Back from a few days off. I planned a comprehensive note but the oil section got long, so I'll write in two installments (the second on broader risk assets, positioning, and funding). Many readers are frustrated with the oil notes (subscribers drifting away), but oil and related trades have turned into the most convex opportunity I have seen in my 25yr+ career. Can you blame me for not caring about semis or Korea or AI or SpaceX?
A few things I believe: Crude oil is set up for a dramatic upside move due to a confluence of fundamentals, positioning, and sentiment. The market's complacency stems from a temporary, unsustainable condition where historic supply/demand deficits are being bridged by visible and shadow inventory draws that should turn into outright shortages in the coming months, including in the US. There is a +15-20% day in oil not far off. My bullish oil thesis in January pre-dated the Iran drama and had nothing to do with Iran.
The Fundamentals (Inventory Draws are Visible Yet Ignored) The fundamentals around inventory draws are visible yet ignored (charts from Morgan Stanley). Key points: US commercial inventories bottom around 370mmbbls (+/-) due to linefill (pipelines can't be empty), minimum refinery stocks, and tank deadstock (June Goh at Sparta). US SPR discharges did the heavy drawing but the pace eases from May's ~1.4mmbpd peak to ~1.0-1.2 in June-July and ~0.8 in August (barring another release, likely announced in June) — the DOE warned (T4 RFP) that "as SPR inventory is depleted, rates into the terminals will degrade" (salt-cavern water-displacement pumping slows at lower fill). Cushing (WTI delivery point) tank minimums ~17-20mmbbls; checks with HFI Research and refiners reveal a growing realization that Cushing should see tank bottoms by June 30th — shocking this isn't in the market. US gasoline operational minimums ~200mmbbls; distillates ~100-110mmbbls. MS Exhibit 41: bears point to commercial (ex-SPR) stocks at the same level as last year — but prices are too low with zero US demand destruction, and draws will accelerate from 1) rising refinery runs, 2) lower imports, 3) rising exports. Largest weekly aggregate crude draw in history two weeks ago; further historic draws in June.
Demand Destruction Claims of several mbpd of global demand destruction are modest at best (some Asia WFH, some Europe/Asia flight cancellations). Inventory destocking is masquerading as demand destruction — observers see declining imports and assume demand is falling (global imports actually bottomed in May and are rebounding; China's declining imports are backward-looking and only part of the picture). Energy Aspects: Kayrros connected-vehicle data shows limited US demand response to higher prices; some institutional research claims US gasoline consumption fell ~-3% y/y in March from credit-card spending, but petrol-station transaction counts remained firm; spending data must be deflated by price (consumers seek cheaper stations/ lower-grade fuel when prices spike, overstating the implied drop). Behavioral note: more customers paying cash inside (to avoid the >3% card fee), which shows up as a card-spending decline not reflecting real demand decline.
The Sequence Matters US gasoline is the most immediate problem. This week's DOE may show a small seasonal gasoline build (Memorial Day wholesaler delivery timing). From the June 10th DOE report gasoline draws rise again; approaching operational minimums without demand destruction, it's very possible to see outright fueling lines/shortages in PADD1 (East Coast) by end-June. Gasoline is ~two weeks from a pricing problem; normally he'd buy $3.50/$4 call spreads on July RBOB (late-June expiry) for 3c, but the problem may manifest fast enough to elicit a Trump regulatory/political response. Per "Shrub's Razor" (the funniest/most absurd outcome is the most likely), he expects "the most business-friendly administration in history" to reenact gasoline export controls (removed by Obama), à la Nixon's 1970s price controls — starting with gasoline, probably partial ("just the tip"), aligned with Trump's politics (e.g. allies Argentina/Japan get gasoline; Europe/NATO don't). The damage hits the whole futures complex (RBOB, HO, even WTI), so he might not get paid on gasoline trades. "Oil traders are the new vigilantes of the 2020s." Crude becomes an obvious US inventory problem only weeks behind gasoline; by mid-July US exports must be priced out.
