1. Bottom Line & Directional Bias
We are LONG copper (HG=F), invalidated by a daily close below the S1 pivot at 6.6997 (settle basis). Three reasons. First, supply risk is live: unions say the Centinela strike could cut November copper output by half (SMM, 2026-10-13), and Ivanhoe's Q3 Kamoa-Kakula saleable output of 76,401 mt is trending toward the lower end of guidance, while LME warehouse stock fell 11,625 MT over the last 20 reports to 231,275 MT [2026-10-12] and COMEX registered stock dropped 5,700 short ton (-1.2%) to 470,200 [2026-10-09]. Second, the technicals support the upside: the settle of 6.726 [2026-10-13] sits 60.1% up the 20-day channel of 6.4215–6.9285, with 5D +1.15% and 20D +4.38% (settle), and the last completed weekly bar (2026-10-05–2026-10-09) closed at 6.692, +2.18% w/w. Third, macro transmits positively: DXY 102.2 and the 10-year at 5.22% [2026-10-13] are both softer on the day, and the copper/gold ratio at 1.61 stands at the 99.6% 1Y percentile, a pro-growth signal. Positioning is not yet crowded (netPct 20.11%, 61.66 percentile), leaving room for trend followers to add. In early Asian trade on 2026-10-14 (as of 08:00), HG=F last printed 6.7185, -0.11% versus the prior settle, holding above S1. A daily close below 6.6997 would negate the call; upside targets are R1 6.7887 then R2 6.8303.
2. Price Action & Technical Analysis
The prior session settle was 6.726 [2026-10-13], down 0.31% on the day (settle), but the broader tape is firm: 5D +1.15% and 20D +4.38% (settle). ATR14 is 0.1146, about 1.7% of price as the full expected daily range, and RV20 is 17.6%, so realized moves remain contained relative to the option market's event pricing. The 20-day channel runs 6.4215–6.9285 with the settle at the 60.1% position, i.e., in the upper half but not extended; the 52-week range is 4.9045–6.96, so price is near the top of the annual band. The classic pivot ladder from 2026-10-12 H/L/C governs the 2026-10-13 session: P 6.7413, R1 6.7887, R2 6.8303, S1 6.6997, S2 6.6523. The settle of 6.726 sits just below pivot P, so the immediate test is 6.7413; a break above it targets R1 6.7887 (about 0.5×ATR beyond P), then R2 6.8303. On the downside, S1 6.6997 is the line that invalidates the long, with S2 6.6523 next. The last five settled bars show a recovery from the 10-08 washout (C 6.5685, L 6.5475) through 10-09 (C 6.692) and 10-12 (C 6.747, H 6.783) to 10-13 (C 6.726, H 6.778, L 6.703), a higher-low sequence. The last completed weekly bar (2026-10-05–2026-10-09) was O 6.573, H 6.763, L 6.5115, C 6.692, +2.18% w/w; the current week (from 2026-10-12, two sessions) is unfinished and last printed 6.726 (+0.51%), so no weekly-close conclusions are drawn. In early Asian trade on 2026-10-14 (as of 08:00), HG=F last printed 6.7185, -0.11% versus the prior settle, with an intraday high of 6.7335 and low of 6.7185, holding above S1 6.6997. The view: constructive while above 6.6997, with 6.7413 the first gate and 6.7887 the measured target.
