Not financial advice. For informational purposes only. Do your own research before making investment decisions.
Commodity Complex
Live YTD + Compound Crisis Thesis
17 front-month futures across Energy, Metals, Grains, Softs, and Livestock. Live YTD from Yahoo Finance. The dispersion between sectors is the trade.
Macro View
COMMODITY COMPLEX · COMPOUND CRISIS REGIME
Live YTD across 17 contracts · Molecules · atoms · calories — the whole hard-asset complex vs overvalued equities (per Jeff Currie, mid-Aug 2026) · Dispersion is still the trade
Live YTD Performance — Commodity Complex
Sorted by year-to-date return · Color-coded by sector · Hover for live price
Energy leads on Hormuz disruption. Refined products outperform crude — logistics, not just barrels. Softs at the bottom (cocoa -30%, coffee -22%) are positioned AGAINST the El Niño thesis — that's the asymmetric setup.
Yahoo Finance · As of Sep 14, 2026Hormuz disruption, supply premium, refined product squeeze
Geopolitical safe haven, real rate compression, CB buying
Fertilizer input shock (urea +50%), 2026 acreage uncertainty
Energy-cost pass-through vs. China demand wobble
Range-bound; feed costs partly offsetting tight supply
Cocoa/coffee unwinding 2024-25 supply shock as harvests improve
Compound Catalyst Thesis — El Niño + Hormuz
Why the softs at the bottom of the chart may be the most asymmetric long opportunity
Core thesis
The May setup — softs priced for weather recovery while a strong El Niño loomed — has largely played out: cocoa off its highs but still deficit-framed, sugar deficit-framed but now facing a genuine mill-mix headwind, all major sugar forecasters still in deficit. El Niño is no longer a probability, it's an active NOAA Advisory forecast to peak very strong Oct–Jan; the Sept 10 CPC update trimmed the projected peak strength only at the margin (still >90% very strong, 75% chance of a historic event). The remaining trade is duration and magnitude: the Oct–Dec West African pod-fill window (the season-start delay flagged by Ivory Coast's regulator got MORE specific this week, not less — weekly arrivals below 15k t in September), the Sep–Nov Brazilian coffee flowering (which keeps moving against the bull case as Brazil's crop forecast holds), and the 2027/28 crop-cycle effects that lag the peak. This week delivered the clearest test yet of the 'every leg moves together' assumption, and the answer was firmly no: the Sept 11 WASDE was bearish for grains, coffee extended its decline rather than reversing, cocoa cooled on price even as its deficit case held, and sugar picked up a genuine new headwind (Brazilian mills reverting toward sugar production as prices rose, per UNICA's 2H-August report — see /sugar). The frame has also widened and, this week, been genuinely tested: per Jeff Currie's molecules/atoms/calories triad, the 'atoms' (copper) leg took a tariff-headline shock, and even the core monetary-fragility bid (gold) fell for a third straight week as Fed rate-hike odds jumped and the latest Treasury buyback operation underwhelmed. None of this breaks the compound thesis, but it is real evidence that a hawkish-Fed, jawbone-heavy environment can push back against every leg of the hard-asset trade simultaneously — dispersion within the complex, not a clean one-way regime call.
Compound Exposure Mapping
Crops ranked by combined El Niño weather risk AND fertilizer cost transmission, vs. how much is already priced in
| Crop | El Niño Risk | YTD 2026 | ||
|---|---|---|---|---|
| CocoaPriority Fert: Moderate | Very High | Moderate | +1.4%● | Forecasters still diverging: Guan Chong's 300-400k tonne global 2026/27 DEFICIT call stands unchallenged, Hedgepoint confirmed its surplus cut to ~111k t (from 325k), StoneX's ~25k t figure not re-confirmed this week. Ivory Coast's season-delay warning got more specific: weekly arrivals expected below 15k t in September and 25k t in October, ~900k t at ports Oct-Dec vs. 1.1M t a year ago — real port-congestion risk ahead of EUDR; Ghana's ~6% farmgate price hike adds a fresh smuggling-incentive risk. Two-sided risk persists (ICE stocks at 2-yr high vs weak crop surveys) into the Oct-Dec pod-fill window |
| Coffee (Arabica)Priority Fert: Moderate | Very High | Moderate | -20.0%● | Trigger keeps moving the WRONG way: Brazil's 2026/27 crop forecast (Safras & Mercado: 75.65M bags, 49.95M Arabica) on strong Minas Gerais rains still stands, and no new rainfall data this week reversed it — the opposite of the disappointing-rainfall trigger this thesis needs. ICE-certified stocks are still near the 1999 low, but the live price/weather picture argues the setup keeps weakening, not strengthening, into the Sep-Oct flowering window |
| Sugar #11Priority Fert: High | Very High | High | +30.6%● | Mixed this week, not one-directional: India's monsoon deficit is unchanged/deepened (June-Aug -13.8%, August itself -16%, Sept outlook still below-normal — deficit case intact), but Brazil's UNICA 2H-August report showed mills reverting toward sugar from ethanol as prices rallied (mix back to ~46.4% from ~41.4% in July, essentially flat YoY) — a genuine new headwind to the deficit case. India's 1 MMT nil-duty import window (to Oct 31) remains open with no usage data; domestic Indian ex-mill prices have been falling on the quota + stock-limit combination even as global futures held firm (POSITION HELD, conviction trimmed — see /sugar) |
| Wheat (KC) Fert: Very High | Very High | Very High | +44.2%● | The de-escalation test kept failing to resolve either way: Turkey's Black Sea shipping-corridor proposal has stalled — Russia rejected it, and Moscow said just before Putin's 'peace deal remains possible' remark that there were no grounds for restoring the Black Sea grain deal. Prices are unwinding the war premium on rhetoric while the underlying disruption remains in place — the Sept 11 WASDE left wheat ending stocks unchanged, no new signal there. Sits on top of the still-unresolved 61-yr-low US winter crop |
| Rice Fert: Very High | Very High | Very High | Modest | Partially |
| Palm Oil Fert: Moderate | High | Moderate | Up modestly | Partially |
| Corn Fert: Very High | Mixed (US neutral) | Very High | +21.0%● | Sept 11 WASDE cut yield to 178.5 bu/acre (production -213M bu to 15.8B) — largely pre-priced by the trade, so the report didn't move the market further; ending stocks fell to 1.6B bu, season-average price raised to $4.80 |
| Soybeans Fert: Moderate | Negative (ARG benefits) | Moderate | +26.0%● | Sept 11 WASDE raised yield to 52.8 bu/acre and production to a near-record 4.5B bu — read as bearish ('no bullish surprises'), sold off hard despite ending stocks falling to 310M bu and price raised to $12.00; mostly priced |
Priced-in summary
Priority-1 rows (Cocoa / Coffee / Sugar) are the asymmetric setup — high El Niño risk + market positioned the WRONG direction. YTD values for live-tracked symbols pull from Yahoo; Rice / Palm Oil show the static May 12 labels.
