# Copper fell toward its stop even as strike risk rose

**Research cutoff:** September 28, 2026, 12:10 a.m. Toronto  
**Review:** Reopened-session monitor; changes since September 27 at 6:05 p.m. Toronto  
**Economic decision:** Hold all three paper exposures, make no additions, and enter no new trade.

The strongest signal is uncomfortable rather than dramatic: **December copper fell toward the portfolio's 6.60 invalidation even while an Escondida labor vote raised the possibility of a strike**. The delayed same-month standard `HGZ26` proxy was **$6.6670/lb** at 11:56 p.m. Toronto, about 1.5% below Friday's 6.7660 settlement. That price action says the disruption headline is not enough to offset softer demand evidence, contango and crowded speculative length.

The correct response is disciplined: **hold the existing +3 `MHGZ26`, do not add, and do not widen the 6.60 stop**. The proxy's session low was 6.6515, above the stop, but a standard-contract range cannot prove the exact micro never touched it. This report therefore records a valuation mark, not an assumed exit or a fabricated stop path. [Delayed HGZ26 quote](https://www.marketwatch.com/investing/future/hgz26)

No paper trade was opened, added, trimmed or closed. The simulation only updated dated marks and retained the existing positions.

## Portfolio and decisions

| Paper exposure | Latest reference | Gross open P&L | Decision |
|---|---:|---:|---|
| +3 December Micro Copper `MHGZ26` | **6.6670** via delayed `HGZ26` proxy | **−$937.50** | **Hold; no add; stop 6.60** |
| −8 December Micro Euro `M6EZ26` | **1.1424** exact delayed | **−$20.00** | **Hold; no add; stop 1.1550** |
| +3 March Micro Soybean Oil / −3 March Micro Soybean Meal | **68.22¢ / $364.90** via delayed standard proxies | **+$82.20** | **Hold; no add before Sep. 30** |

Paper NAV is **$999,861.70**, with **−$875.30** gross open P&L and **−$138.30** cumulative net P&L. Cash remains $1,000,737; the portfolio holds $41,000 of conservative modeled margin reserves. The ledger's entry-based buffered open-risk measure is **$4,904.35, or 0.49% of NAV**. That is deliberately conservative and differs from the smaller current-mark-to-stop distance after market moves.

The copper and oilseed marks use compatible standard-contract proxies in the same quoted units. They are not represented as observed micro trades. Oil and meal were observed four minutes apart, so the basket mark is useful for valuation but not a synchronized executable package.

## 1. Copper: physical scarcity is real, but the market is rejecting the easy headline

### What changed

China's January–August industrial profits rose **15.7% year over year**, slowing from **17.6%** through July. The level is still strong, but the deceleration weakens the demand-confirmation side of the copper thesis. [China National Bureau of Statistics](https://www.stats.gov.cn/sj/zxfbhjd/202609/t20260928_1965425.html)

At the same time, Escondida supervisors were scheduled to vote September 28–30 on BHP's contract offer. A rejection would begin mandatory government mediation before a strike; it would not instantly remove mine supply. Operations were already gradually resuming after the prior accident. Treat the vote as a catalyst, not confirmed lost production. [Escondida vote](https://www.marketscreener.com/news/bhp-escondida-mine-supervisors-union-urges-rejection-of-contract-offer-strike-looms-ce785ad9df8ff521), [operations update](https://www.investing.com/news/stock-market-news/unions-at-chiles-escondida-mine-reject-bid-to-pause-contract-talks-after-workers-death-4917884)

The physical evidence remains supportive but is not new overnight. SHFE copper inventories were 47,147 tonnes, down 15.9% in the week, while LME cash copper traded at a $124.75/tonne premium to three months on September 24. That London backwardation signals nearby tightness. It should not be confused with the COMEX curve: delayed December at 6.6670 versus March around 6.7660 implies roughly **9.9¢/lb of contango**, using asynchronous observations. [Copper inventories and LME spread](https://www.brecorder.com/news/40441192/)

Positioning is the other warning. September 22 CFTC futures-only data show non-commercial copper net length of **90,522 contracts**, up 15,388 in a week. Regional scarcity is not an undiscovered view; crowded ownership can accelerate liquidation when demand evidence disappoints. [CFTC metals positioning](https://www.cftc.gov/dea/futures/deacmxsf.htm)

### Payoff and invalidation

The prior November 13 scenario tree—26% at 7.20, 41% at 6.85 and 33% at 6.60—has a weighted endpoint of 6.8245. From the 6.6670 proxy, that is about **+$1,181 gross forward value** for three micros. But only $502.50 of raw distance remains to the stop before friction and any gap. Small probability changes can erase the apparent edge.

**Decision: Hold +3 `MHGZ26`; no add.** Keep the 6.60 stop, 7.20 target and November 13 time exit. A break below 6.60 ends the trade; labor headlines do not justify widening protection.

