# Growth surged, rates repriced—and copper stayed pending

**Research cutoff: September 23, 2026, 12:05 p.m. Toronto.** Exact-contract observations carry their own timestamps. Delayed prices are research marks, not executable quotes; asynchronous spread legs are diagnostic only. This is a simulated portfolio, not a brokerage account or personalized investment advice.

## Decision first

**Remain flat. Keep the existing conditional `+3 MHGZ26` Micro Copper instruction pending, but fill zero contracts.** Fresh timestamped `MHGZ26` chronology was unavailable. Standard December copper `HGZ26` traded at **$6.7795/lb at 11:22 a.m. Toronto**, inside the micro plan's price zone, but it is a different contract and cannot prove the required exact-micro sequence: an `MHGZ26` trade at $6.78–$6.80 followed by a genuinely later reclaim of $6.80. No proxy fill, cancellation or size increase was applied. [Standard-copper corroboration](https://www.wsj.com/market-data/quotes/futures/HGZ26/contracts)

The paper portfolio remains **flat with NAV and cash of $999,916**. Realized gross P&L is −$77, commissions are $7, modeled slippage of $2 is embedded in fills, and realized net P&L is **−$84**. There are no positions, new fills, margin use or open risk. One conditional instruction remains pending until its original **September 26, 12:12:50 a.m. Toronto** expiry unless its pre-entry invalidation occurs first.

The retained plan is unchanged: three December 2026 Micro Copper contracts after the exact trigger, a $6.60 stop, $7.20 target and November 13 mandatory exit. The portfolio's conservative $6,000 paper margin reserve per micro is an assumption, not an exchange initial-margin quote. No automatic roll is authorized.

## The midday regime shock

U.S. flash business activity accelerated rather than cooled. The September composite PMI rose to **58.4** from 56.0, services reached **58.7**, and manufacturing rose to **57.0**. Manufacturing input prices jumped to **66.4**, the highest since October 2022, while new orders reached 58.2. The combination raises the probability of resilient nominal growth with persistent inflation—not a clean growth signal for risk assets. [S&P Global data reported by Reuters](https://www.reuters.com/business/us-business-activity-more-than-five-year-high-inflation-pressures-building-2026-09-23/)

Fed Governor Michael Barr reinforced that interpretation: he described growth as strong and employment as solid, judged inflation risk greater, supported the latest rate increase and said further policy adjustments would likely be needed. [Federal Reserve](https://www.federalreserve.gov/newsevents/speech/barr20260923a.htm)

By about 11:04 a.m. Toronto, the two-year Treasury yield was **4.862%**, up 8.49 basis points, and the ten-year was **5.054%**, up 8.7 basis points. October hike odds rose to 73% from 53% earlier in the day. EUR/USD fell to 1.1389 and USD/JPY rose to 158.25. These are reported cash-market observations, not futures execution inputs. [Reuters market report](https://app.bigdata.com/documents/F1625ED007038BAE157E5C7F151FBAEE)

The four-month regime weights now assign 46% to persistent inflation/friction, 31% to resilient growth/normalization, 12% to renewed supply shock, 8% to growth break and 3% to AI/liquidity disappointment. These are analyst judgments, not market-implied probabilities.

## Copper: positive physical evidence, thinner payoff

Chinese physical data still support the long thesis. National copper inventories fell **14,500 tonnes to 74,800 tonnes** week over week, while the copper-cathode-rod operating rate was expected to rise 2.3 percentage points to 68.12% as pre-holiday demand improved. The signal is constructive, though it is not an exact futures trigger. [Shanghai Metals Market](https://news.metal.com/newscontent/104126401-national-copper-inventories-continue-to-decline-pre-holiday-stockpiling-supports-demand-recovery-smm-weekly-data)