The Dispersion Opportunity
The May dispersion trade (softs priced against El Niño) has largely converged, but the last two weeks are the clearest evidence yet that it doesn't converge uniformly — and this week widened the gap further. Cocoa cooled on price (~$5,985/t, Sept 12, off the ~$6,171-6,233 highs) even as its deficit case held or strengthened (Guan Chong's call unchallenged, Ivory Coast's delay warning more specific); coffee kept falling on fundamentals, not just price (Dec arabica to $2.84/lb, -16.4% over 30 days, as Brazil's raised crop forecast held); and sugar picked up a genuine new fundamental headwind (Brazilian mills reverting to sugar from ethanol per UNICA's 2H-August report) even as its other legs (India, El Niño) held. Wheat kept unwinding its Black Sea risk premium on stalled diplomacy (Turkey's corridor plan rejected by Russia) rather than any actual resolution. Even the broader hard-asset/monetary frame took a real hit this week: gold fell a third straight week as Fed rate-hike odds jumped and the first Treasury buyback operation under the new window underwhelmed — the clearest evidence yet that 'molecules, atoms, calories vs. overvalued equities' is not a one-way trade even when every individual physical-disruption thesis (Hormuz, refinery strikes, El Niño) remains intact. What remains unpriced: the 2027/28 lagged effects (the 2011 analogue: the 36¢ sugar ATH came AFTER consecutive damaged seasons) and Ivory Coast's now-more-specific season-delay data. The asymmetry reads differently leg by leg right now — cocoa and wheat look like price pullbacks inside intact stories (potentially better entries), coffee looks like a thesis that keeps getting more wrong on the data, and sugar looks like a thesis absorbing its first real fundamental pushback. Per Jeff Currie's widened frame (mid-Aug 2026), the hard-asset complex vs. overvalued equities call is being genuinely tested this week, not just extended.
Long Sugar #11 — HELD since May 22, added Aug 2026
SBWhy
Triple-exposed — Brazil cane (drought sensitivity), Indian monsoon (rainfall failure risk), Australian crush — plus fertilizer compounding (heavily N-dependent crop). The multi-year-low entry thesis has played out and largely held, but picked up a genuine new headwind this week: UNICA's 2H-August Centre-South report showed Brazilian mills pulling back from ethanol toward sugar as prices rallied (sugar mix back to ~46.4% from ~41.4% in July, essentially flat year-over-year vs. the prior sharp ethanol-pivot narrative) — the clearest evidence yet that the mill-mix-shift leg of the deficit thesis can reverse when prices climb far enough. The India leg is unchanged and still supportive: June-Aug monsoon deficit finished at 13.8% (IMD), August itself the 7th-driest month since 2001, September outlook still below-normal. Front-month price action this week could not be reliably confirmed (source access issues); last confirmed settle is 18.07¢ (Sept 4). Full detail and conviction level (trimmed from 80 to 77 on the mill-mix reversal) on /sugar.
Expression
Executed: 2× Feb'27 18¢ calls, plus 1× Feb'27 19.5¢ call added Aug 2026 for convexity. Add trigger per plan: deepening UNICA ethanol-pivot data. Any further adds: consider later months (SBK7/SBN7) or higher strikes. Live marks on /sugar.