## 2. Euro: rate support remains, but the short is already crowded

Exact delayed `M6EZ26` traded **1.1424** at 11:57 p.m. Toronto, near the 1.1422 assumed paper entry. The contract's session range was 1.1411–1.1428. March was about 43 pips above December at a nearby but asynchronous time, showing upward curve carry that is already embedded rather than a separate guaranteed gain. [Exact M6EZ26 quote](https://www.wsj.com/market-data/quotes/futures/M6EZ26)

The policy mechanism still favors the dollar. The Fed's target range is 3.75%–4.00%, while the ECB deposit rate is 2.50%, a midpoint gap of 137.5 basis points. Yet markets already assigned roughly a two-thirds probability to another October Fed hike, and CFTC non-commercial euro positioning was net short 52,334 contracts on September 22. The direction is plausible; it is not contrarian. [Federal Reserve](https://www.federalreserve.gov/newsevents/pressreleases/monetary20260916a.htm), [ECB](https://www.ecb.europa.eu/press/pr/date/2026/html/ecb.mp260910~314e508016.en.html), [CFTC FX positioning](https://www.cftc.gov/dea/futures/deacmelf.htm)

A fresh 30%/40%/30% tree at 1.1150/1.1300/1.1550 gives about **+$900 forward net** for eight micros after an assumed $40 exit reserve. That supports the existing starter, not an addition. Softer U.S. inflation, a hawkish ECB surprise or durable geopolitical de-escalation could squeeze the crowded short.

**Decision: Hold −8 `M6EZ26`; no add.** Keep the 1.1550 stop, 1.1150 target and November 30 exit.

## 3. Soybean oil over meal: a better mark, but not better confirmation

Delayed standard proxies marked March soybean oil at **68.22¢/lb** and meal at **$364.90/short ton**. For the portfolio's basket definition,

`V = 60 × oil − 10 × meal = 444.20`,

versus 416.80 at entry. Across three pairs, that is **+$82.20 gross**. Crucially, the improvement came from weaker meal, not a fresh oil-demand release.

The structural case remains intact. USDA projects 2026/27 soybean-oil biofuel use at **17.8 billion pounds**, up 3.0 billion year over year. EIA reported actual June soybean-oil biofuel consumption of 1.556 billion pounds, versus 1.434 billion in May and 1.043 billion in June 2025. More crushing creates meal as a byproduct, while oil receives the stronger policy-driven demand pull. [USDA Oil Crops Outlook](https://www.ers.usda.gov/media/29635/ocs-26i.pdf?v=83674), [EIA soybean-oil use](https://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?f=M&n=PET&s=M_EPOOBDSO_YIFBP_NUS_MMLB)

The counter-case is meaningful. September 22 managed-money positioning showed oil net long 92,137 contracts but falling, while meal net length increased to 191,087. Meal exports can absorb supply, and March's static curve carries adversely versus nearby contracts. [CFTC agriculture positioning](https://www.cftc.gov/dea/futures/ag_sf.htm)

For the next 72 hours, judgmental 25% bull, 55% base and 20% bear outcomes at 69.50/355, 68.50/363 and 66.50/375 produce only **+$38.40 expected value** after an assumed $30 exit reserve. The edge disappears if roughly 3.4 probability points move from bull to bear.

**Decision: Hold the +3 `MZLH27` / −3 `MZMH27` package; no add.** Keep the combined basket stop and January 27 time exit. September 30 Grain Stocks and the EIA biofuel update can materially reprice both legs.

## 4. Challenger scan: no stronger replacement

| Market | Current assessment | Decision |
|---|---|---|
| WTI crude | November around $93.84 versus March near $83.03; extreme backwardation already prices scarcity | **Avoid chasing** |
| U.S. Treasuries | Sep. 25 cash curve: 2Y 4.81%, 10Y 5.17%; tightening and term premium are visible | **Avoid new outright short** |
| AUD | RBA decision imminent and a hike is heavily priced | **Wait through decision** |
| U.S. equity indices | Resilient earnings compete with 5%+ long yields | **Wait; no clean asymmetry** |
| Gold | Defensive demand competes with high real yields and contango | **Wait** |
| Corn / wheat | Report risk and crowded positioning weaken fresh entries | **Wait** |
| Live cattle | Tight supply is interesting, but contract size and recent rebound reduce entry quality | **Watch, not enter** |

WTI is the clearest rejected challenger. Delayed benchmark quotes showed November at $93.84 and March at $83.03—a **$10.81 backwardation**. Middle East exports are recovering but remain below prewar levels, while diplomacy is unresolved. An illustrative March-micro tree produced only about +$102 expected value against roughly $823 of risk to a $75 stop. The curve pays for prompt scarcity already; neither a chase-long nor a geopolitical short meets the bar. [WTI curve](https://ca.investing.com/commodities/crude-oil-contracts), [physical flows](https://www.reuters.com/business/energy/mideast-oil-exports-rebound-september-saudi-arabia-boosts-shipments-2026-09-28/)

## 5. Next 72 hours

- **September 28, 4:00 p.m. Toronto:** USDA Crop Progress.
- **September 29, 12:30 a.m.:** RBA decision; 1:30 a.m. press conference.
- **September 29, 8:30 a.m.:** Canada GDP.
- **September 29, 10:00 a.m.:** U.S. JOLTS and consumer confidence.
- **September 29, 9:30 p.m.:** China PMIs—the most important near-term demand test for copper.
- **September 30, 8:30 a.m.:** U.S. PCE, income/outlays and third-estimate Q2 GDP.
- **September 30, noon:** USDA Grain Stocks and Small Grains.
- **September 30, 7:50 p.m.:** Bank of Japan September summary of opinions.

## Bottom line

Copper's decline is more informative than the strike headline. The long still has positive modeled upside, but it now sits close to invalidation and is owned by a crowded speculative cohort. **Hold without adding and let 6.60 decide.** The euro short remains the cleanest of the three active mechanisms, while the oil/meal package is a small structural diversifier whose near-term edge is too thin to increase before September 30 data.

All executions and results are simulated. Proxy prices are labeled as proxies, stops are not guaranteed fills, and no real order was placed.