Brent printed above 100 dollars for the first time in two months, and by Thursday every desk I watch had quietly filed the same verdict. This was a war premium, not a new floor. The outright looked like a supply shock. Everything underneath it, the curve, the options tape, the ETF flows, the refiner equities, was voting that the shock was temporary.
The tape that everyone saw
The outright is where the fear lived. WTI rose 5.2 percent on the week to settle at 89.31, after tagging 92.19 on Wednesday. Brent added 6.3 percent to close at 96.78, after breaking 100.69 midweek, its first triple-digit print since May. I have sat through enough of these to know the outright is the part of the market that reacts, not the part that thinks, and the reaction here had a genuine cause. The Iran conflict widened into a two-chokepoint problem in a few sessions. Houthi strikes reopened a Red Sea front while the Strait of Hormuz stayed under threat, and a CPC pipeline halt forced temporary Kazakh output cuts on top. Three supply-disruption vectors firing at once is not a headline, it is a physical event.
On Monday, Brent closed above its 50-day moving average for the first time in two months, and an Invesco cross-commodity ETF logged a record daily inflow north of 660 million dollars. That number is a tell about who was buying and how, and I will come back to it. By Thursday the Brent print had faded back to 96.78. Crude kept moving through Middle East routes despite the shooting, and Washington signaled the US-Iran channel was still open. The outright had done its job. It scared people. Now I wanted to know what the instruments that actually price risk were saying, and they were saying something very different.
The curve is where I look first
Forget the flat price for a moment. The signal of the week was in the time spreads, and it was violent. The WTI prompt spread, M1 minus M2, went from a near-flat plus 0.57 on Monday to plus 3.19 Tuesday, plus 5.18 Wednesday, then eased to plus 4.16 Thursday. The M1-M7 ran plus 9.22, plus 12.01, plus 16.57, plus 13.84. Brent moved in lockstep, its prompt from plus 2.49 to plus 6.43 at the Wednesday peak, and its M1-M7 hitting plus 18.79, the widest in more than two months.
A nine-fold move in the WTI prompt spread over three sessions is the physical market screaming that barrels are wanted now, not later. For context on how fast the regime flipped, that same prompt spread had briefly gone into contango as recently as July 2. Three weeks later it was in five-dollar backwardation. When I see steep backwardation across every tenor, I read genuine near-term scarcity, not a financial squeeze. But the part that told me the most came Thursday, when the WTI prompt eased to 4.16 and Brent M1-M7 pulled back to 15.50. The curve is the market’s real-time referendum on whether a disruption is permanent or temporary, and that partial unwind, arriving before any ceasefire and while the outright was still elevated, was the first ballot cast for temporary. If Hormuz stays open and the US-Iran channel holds, that backwardation compresses fast, and the geopolitical premium in the flat price goes with it.
The options desk was selling the spike
The volatility surface agreed with the curve, and the composition of the flow is what convinced me. Implied vol on Brent and WTI hit its highest since May 22 on Wednesday and Thursday as Brent broke 100, with futures volume and open interest climbing together. That much is mechanical. Prices move, vol lifts. What matters is what people did with the vol, and what they did was defensive.
Brent options set a record volume on Thursday, and puts featured heavily, including October strikes, a structural hedge against the geopolitical premium unwinding hard. Narrow put spreads down in the low 70s had been bought in size as early as the prior week, and on Tuesday those put spreads were rolling up as the market climbed toward 95. I have put on that exact trade myself more times than I can count. It is the classic sell-the-spike hedge, large holders locking in gains and paying for downside while the rally is still loud. WTI’s second-month 25-delta skew had reached its most bullish since April the week before, but as the 100 level broke the flow turned defensive. Nobody I watched was chasing calls into triple digits.
The ETF tape said the same thing, and here is where that 660 million dollar inflow comes back. The USO oil fund posted its biggest daily outflow since early April on Wednesday, 277 million dollars, right at the highs. ETF longs used the 100 print as an exit, not an entry. CFTC data showed net bullish NYMEX WTI bets at a three-week high through July 21, and Brent net length up 7,767 contracts, but the positioning was carried with protection underneath it. Long the barrel, hedged the downside. That is not conviction we are going higher. That is a desk that wants a tail without betting the ranch on it.