Long Cocoa — Best odds on the board per the Aug 19 check
CCWhy
Odds check (Aug 19, spot ~6,070, realized vol ~58%): implied P(>6,520 — the July high) ≈ 35%, P(>8,500 — the suggested spread's short leg) ≈ 15%. Both targets have deep recent precedent: 6,520+ traded in 19 of the last 154 months and 8,500+ in 13 of them (the 2024-25 shortage regime, ATH ~12,260). Compare coffee, where the equivalent tail strike has NEVER traded. So the market pays ~6:1 on a level cocoa occupied for over a year, with a dated catalyst (Oct-Dec pod-fill at the El Niño peak) still ahead. The two-sided caveats stand — ICE inventories at a 2-yr high, Ivory Coast arrivals +20% YoY — which is exactly why the odds are this fat. UPDATE (Aug 30): spot has since jumped to ~$6,233/t on a fresh 4.2% single-day move, already inside the 6,520 target zone, as Ivory Coast (1.35-1.45 MMT) and Ghana (450-550k MT) 2026/27 estimates worsened further — the odds picture above needs re-pricing at the higher spot. UPDATE (Sept 4): spot has cooled slightly to ~$6,171/t, still inside the 6,520 target zone, as the market digests genuinely divergent forecaster views — Guan Chong (the world's #4 cocoa grinder) warned Sept 2 of a 300-400k tonne global 2026/27 DEFICIT, while StoneX and Hedgepoint still project a surplus, just narrowing sharply (StoneX to ~25k tonnes from 422k this season; Hedgepoint to ~111k tonnes from 325k). Ivory Coast's regulator also flagged Aug 26 that the new season could run 8-10 weeks late, risking a Q4 arrivals/port-congestion crunch — a fresh, dated catalyst that sits squarely inside the Oct-Dec pod-fill window this trade is underwritten by. UPDATE (Sept 12): spot cooled further to ~$5,985/t (Dec contract, -2.95% that session), back below the 6,520 target zone entirely. The deficit case itself didn't weaken — Guan Chong's call stands and Ivory Coast's delay warning got more specific (arrivals below 15k t in September, 25k t in October, ~900k t at ports Oct-Dec vs. 1.1M t last year) — so this reads as a price pullback inside a still-intact fundamental story, not a reversal of the thesis, and arguably a better entry than the late-Aug spike.
Expression
Options only (DCC contracts) — outright futures are too volatile and margin-intensive for sizing discipline. With spot ~6,070, Mar'27 call spreads around 7,000/9,000 keep defined risk; the Aug pullback zone is the entry the June chase never offered. Trigger to add: first weak pod-count/arrivals data in Oct.
Sizing note
Small relative to oil book — think 5-10% of oil notional. This is a satellite convex bet, not a portfolio allocation.
Long EU Natural Gas (TTF) — Hard-deadline refill trade
TTFWhy
EU storage was 60.4% full on Aug 15 — roughly 9-10pp below the 5-yr seasonal norm and the lowest early-August reading on record — with Qatari LNG cargoes delayed by shipping disruptions and TTF already ~€62.5/MWh. UPDATE (late Aug): TTF has extended to ~€69/MWh (a 3-year high above €68 around Aug 25, briefly dipping to €65.40 on Aug 26) and storage has risen only to 63% by Aug 26 — described by trackers as the lowest seasonal level since 2021 (or, by some sources, since 2009), still well below the 5-yr average. Even hitting the relaxed 80% Nov-1 target leaves Europe entering winter with buffer in the low-70s vs 90%+ in normal years: no headroom for cold snaps. Per Currie (mid-Aug 2026): 'dangerously, precariously tight' — a read the subsequent price action supports. UPDATE (early Sept): storage improved modestly to 66.4% (Sept 4, up from 64.4% the prior weekend) but remains -16.2% YoY and -20.2% vs. the 5-yr average — some catch-up in the injection pace, not a closing of the gap. TTF itself pushed higher — one report puts it near €80/MWh, the highest since December 2022 (exact date unconfirmed this week, source access blocked), a real move above the ~€69 late-Aug level if it holds. Driver: the Hormuz blockade is now estimated to have disrupted ~20% of global LNG flows, mostly Qatari cargoes — directly tying this trade to the same conflict as the sugar/cocoa/oil legs. Note US natgas is NOT the expression — the US is an island; this must be TTF (or JKM).
Expression
Defined-risk winter calls (Dec'26–Feb'27) rather than outright futures — the payoff is weather-contingent and a mild winter plus LNG recovery kills it. Check IBKR permissions for ICE Endex TTF; US-listed gas ETPs track Henry Hub, not TTF, and do not express this trade. Trigger discipline: if the EU storage trajectory closes the gap to the 5-yr norm by late Sep (the Oct 31 refill-window close), the edge is gone — stand down. So far storage is not closing that gap.
Long Gold — Systemic hedge, now a dip-buy of the January spike
GCWhy
Correction to the earlier framing (Aug 19 review): 'flat YTD' hides the path — gold spiked to $5,586 on Jan 29, 2026, held $5,300-5,400 into early March, and has retraced ~21% to ~$4,420. It HAS run; Currie's mid-Aug re-entry is a dip-buy of that spike, not a pre-breakout entry. Odds check (realized vol ~29%): implied P(>5,000) ≈ 24% — a retest of levels that traded for weeks in Q1 — and P(>6,000) ≈ 5.5% (never traded). The thesis (30-yr UST at 5.22%, 65-month CPI streak, Japan selling USTs to defend the yen, US selling euro reserves — 'when one breaks, they probably all break') is unchanged; what changed is knowing you're buying a 5-month drawdown, with the risk that January WAS the blow-off. Then the Aug 19 buyback announcement provided the first live test of the thesis: within 45 minutes of Treasury doubling long-bond buybacks, gold ripped $100 to ~$4,460 — its best gold-ETF day in ~4 years — on FALLING yields plus a -0.8% dollar. That's the tell: gold is trading as the debt-monetization hedge, not the rate-level trade. UPDATE (Aug 30): the move has extended, not faded — gold reached ~$4,647 (highest since mid-May) and silver briefly topped $71 before settling near $67. But the caveat just got less hypothetical: on Aug 28, incoming Fed Chair Warsh's hawkish Jackson Hole remarks pushed 2-year yields from 4.22% to 4.30% — the exact 'surprise hawkish Fed' risk this idea already flagged, now live rather than theoretical, though it hasn't yet dented the gold move. UPDATE (early Sept): the Warsh risk showed up in the tape, briefly — gold pulled back from ~$4,647 to around $4,470, touching a three-week low, before rebounding on dovish comments from other Fed officials. Read that as confirmation the position is a genuine two-sided bet on Fed messaging, not a one-way ratchet. UPDATE (Sept 12): the Warsh risk stopped being brief — gold fell for a THIRD straight week to ~$4,347-4,350 as CME FedWatch odds of a Sept 15-16 25bp hike jumped to ~86% (from ~70% pre-CPI) on continued hawkish commentary plus a hot CPI/PPI mix, and the Sept 9-10 Treasury buyback operation (only $6B, tripling the old ceiling but read as underwhelming) failed to repeat the Aug 19 rally — long yields rose instead of falling. Silver diverged positive (+1.5% to ~$65) the same day gold fell, worth watching as a possible tell that the drawdown is Fed-specific rather than a broad hard-asset unwind. This is now a genuinely live drawdown test of the thesis, not a hypothetical one.
