the only fresh mark is the broad index at 118.0628 on August 21, which shares no scale with the 100.806 narrow marker the July view used, and reconciling the two would mean inventing a number.
What none of it supports is the story the bullish call was told with. That upgrade was earned by a spread through 5.00 and a chokepoint premium, and it was sized on the mechanism rather than on the direction. The spread is 2.41, the premium is invisible in vol, and crude is higher anyway, led by the barrel furthest from any tanker. Right on price and wrong on the reason is the version of right that does not repeat.
Explore these indicators together: Chart Brent Crude Oil, WTI Crude Oil, and 3 more on the Indicators Dashboard