Products, and the one bid that actually held
The refined barrel is where the premium was most defensible, and it separated cleanly into the part that faded and the part that did not. Heating oil ran from 412.66 cents on Monday to 434.16 on Wednesday before easing to 418.06. RBOB gasoline spiked to 349.64 on Wednesday and then handed back essentially the entire war premium in a single session, closing at 339.59 on Thursday once Hormuz flows proved resilient. Gasoline is a headline follower. It gave the premium straight back.
Diesel is the standout, and it is structural, which is why it is the one part of the complex I would not fade here. The ICE gasoil crack hovered near 65 dollars, and the 3-2-1 crack reached as high as 70 dollars, a record, driven by Russia’s diesel export ban and by Hormuz disruptions cutting Middle Eastern product flows. Repsol said it expects refining margins to stay healthy into 2027. Around it, the evidence stacked up the same direction:
- US retail gasoline climbed back above 4 dollars a gallon.
- Cathay Pacific announced passenger fuel surcharges from August 1.
- US refined product exports set a record the prior week on propane and diesel shipments.
- Romania’s Petromidia refinery flagged a possible 10 to 15 percent output cut in August if Kazakh crude deliveries stay halted.
Unlike the gasoline spike, the diesel bid did not fade. Refiners cannot easily substitute away from Middle Eastern grades, and a Russian export ban is not a headline that reverses on a ceasefire tweet. When the rest of the board is pricing a temporary shock and one product is pricing a durable one, the divergence is the information. Diesel is telling me the tightness there has a different half-life than the war premium in crude.
Natural gas told the mirror story
While crude ripped into backwardation, US natural gas did the exact opposite, and the contrast is clean enough to be useful. Henry Hub settled at 2.871 on Thursday, down 1.37 percent on the week, a fifth straight weekly decline and the largest five-week drop since April. The curve mirrored crude inverted. The prompt spread flipped from a small premium of plus 0.027 on Tuesday to a discount of minus 0.017 by Thursday, and the twelve-month spread deepened to minus 0.385. That is persistent contango, ample domestic supply, no near-term scarcity, the photographic negative of crude’s violent backwardation.
Tuesday’s 2.1 percent bounce on hotter weather and higher LNG feedgas was gone by Thursday as production rose, the forecast cooled, and gas got dragged lower inside the commodity baskets sold in response to the oil shock. Hedge funds pushed net bearish Henry Hub bets to a two-year high, net short 50,303 contracts, with short-only positions at 518,422, the most in over two years. That is a crowded short, and crowded shorts are the kindling for a squeeze if LNG demand or a weather shock arrives. Europe went the other way entirely. TTF rose 11 percent on the week and more than 45 percent in July on the Red Sea LNG disruption, with storage at 55 percent against a 70 percent seasonal norm. The Henry Hub and TTF gap resolves one of two ways, through more US LNG heading east, or through a shipping escalation that hurts both. I would not want to be short the wrong one of those.
The equities named the verdict
If you want the cleanest read on what the market actually believes, and not what it is feeling, look at the energy stocks. They diverged from crude in three different directions this week, and each direction is a data point.
Refiners sold off into record crack spreads, and that is the single most important signal of the week. VLO, PSX and MPC had hit record highs on July 20 precisely because the 3-2-1 crack was at a record. Then they reversed hard: PBF minus 6.69 percent, DK minus 5.94, VLO minus 3.91, DINO minus 3.70, MPC minus 3.29, PSX minus 2.59, all of them 8 to 12 points below WTI’s move on the week. The equity market is front-running margin compression. If crude holds at 90 to 100, rising input costs erode the very crack advantage that drove the rally. US refiners have run at or above 95 percent utilization for nearly two months, which is full capacity, which turns any equipment breakdown into a margin event, and roughly 10 percent of global refining capacity is already offline. The options market votes the same way, with PBF implied vol above 70 percent, VLO in the 69th to 88th percentile, and a 5.6 percent implied move priced for PBF’s July 30 earnings. The sector is repricing itself from beneficiary of the oil spike to victim of it.