Expression
Outright futures or a GLD-equivalent for the sleeve. Currie's own caveat: a surprise Fed rate hike (a Warsh-style move) is the main drawdown risk — Warsh's Aug 28 Jackson Hole comments are a live instance of exactly that risk, worth watching for follow-through. Size so that scenario is survivable. High-beta alternative: silver (~$67, briefly $71 intraday) historically outruns gold in monetary-fear regimes as the gold/silver ratio compresses — but its industrial half carries a structural drag (solar makers thrifting silver loadings), so treat it as a leveraged expression of this card, not a substitute.
Long Coffee (Arabica) — Wait for the September rainfall trigger
KCWhy
The setup is real but the odds are now roughly fairly priced (Aug 19 review): backed out of the Mar'27 chain (~26-31% IV), the market implies ~1-in-6 for a retest of 400 and ~1-in-25 for new all-time highs (>470, above the 441 record). The base rate is comparable — three weather rallies in 15 years, each roughly doubling the price (2014 drought, 2021 frost, 2024/25 shortage; KC spent nine months of 2025 above 400) — and the cleanest counterexample cuts against the thesis: the 2015-16 super El Niño produced NO coffee rally (KC highs 185/176). The tensions that keep it on the board: ICE-certified stocks lowest since late 2023, Brazil Jan-Jul exports -12.2%, spec positioning flat, and the record 69.3M-bag 2026/27 forecast still depends on Sep-Oct flowering rains arriving on time. UPDATE (Sept 2-5): the live data moved against this trade this week, honestly. Safras & Mercado raised Brazil's 2026/27 crop forecast to 75.65M bags (49.95M Arabica) on the back of Minas Gerais rains running as high as 127% of the historical average in the latest week — the opposite of the disappointing-flowering-rain trigger this idea needs. Dec arabica sold off >3% to a ~40-day low below 300¢ on the news before a partial one-day bounce (+1.14% Sept 3). This does not kill the thesis (El Niño's peak intensity is still Nov-Feb, and one good rain week doesn't undo a season), but it meaningfully lowers the odds the September trigger fires as written. UPDATE (Sept 11): the decline extended, not reversed — Dec arabica closed $2.84/lb, down ~16.4% over 30 days. No fresh rainfall data surfaced this week to reverse the Brazil crop-forecast narrative; the entry trigger has still not fired and is further from firing than it was two weeks ago.
Expression
NO ENTRY WITHOUT A TRIGGER — and the trigger just got harder to satisfy: this week's rain data argues AGAINST 'Minas Gerais rainfall disappointing in September' rather than for it. Keep watching Cooxupé/Cepea field reports for a reversal back to dryness through the rest of September before considering Dec'26 OTM calls (expiry ~mid-Nov, just past the flowering verdict) or a Mar'27 call spread (~430/500 for a ~$1k debit at current vol). Until the trigger fires, hold fire — and this week is a reminder that the base case (no entry) is currently doing its job.
Long Distillates (HO) — Strongest fundamentals, worst entry
HOWhy
The most confirmed leg of the Currie thesis: the US diesel crack broke $100/bbl for the FIRST TIME EVER on Aug 17 ($102.20 intraday) and settled in triple digits from Aug 18 onward — confirmed structural, not a spike. Distillate stocks kept falling, ~105.6M bbl for the week ended Aug 14, the lowest for the date since the mid-1990s, ~12-13% below the 5-yr average. Refinery strikes have only intensified since (Kirishi hit Aug 30, one of Russia's largest, weeks after resuming operations) and Russia's own export constraints continue to stack on the SPR grade mismatch. Structural tightness into winter — but it is also the most repriced idea on this page (HO +106% YTD), so the risk-adjusted entry is poor at the record print. Odds check (Aug 19, spot ~$4.32, realized vol ~57%): implied P(>$5.15 — the 2022 all-time-high zone, traded in just 1 of 154 months) ≈ 26%, P(>$6.00 — never traded) ≈ 15%. Like coffee's 470 strike, the upside targets here are record prints, not revisits — the odds look generous only because crisis-year vol is in the input. Confirms the tier-3 'worst entry' call. UPDATE (Sept 13): the structural case just got stronger, not weaker — Ukraine's cumulative refinery-strike tally is now 60+ hits on 24+ Russian refineries (21 in August alone, the war's highest single month), Russian refinery output fell to ~3.8 mb/d (from ~5.0 mb/d), Russia extended its diesel export ban through Sept 30, and the crack spread hit a fresh record $108.02/bbl with East Coast distillate stocks down to just 19.3M barrels. President Trump publicly called on Ukraine (Sept 13) to halt strikes on Russian diesel targets specifically — the first political pressure aimed directly at this transmission mechanism, worth watching as the one plausible near-term de-escalation path for this trade.