Bringing It All Home Historic draws in US crude and products are inevitable — baked in. Almost nobody does this work (energy talent is non-existent after 15 years as a 3-4% index-weight backwater; follow HFI Research). Triangulate forward inventories via producer data, announced refinery runs, ships signaling to the Gulf of America, etc. Bottom line: the largest draws in US crude stockpiles in the history of the data over the next 6-8 weeks. "Inevitable" and "imminent" aren't always the same — the frustration window. (The Big Short analogy: Vinny/Porter vs Greg Lippman — the math is baked in but marked against you.) We've crossed the Event Horizon regardless of whether the Iran war ends tomorrow; Twitter/Truth Social is noise. Some "Tom Hanks has Covid" moment will touch off non-linear "oh sh*t" awareness. Now inevitable and imminent are the same — weeks/months has become days/weeks.
Positioning & Sentiment VaR shock and degrossing by physical traders (blew through risk limits in March/April volatility) now trade smaller or flip futures into options due to Trump-tweet stop-outs → prompt futures increasingly illiquid. The futures→options substitution sets up a fragile profile where market makers are short calls; with oil vol now lower, they could be forced to chase prices higher and buy futures in an illiquid market to hedge rising delta and vanna (like the semiconductor delta/vanna squeeze). Speculative length largely wrung out — large+small speculators' net-long WTI is not far above the 2023-24 lows; it's the options exposure keeping the net long elevated (dangerous if oil pops on dealer hedging). Retail Reddit crowd is short via SCO (2x inverse WTI). USO (WTI front-month ETF): $1.7bn assets, down from $2.8bn on Mar 12 (-$1.1bn / -40%), -$900mn from outflows since Mar 31, +87% YTD, -15% from its May-19 high. BNO (Brent): -16% since its May-4 high, -25% of assets ($1bn→$765mn), +78% YTD. SCO ("2x short oil"): assets exploded from $100mn at war onset to ~$1.5bn (almost as big as USO!), ~$700mn inflows since Apr 1, $1.6bn inflows since the Mar-1 war start — yet -17% since Mar 31, -53% since the war, -65% YTD; retail keeps plowing in. CFTC CoT: producer/merchant (commercials) much more net long since the war (refiners buying? producers reducing hedges?); managed money mostly flat; "other reportable" (prop shops, Chinese traders, family offices, "Cloudbears") cut net long >30% since early April; swap dealers imploded from ~350k to 660k net short (Wall Street desks hedging long OTC exposure for unknown clients). Bottom line: producer/merchant commercials are buying like crazy while Wall Street synthetics sell to them (see Exxon's Bernstein-conference comments; Chevron CEO on Bloomberg). Sentiment is poor — many traders have thrown in the towel after ill-timed call-option trades since March.
Iran Doesn't Actually Matter As in January, no Iran in the thesis. No grand bargain likely — Iran needs to make this hurt; having found economic leverage they won't relinquish it; their optimal strategy (constrain the oil trade) doesn't change regardless of others' actions (a democratic leader facing midterms amid $8+ gasoline is in crisis). All noise. US commercial inventories head below 400mmbbls even if the Strait opens tomorrow; then US exports must be priced out; refiners build stocks to capture $50 cracks rather than cut runs. He wishes Trump/Iran would sign a ceasefire so crude sells off $5 and reverses on a news-failure — then only the math is left (boats per day, time to drain Gulf stocks, restart wells, ramp production). Every physical trader (esp. Asia) is waiting for a peace-deal/TACO dip to buy — and "when everyone is waiting for something, the market won't let them in." His largest position by a wide margin is now BNO (Brent ETF, for roll yield), primarily calls across Jul-26→Jan-27 strikes, bought aggressively as recently as Friday.
"Please do your own work." — Paulo aka Cloudbear