The E&Ps captured only a fraction of the crude move. XOM led at plus 3.45 percent, helped by its integrated model, then FANG plus 2.44, COP plus 2.35, DVN plus 2.13, EOG plus 2.06, CVX plus 1.95, APA plus 1.76, OXY plus 1.42. Call it 27 to 66 percent of the crude move, a beta discount of roughly three points. Some of that is the broad tape, with the S&P down 1.28 percent and the Nasdaq down 3.52 on the week. The rest is a geopolitical risk discount, the market pricing that a ceasefire or a Hormuz reopening could unwind the crude spike and leave these names stranded at elevated valuations. The record ETF inflow says the same thing from the other side. Investors wanted direct commodity exposure this week, not equity proxies, because a proxy carries the risk that the relationship breaks.
Tankers are the paradox, and they are the tell I trust most. Hormuz disruptions and Red Sea attacks should be a windfall through longer routes and higher day rates, yet the sector was flat to negative: FRO plus 2.18, INSW plus 2.00, ASC plus 1.96, TNK plus 0.33, DHT minus 0.11, STNG minus 0.21, NAT minus 0.31, TK minus 1.15. The trapped-oil problem explains it. Crude in floating storage rose 31 percent in the week to July 17 to 98 million barrels, with Middle East floating storage up 144 percent, and a tanker sitting as a stationary warehouse does not earn a spot voyage rate. Add the transit risk, two UAE-linked vessels hit by projectiles crossing Hormuz in mid-July, and the operational uncertainty reads as a negative. DHT put its Q2 fleet TCE near 126,700 dollars a day with spot VLCCs around 162,600, and the stock barely moved, which tells me the strong rates are already in the price. Evercore cut its Scorpio target from 98 to 94 on Hormuz resumption risk. The muted tanker response is, to me, the clearest single statement in the complex that the market does not believe this disruption is structural. If Hormuz were permanently impaired, day rates would surge and these stocks would re-rate. They did neither.
Notes from the desk
Put the ballots side by side and they read as one sentence. The time spread ripped and then unwound before any ceasefire. The options desk bought puts into the rally. ETF longs sold the 100 print. Refiners sold off into a record crack because input costs threaten the margin. Tankers stayed flat through a live supply shock. The only place the premium held cleanly was diesel, where the tightness is structural and the Russian ban does not reverse on a headline. Every market that mattered voted the spike was temporary, and they voted it while the flat price was still frightening people. That is the part worth remembering. The instruments that price risk turned before the instruments that broadcast it.
The way I map these episodes, a story like this one is not an oil call, it is a load-bearing column test. The question is never simply where Brent settles, it is whether the disruption is a temporary stress on a structure that holds, or a crack in the structure itself. This week the complex answered temporary, in five separate voices, cleanly enough that I would trust the vote over my own gut reaction to a 100-dollar print. The framework behind that reading is the subject of my working paper, Convergent Faults, and of Beyond Gamma Exposure, on what a derivatives desk actually watches when a regime starts to move. Those works are the framework. This note is the week’s tape.
The coming week has one binary that sets the tone. PBF earnings land July 30, pre-market, the first major refiner print into the crack spike, and the options are pricing a 5.6 percent move for a reason. Above all of it sits the macro switch. Any credible Hormuz or Iran ceasefire signal reverses the whole board at once, crude lower, refiners higher on input relief, tankers lower on rate normalization, E&Ps lower on the revenue headwind. The vote this week was that the spike was temporary. The next few sessions of Hormuz traffic data will tell us whether the market called it right. My money is on the tape that turned first.
Sources: Bloomberg, Bloomberg First Word, Barron’s and Dow Jones, week of July 21 to 25, 2026. CFTC Money Managers commodity positions for July 21. Vortexa floating-storage data to July 17. Baker Hughes US rig count. Access the full desk at crossvol.com.