Expression
Don't chase the record: buy pullbacks or use defined-risk Dec/Jan calls. The crack spread (long HO / short CL) isolates refining tightness from crude-level ceasefire risk. The oil book already carries correlated exposure — size as a satellite, not a new core.
Sizing note
Overlaps the existing oil position — treat as an extension of the molecules leg, not a diversifier.
Long KC Wheat — Compound winner, now substantially repriced
KEWhy
The +9% 'room to extend' call extended further still, then reversed on the first real diplomatic headline: CBOT wheat pushed to $7.95/bu (Sept 2-3, a fresh 3.5-year high) as Black Sea loadings stayed down ~60% over the trailing 5 weeks, before pulling back to $7.42 by Sept 5 after Putin said a peace deal 'remains possible' and Turkey proposed a Black Sea grain-passage plan. This is precisely the scenario flagged two weeks ago as the risk to chasing this trade: a ceasefire/passage-plan headline doesn't fix the underlying 61-yr-low US winter wheat crop, but it can and did remove $0.50+/bu of the Black Sea risk premium in three trading sessions. Odds check (Aug 19, ZW ~684¢): implied P(>800¢) ≈ 18%; wheat briefly got within $0.05 of that level (Sept 2-3) and has since given most of that move back. UPDATE (Sept 13): the unwind continued and the diplomatic track stalled rather than progressed — CBOT eased further to ~$7.53/bu and UK wheat to £212.50/t, but Turkey's shipping-corridor proposal was rejected by Russia, and Moscow said just before Putin's remark that there were no grounds for restoring the Black Sea grain deal. The Sept 11 WASDE left wheat ending stocks unchanged, offering no new signal. Read together: the risk premium keeps deflating on rhetoric even though nothing concrete has actually been agreed — exactly the kind of gap between price action and physical confirmation this idea already flagged as the risk.
Expression
This week is the clearest evidence yet that most of the asymmetry has been captured and is two-way, not one-way — the same headline risk that built the position can unwind it in days. Prefer buying dips on confirmed Black Sea de-escalation (the underlying US-crop supply math survives a ceasefire) over chasing spikes; use defined-risk calls, sized smaller than the softs trades, and treat the Putin/Turkey headlines as the live test of how much of this move is diplomacy-reversible.
Long Platinum — Deep value after a positioning crash, no dated catalyst
PLWhy
The -19% YTD (~$1,727) is a burst positioning bubble, not a broken fundamental story: platinum spiked toward ~$2,900 around the turn of the year on investment inflows, then crashed on ETF liquidation. Underneath, the market is in its 4th consecutive annual deficit and WPIC projects deficits averaging ~331koz/yr through 2030 — 2026 demand falls 8% only because last year's investment mania isn't repeating (industrial +11%, auto just -3%). It trades at ~0.39× gold vs a historical premium. Currie kicker: South African supply is extremely energy-intensive, so the diesel/power squeeze curbs output at exactly the wrong time. The catch: no dated catalyst — this is a regime trade that fires with the monetary leg, and 'structural deficit' hasn't stopped a 40% drawdown before. Flows set the price here, not the deficit. Odds check (Aug 19, spot ~$1,743, realized vol ~56% post-crash): implied P(>$2,200) ≈ 22% — a +26% move to a level that traded only 2 of the last 154 months — and P(>$2,900, above the $2,852 spike high) ≈ 7%. The deficit story doesn't need new highs to pay: the 2,200 zone is the realistic target, and ~4.5:1 implied odds against a persistent structural deficit is the actual bet.
Expression
The 50-oz contract ($50 per $1 move) makes PL one of the most retail-sizable metals — outright futures or long-dated calls, but options are thinner than gold/silver so check spreads. Expect the ETF overhang to sell rallies: scale in on weakness rather than chase, and treat it as the deep-value satellite of the atoms leg.
Timing Windows — When Impacts Hit the Tape
Catalyst calendar for the El Niño + Hormuz compound thesis
India monsoon
LIVE, STILL NO RECOVERY: the full June-Aug season closed at a 13.8% cumulative deficit (IMD) — August itself was a -16% monthly deficit (Skymet), the southern peninsula the worst-hit region at -27% as of Sept 7. Both IMD (<91% of LPA) and Skymet (-20% of LPA) September outlooks point to continued below-normal rainfall. No reversal toward normal has shown up yet. Drives sugar, rice, cotton, Indian wheat.
Australian ABARES estimates
Wheat production downgrades if El Niño-driven dry pattern develops. First major Australian harvest signal.
Brazilian Centre-South cane
REVERSED, not deepened: UNICA's 2H-August 2026 report published (~Sept 10-11) and shows sugar mix back to 46.43% (vs. 46.73% a year earlier, essentially flat YoY) — a sharp reversal from the ~41-42% mix seen through July, as high sugar prices (ICE up ~40% off its June low) pulled mills back from ethanol. Crush +2.08% YoY, sugar output +0.69% YoY (vs. -17.6% YoY in 2H-July). This converges toward, not further below, USDA's full-season 2026/27 Sugar Annual forecast of ~48% sugar/52% ethanol. This is the clearest evidence yet that the mill-mix-shift leg of the sugar deficit thesis can reverse — a genuine headwind, flagged explicitly on /sugar. UNICA biweekly reports remain the add/trim trigger for the sugar position.