At a hypothetical first qualifying `MHGZ26` reclaim at **$6.8005**, the post-PMI scenario tree is:

| Conditional path | Weight | Terminal price | Three-contract P&L after $22.80 estimated friction |
|---|---:|---:|---:|
| Bull: tight regional supply meets resilient demand | 25% | $7.20 | +$2,973.45 |
| Base: support holds but higher rates cap the move | 35% | $6.90 | +$723.45 |
| Bear: demand/rates shock reaches the stop | 40% | $6.60 | −$1,526.55 |
| **Weighted** | **100%** | — | **+$385.95** |

Three micros represent 7,500 pounds. The raw stop loss from $6.8005 to $6.60 is $1,503.75; the 1.25 buffer plus $28.50 risk-accounting friction gives approximately **$1,908.19**, or **0.1908% of NAV**. The conditional target-to-buffered-risk ratio is about **1.56:1**. Every extra cent paid costs $75 across the package, and the scenario EV becomes negative above approximately **$6.852**.

This is now a fragile positive idea. Moving ten probability points from the bull case to the bear case lowers EV by $450 to **−$64.05**. That sensitivity is why the exact trigger remains decisive and why a standard-contract print cannot be substituted.

Carry is adverse for the long: later copper contracts remain above December. Exact synchronized micro spread quotes were unavailable, so no executable roll cost is claimed. The plan exits by November 13, ahead of the December contract's November 25 last trade, and does not roll. [CME Micro Copper calendar](https://www.cmegroup.com/markets/metals/base/micro-copper.calendar.html)

## Gold: lower price still does not repair the long

Exact `MGCZ26` was **$4,330.40/oz at 10:56 a.m. Toronto**, versus the $4,399.30 simulated exit. One Micro Gold contract is 10 ounces, so the completed exit avoided **$689** of subsequent gross downside. That is a counterfactual diagnostic, not realized profit; realized net portfolio P&L remains −$84. [Exact Micro Gold quote](https://www.wsj.com/market-data/quotes/futures/MGCZ26)

The decision remains **Avoid immediate re-entry**. A post-PMI tree using 20% at $4,800, 35% at $4,400 and 45% at $3,950 has an expected terminal price of $4,277.50. After an indicative $36.70/oz roll debit and $24 assumed round-trip costs, expected value is approximately **−$920 per micro**, or −2.12% of outright notional. Central-bank and ETF demand remain supportive, but crowded positioning, soft Asian physical demand, contango and higher rates outweigh the improved spot valuation. The negative tree is a rejection of the long, not a short recommendation.

## Oil inventory build weakens the long—but does not justify a short

For the week ended September 18, EIA reported U.S. commercial crude inventories up **3.0 million barrels to 426.4 million**, versus an expected 0.6-million-barrel decline. Cushing stocks rose 2.3 million barrels to 23.7 million, refinery runs fell 519,000 barrels per day and utilization fell 2.8 percentage points to 94.0%. Gasoline inventories fell 1.7 million barrels and distillates fell 0.4 million, so the product side was less bearish than the crude build. [EIA Weekly Petroleum Status Report](https://ir.eia.gov/wpsr/overview.pdf)

Exact standard `CLZ26` was **$88.96/bbl at 11:19 a.m. Toronto**, with an $86.53–$89.55 range. **Avoid the December crude long.** The inventory build and export decline weaken it, but product draws, backwardation and geopolitical supply risk make a new short unattractive. Exact `MCLZ26` was unavailable, so no micro execution is inferred.