August CPI print + WASDE report — PUBLISHED
Both landed on schedule Sept 11. CPI confirmed the diesel pass-through Currie flagged: headline +0.4% m/m/+3.4% y/y, energy +2.1% m/m, gasoline +3.9% m/m (over a third of the headline increase), energy +16.3% y/y (up from +14.7% y/y in July) — core CPI eased slightly to 2.4% y/y. WASDE read bearish for grains: corn yield cut to 178.5 bu/acre (largely pre-priced, didn't move the market much), soybean production came in near-record (52.8 bu/acre, 4.5B bu) and sold off hard on 'no bullish surprises,' wheat stocks unchanged.
Treasury upsized buyback window — OPENED, first result underwhelming
Opened as scheduled Sept 9. The first operation (Sept 9-10) bought up to $6B — tripling the old $2B ceiling but read by the market as underwhelming relative to Treasury Secretary Bessent's earlier framing. The result was the opposite of the Aug 19 pattern: long yields ROSE rather than fell (10-yr to ~4.97-4.98%, highest since Oct 2023), and gold fell for a third straight week to ~$4,347-4,350 rather than rallying. Combined with CME FedWatch pricing ~86% odds of a 25bp hike at the Sept 15-16 FOMC, this is the clearest sign yet that the 'buybacks hold the curve down' mechanism is being challenged, not confirmed. Watch the remaining Sept-Nov operations and the actual FOMC decision as the next tests.
Brazilian coffee flowering
Critical period for the 2027 crop. Dry conditions during flowering trigger sharp arabica futures moves. This is the explicit ENTRY TRIGGER for the coffee trade idea, and the live data has now moved the WRONG way for the bull case for two straight weeks: parts of Minas Gerais logged rainfall as high as 127% of the historical average, Safras & Mercado raised its Brazil 2026/27 crop forecast to a record 75.65M bags (49.95M Arabica), and Dec arabica has fallen further since (closing $2.84/lb Sept 11, down ~16.4% over 30 days) rather than bouncing. ICE-certified stocks remain near the 1999 low, keeping the setup alive in principle, but no fresh dry-weather data surfaced this week to revive it. No entry trigger has fired, and it is further from firing than two weeks ago.
EU gas refill window closes
Per Currie (mid-Aug 2026): European refill season is well behind schedule, and winter distillate + gas inventories should be rebuilding now and aren't — 'dangerously, precariously tight.' Storage improved modestly to 66.4% (Sept 4, from 64.4% the prior weekend) but remains -16.2% YoY and -20.2% vs. the 5-yr average — some catch-up in pace, not a closing of the gap. TTF itself pushed higher still, with one report citing ~€80/MWh (highest since Dec 2022, exact date unconfirmed) versus ~€69 in late August. The Hormuz blockade is now estimated to have disrupted ~20% of global LNG flows (mostly Qatari cargoes), directly linking this window to the same conflict as the rest of this page. The refill window effectively closes ~Oct 31; whatever isn't in storage by then is winter's problem.
West African cocoa main crop
Pod-fill phase. This is when El Niño impacts on cocoa get priced in most aggressively. The big catalyst window — now complicated further by the season-start delay: Ivory Coast's Conseil du Café-Cacao's Aug 26 warning (8-10 weeks late on weather and a lingering mid-crop) was detailed further this week — weekly arrivals are now expected below 15,000 t in September and 25,000 t in October, with ~900,000 t expected at Ivorian ports Oct-Dec vs. 1.1 million t in the same window last year, risking real port congestion at Abidjan/San Pedro ahead of EUDR. Ghana's ~6% farmgate price hike adds a fresh cross-border smuggling-incentive risk for Ivorian beans. A port-congestion and timing risk layered on top of the weather risk this window was already tracking.
El Niño peak intensity
NOAA expects this is when atmospheric coupling is strongest. Maximum impact on global weather patterns.
Show structural context (4 sections)
Energy is the standout
Refined products (RBOB, heating oil) are outperforming crude as the Hormuz disruption hits product flow more than crude supply — confirming the structural thesis that this isn't just a barrel issue, it's a logistics issue. This week sharpened that further: the diesel crack spread hit a fresh record $108.02/bbl, and August CPI (released Sept 11) showed real pass-through — energy +2.1% m/m, gasoline +3.9% m/m, energy +16.3% y/y, up from +14.7% y/y in July.
Grains are the second-order Hormuz trade — WASDE landed bearish, and the Black Sea rally kept unwinding
Corn and soy strength began as fertilizer cost transmission and a direct US supply shock. The Sept 11 WASDE report published on schedule and disappointed the bulls: corn's yield cut (to 178.5 bu/acre) was largely pre-priced and didn't move the market, while soybean production came in near-record (52.8 bu/acre, 4.5B bu) and sold off hard (Nov beans -35.75¢) on 'no bullish surprises'; wheat stocks were unchanged. Separately, the Black Sea escalation that drove CBOT wheat to $7.95/bu (Sept 2-3) kept unwinding — CBOT eased further to ~$7.53/bu and UK wheat to £212.50/t — but Turkey's shipping-corridor proposal has stalled: Russia rejected it, and Moscow said just before Putin's 'peace deal remains possible' remark that there were no grounds for restoring the Black Sea grain deal. Read together: prices are giving back the war premium on rhetoric, while the underlying disruption (Black Sea loadings, the US crop shock) remains structurally unresolved — exactly the 'jawbone vs. physical confirmation' distinction this page's oil-side sibling tracks.
Softs are diverging further, not converging — and sugar picked up a genuine new headwind
In May this page argued softs were priced for the wrong direction. Cocoa and coffee kept diverging from each other this week: cocoa cooled to ~$5,985/t (Dec contract, Sept 12, -2.95% that session) off its highs even as Guan Chong's 300-400k tonne 2026/27 deficit call stood and Ivory Coast's season-delay warning got more specific (weekly arrivals below 15k t in September, 25k t in October, ~900k t expected at ports Oct-Dec vs. 1.1M t a year ago); coffee kept falling — Dec arabica closed $2.84/lb (Sept 11), down ~16.4% over 30 days, extending rather than reversing the break from Brazil's raised crop forecast. Sugar's picture got more complicated: UNICA's 2H-August Brazilian Centre-South report showed mills pulling back from ethanol toward sugar as prices rallied (sugar mix back to ~46.4% from ~41.4% in July, essentially flat year-over-year) — a real, if moderate, reversal of the mill-mix-shift leg of the deficit thesis, even as India's monsoon and El Niño legs held (see /sugar for full detail). None of the three crops is moving in lockstep with the others right now.
Metals dispersion resolved — copper joined the bid, cooled, then took a tariff-headline shock
The May read was that copper (+8%) lagging silver/gold flagged weak industrial demand. That's now stale: copper set an all-time COMEX high (~$6.77/lb) on Aug 7, has since eased to ~$6.47/lb, and took a genuinely new hit this week — a single-session drop of nearly 5% on reports the US is reconsidering import tariffs on refined copper. Still +41.2% y/y, so the structural (electrification/AI demand, tariff) thesis isn't broken, but this is a real two-sided week for the 'atoms' leg, not a clean continuation.
Refinery strikes are the new structural regime — and the tally keeps growing
The count keeps climbing: Ukraine's cumulative tally is now 60+ strikes across 24+ Russian refineries, with 21 strikes in August alone — the single highest month of the war. Russian refinery output fell from ~5.0 mb/d to ~3.8 mb/d, and Russia extended its diesel export ban through Sept 30. Cheap drones make energy infrastructure trivially attackable — targeting refineries is now a proven strategy other actors will replicate. The transmission is diesel: the crack spread hit a fresh record $108.02/bbl this week, with East Coast US distillate inventories at just 19.3M barrels. Notably, President Trump publicly called on Ukraine (Sept 13) to halt strikes on Russian diesel targets specifically — the first sign of political pressure aimed at this exact transmission mechanism, not yet reflected in any actual de-escalation.
The monetary endgame is the hard-asset bid — but this week the bid itself cracked
Per Currie (mid-Aug 2026): the endgame sequence is oil pops → the long end reprices well above 5.25% → financial repression → equity de-rating, expressed as HALO (Hard Assets, Local Operations) — long molecules, atoms, and calories against overvalued US equities. This week tested the hard-asset half of that trade directly: gold fell for a third straight week to ~$4,347-4,350 as rate-hike odds jumped (CME FedWatch ~86% for a 25bp hike at the Sept 15-16 FOMC, up from ~70% pre-CPI) on hawkish Fed commentary plus a hot CPI/PPI mix — the opposite of a debasement trade. Silver, notably, diverged positive (+1.5% the same day). The 10-year yield itself climbed to roughly 4.97-4.98%, its highest since October 2023. This is the clearest evidence yet that a genuinely hawkish Fed path can pressure gold even inside a monetary-fragility narrative — the endgame thesis isn't dead, but this week argues the path there is less linear than the mid-August framing implied.
Financial repression's next test arrived — and came up short
Currie's endgame step 3 (the Aug 19 Treasury buyback pivot) got its first live test this week: the Sept 9-Nov 4 buyback window opened as scheduled, and the first operation (Sept 9-10) bought up to $6B in longer-dated debt — tripling the prior $2B ceiling, but read by the market as underwhelming relative to Treasury Secretary Bessent's earlier framing. The result was the opposite of the Aug 19 pattern: long yields ROSE rather than fell on the news, and gold extended its decline rather than rallying. Combined with incoming Fed Chair Kevin Warsh's continued hawkish trajectory (CME FedWatch now ~86% for a Sept 15-16 hike), this is the clearest sign yet that the 'buybacks hold the curve down' mechanism is being actively challenged rather than confirmed — watch the remaining Sept-Nov buyback operations and the actual FOMC decision as the next real tests.