## Cross-asset board

| Market | Current evidence | Decision and paper status |
|---|---|---|
| `MHGZ26` Micro Copper | Fresh exact chronology unavailable; prior exact micro 6.8075 at 2:21:19 a.m. Standard `HGZ26` 6.7795 at 11:22 a.m. is corroboration only. | **Wait.** Keep the existing +3 conditional plan pending; zero filled. Require the exact micro test and later reclaim. |
| `MESZ26` Micro S&P | **7,778.75 at 12:01:56 p.m.**, bid 7,778.50 × 33 / ask 7,778.75 × 89, down 0.68%; volume 553,320. | **Avoid immediate short chase; no order.** Reassess only after a failed 7,790–7,800 rebound or at the close. |
| `MNQZ26` Micro Nasdaq | **30,735.00 at 12:02:14 p.m.**, bid 30,735.00 × 8 / ask 30,735.25 × 1; volume 1,565,016. | **Avoid chase.** The same rate shock applies, but no distinct stop-defined edge beat the S&P illustration. |
| `ZNZ26` 10-year note | **105-08 at 12:01:45 p.m.**, bid 105-07½ / ask 105-08; volume 1,689,381. | **Avoid duration catch.** Strong activity, input inflation and Barr's guidance oppose a premature long; available two-year micro liquidity was inadequate. |
| `CLZ26` WTI | **$88.96 at 11:19 a.m.**; EIA crude +3.0 million barrels, product stocks lower. | **Avoid long and short; no order.** Mixed physical evidence and unavailable exact micro execution. |
| `NGF27−NGH27` gas | 3.853 at 11:46 versus 2.848 at 11:47; asynchronous premium **$1.005/MMBtu**. | **Wait / no order.** Reassess with synchronized legs after September 24 storage. |
| `ZCZ26` corn | **528.75¢ at 10:49 a.m.**, volume 95,250. | **Avoid.** The old 525¢ gate expired. One standard contract to the historical 500¢ reference risks about $1,797 buffered before costs, but upside to a 535¢ base is only $312.50 gross. Re-underwrite after export sales. |

## Best rejected opportunity: a post-PMI S&P short

The cleanest challenger is a one-contract `MESZ26` short only after a failed rebound into **7,790–7,800**, not at 7,778.75 after the selloff. An illustrative entry at 7,795, stop at 7,810 and target at 7,765 risks $75 gross; reserving $10 for friction gives $85 risk and $140 net target reward. At a judgmental 45% probability that the target is reached before the stop, EV is **+$16.25**; at 30%, EV is **−$17.50**.

This is rejected rather than staged. The probability is not empirically calibrated, the margin assumption was not verified, and chasing a 0.68% decline creates poor reversal risk. The decisive question is observable: does a rebound into 7,790–7,800 fail before the close? If not, retire the intraday expression.

## What changed since 6:05 a.m.

- U.S. growth surprised sharply higher while input-price pressure accelerated; Fed and bond-market reactions raised the discount-rate hurdle for metals and equities.
- Copper's physical evidence improved, but the scenario edge became more sensitive. The exact micro trigger was not verified and the pending instruction did not fill.
- Exact Micro Gold fell to $4,330.40. That validates the tactical exit without creating portfolio profit or a new entry.
- EIA's crude build weakened the oil long; product draws kept the short from clearing its own hurdle.
- The old March S&P pullback-long thesis is retired. The current December short illustration is deliberately not an order.

## Next 72 hours

- **September 23 close:** determine whether the equity selloff extends or the failed-rebound short premise expires. This is observation, not an automatic trade.
- **September 24, 8:30 a.m. Toronto:** initial jobless claims and USDA export sales. The latter is the next decisive corn check. [USDA export-sales schedule](https://fas.usda.gov/report-release-announcement/weekly-export-sales-300)
- **September 24, 10:00 a.m.:** U.S. new-home sales; **10:30 a.m.:** EIA natural-gas storage. The gas spread requires synchronized exact legs after the release. [EIA natural gas](https://ir.eia.gov/ngs/schedule.html)
- **September 25:** durable-goods orders, final Michigan sentiment and inflation expectations, followed by CFTC positioning at 3:30 p.m. [Census release calendar](https://www.census.gov/economic-indicators/calendar-listview.html), [CFTC schedule](https://www.cftc.gov/MarketReports/CommitmentsofTraders/ReleaseSchedule/index.htm)

The research horizon remains January 23, 2027. Contract clocks override that horizon: the only pending expression expires September 26 if it does not qualify and, if filled, exits by November 13 with no automatic roll.