Live Prices Table
All 17 contracts · Click headers to sort
| Commodity | Price | YTD ▼ | Symbol | ||
|---|---|---|---|---|---|
Heating Oil Energy | Energy | 4.830 $/gal | +128.3% | +6.37% | HO=F |
RBOB Gasoline Energy | Energy | 3.210 $/gal | +89.2% | -0.04% | RB=F |
WTI Crude Energy | Energy | 102.85 $/bbl | +79.4% | +12.43% | CL=F |
Brent Crude Energy | Energy | 107.67 $/bbl | +77.2% | +11.83% | BZ=F |
Wheat Grains | Grains | 730.50 ¢/bu | +44.2% | +2.03% | ZW=F |
Cotton Softs | Softs | 85.060 ¢/lb | +32.9% | +3.20% | CT=F |
Sugar Softs | Softs | 19.070 ¢/lb | +30.6% | +5.53% | SB=F |
Soybeans Grains | Grains | 1297.50 ¢/bu | +26.0% | +0.29% | ZS=F |
Corn Grains | Grains | 529.50 ¢/bu | +21.0% | +3.42% | ZC=F |
Copper Industrial Metals | Industrial Metals | 6.440 $/lb | +14.1% | -2.46% | HG=F |
Cocoa Softs | Softs | 5954.00 $/MT | +1.4% | -2.10% | CC=F |
Gold Precious Metals | Precious Metals | 4329.70 $/oz | +0.3% | -2.26% | GC=F |
Live Cattle Livestock | Livestock | 222.23 ¢/lb | -5.8% | +3.70% | LE=F |
Silver Precious Metals | Precious Metals | 63.250 $/oz | -10.4% | -4.24% | SI=F |
Platinum Precious Metals | Precious Metals | 1770.10 $/oz | -16.7% | -2.80% | PL=F |
Natural Gas Energy | Energy | 2.900 $/MMBtu | -19.9% | -2.55% | NG=F |
Coffee Softs | Softs | 286.00 ¢/lb | -20.0% | -11.80% | KC=F |
What kills this trade
- Weaker-than-forecast El Niño peak — the event is confirmed (NOAA Advisory) and the Sept 10 CPC Diagnostic Discussion still shows very-strong-peak odds above 90% with a 75% chance of a 'historic' event (RONI ≥ +2.5°C, exceeding every El Niño since 1950) — a marginal trim from August's 95%/69% split (the SPEAR model's Sept run projects a slightly lower peak), so residual risk keeps shrinking but isn't zero. Related: the Indian monsoon closed its June-Aug season at a 13.8% cumulative deficit (IMD), August itself a -16% monthly deficit (Skymet), and both IMD's and Skymet's September outlooks still point below-normal — a genuine catch-up to normal would still weaken the India leg, but the data continues to move the other way, not toward recovery.
- Combined unwind — Hormuz resolves AND El Niño underperforms. Worst case for both legs; size accordingly. This week offered the first real (if stalled) test of the Hormuz half: Iran negotiated a temporary shipping-lane agreement with Oman, due to be presented to Gulf nations Sept 14 — the most concrete de-escalation process attempted since the war began — but that meeting was just postponed (Bahrain refusing to participate, Houthi escalation in Yemen complicating), while an Iranian vessel was struck near Hengam/Qeshm Island around Sept 12-13 and Iran announced plans for a wider maritime 'exclusion zone.' Net: genuinely two-sided, not a clean de-escalation. The standing unwind's monetary leg (yen/UST stabilization) got a live test too: the Sept 9-10 Treasury buyback bought only $6B (versus Bessent's bigger framing) and long yields ROSE rather than fell — the opposite of the Aug 19 pattern — while gold fell a third straight week on hawkish Fed odds. That is arguably the clearer near-term unwind signal this week: not Hormuz resolving, but the monetary-repression trade failing its own test. The wheat/Black Sea leg's diplomacy-reversible risk kept playing out: Turkey's shipping-corridor proposal stalled (Russia rejected it) even as wheat kept giving back its war premium on rhetoric alone (CBOT to ~$7.53/bu).
- Surprise Fed hike — Currie's own caveat (mid-Aug 2026): a Warsh-style surprise rate hike would knock gold and the whole hard-asset complex temporarily. This risk stopped being hypothetical and became the week's dominant macro story: CME FedWatch now prices ~86% odds of a 25bp hike at the Sept 15-16 FOMC (up from ~70% pre-CPI) on continued hawkish commentary plus hot CPI/PPI data, and gold fell for a THIRD straight week to ~$4,347-4,350 — no longer a brief pullback but a sustained multi-week decline. This is now the most live risk on the page, not the most theoretical one.
- Cocoa-specific: faster-than-expected Ghana/Ivory Coast harvest recovery announcement before El Niño impacts show in production data. This risk keeps moving the WRONG way, not the right way — Ivory Coast's season-delay warning got MORE specific this week (arrivals below 15k t in September), Guan Chong's deficit call stands, and Ghana's farmgate price hike adds a fresh cross-border smuggling risk — so the near-term risk to this specific trade currently looks smaller, not larger, even as spot cooled on price alone.
- Coffee-specific: this risk is no longer a single live data point — it has now persisted for two straight weeks. Brazil's 2026/27 crop forecast (raised Sept 2 on strong Minas Gerais rains) has not been reversed by any fresh data, and Dec arabica has fallen further (to $2.84/lb, -16.4% over 30 days). The entry trigger for the coffee trade has not fired and is further from firing than two weeks ago. Sugar-specific: this risk fired partially this week — UNICA's 2H-August report showed Brazilian mills reverting toward sugar from ethanol as prices rallied (mix back to ~46.4% from ~41.4% in July) — a genuine, if moderate, headwind to the deficit case that the sugar position (see /sugar) has already priced into a trimmed conviction level (80 → 77).
Connection to existing oil book
This thesis is complementary, not redundant, to the Hormuz oil position. Both express variants of the same macro view (global supply shock, geopolitical/weather catalyst convergence), but they're driven by largely independent transmission mechanisms and will move on different news flow. Suggested allocation: 20-30% of oil notional, spread across 2-3 of the trades above. Cleanest portfolio construction: sugar as core (lowest vol), cocoa as convex satellite (options only), coffee as opportunistic add if it breaks lower. In Jeff Currie's HALO framing (mid-Aug 2026), the oil + sugar book already covers the molecules + calories legs — gold/copper is the missing atoms/monetary leg, which is what the new gold idea addresses.