3. Supply-Demand Balance & Fundamental Drivers
The physical balance is tightening at the margin. COMEX registered stock stood at 470,200 short ton [2026-10-09], down 5,700 short ton (-1.2%) d/d (as of 2026-10-09, prev 2026-10-08). LME warehouse stock was 231,275 MT [2026-10-12], down 1,750 MT (-0.75%) d/d (as of 2026-10-12, prev 2026-10-09), with a 20-report change of -11,625 MT, a persistent drawdown. SHFE warrant stock was 9,407 MT [2026-10-13], up 622 MT (+7.08%) d/d (as of 2026-10-13, prev 2026-10-12), the only counter-tick and a small one in absolute terms. The headline supply risk is the Centinela strike, which unions say could cut November copper output by half (SMM, 2026-10-13); against already-falling exchange inventories, even a partial disruption tightens the prompt market. Ivanhoe reported 76,401 mt of Q3 copper in saleable products at Kamoa-Kakula, with guidance trending toward the lower end, reinforcing the supply-side caution. On the demand side, Magna Mining approved the Levack mine restart targeting mid-2028 commercial production, a medium-term supply addition that does nothing for the 2026 balance, and Eldorado's Skouries mine won EU strategic status, again a future-tonnage story. Macro transmits positively: DXY 102.2 and the 10-year at 5.22% [2026-10-13] are both softer on the day, easing the financial-conditions drag on a USD-priced industrial metal, and today's China prints (CPI, PPI, trade, M2, new loans) plus US CPI and retail sales are the near-term catalysts. The term structure is in contango (M1-M2 -0.0305, -0.45%, RY -5.42%), a roll cost for longs rather than a price cap. The view: supply disruption risk plus multi-exchange draws support the long while SHFE's small build stays isolated.
4. Positioning & Fund Flows
CFTC data show managed money still net long but trimming. For the week of 2026-10-06, open interest was 304,033, longs 76,836, shorts 15,702, net 61,134, a weekly change of -16,924; the prior week (2026-09-29) was net 78,058 (Δ -4,464), 2026-09-22 net 82,522 (Δ +17,416), and 2026-09-15 net 65,106 (Δ -17,048). Net length has fallen for two consecutive weeks from the 2026-09-22 peak, yet price has risen over the same window (settle 6.726 [2026-10-13] vs 6.692 at the 2026-10-09 weekly close), so this is not a classic divergence — it is long liquidation into strength, which leaves headroom for re-entry rather than signaling exhaustion. Crowding is moderate: netPct (managed-money net ÷ OI) was 20.11% on 2026-10-06 at the 61.66 percentile of the 3-year window, down from 25.92% (68.51 percentile) on 2026-09-29 and 27.36% (69.05 percentile) on 2026-09-22; the trade is not crowded, and |z| is well inside the 1.5 extreme threshold. The CTA trend-following proxy is at 98 across all four weeks, i.e., trend followers are already maximally long, which cuts both ways: it confirms the uptrend but means incremental buying from that cohort is limited. Commercial hedgers sit at 56.9% (2026-10-06), roughly mid-range. On volatility, RV20 is 17.6% while the cross-asset implied-vol complex is calm (VIX 15.27, 1Y percentile 16%; GVZ 22.48, 1Y percentile 10%; OVX 48.52, 1Y percentile 41%), so optionality is not pricing a large copper-specific event despite the strike risk — a cheap way to express the long. The view: positioning is supportive, not stretched, with the CTA extreme the main caveat.
5. Cross-Asset Relative Value
The copper/gold ratio stands at 1.61 [2026-10-13], at the 99.6% 1Y percentile and 61.64% 3Y percentile: copper is extremely rich relative to gold over the past year, a pro-growth signal that corroborates the long but also flags how much good news is already in the relative price. The gold/silver ratio is 70.18 [2026-10-13], at the 82.54% 1Y percentile and 27.51% 3Y percentile — a high ratio means gold is relatively rich and silver is lagging, so the precious-metals complex is not confirming the risk-on impulse as strongly as copper is. The oil/gold ratio is 0.0225 [2026-10-13], at the 95.24% 1Y percentile and 50.53% 3Y percentile, indicating energy is also firm versus gold, consistent with a broad reflationary tape. The crack spread 3:2:1 is 65.94 USD/bbl [2026-10-13], a healthy refining margin that points to robust crude demand, another pro-cycle cross-check. DXY at 102.2 and the 10-year at 5.22% [2026-10-13] are both softer on the day, removing a near-term headwind for USD-priced copper. The term structure is in contango (M1-M2 -0.0305, -0.45%, RY -5.42%), a roll cost for longs rather than a signal of prompt tightness. The view: cross-asset ratios confirm a pro-growth regime that supports the long, with the copper/gold ratio's 99.6% 1Y percentile the main caution that relative upside may be narrowing.
6. Historical & Seasonal Patterns
The seasonality block covers the same calendar start, the next 20 sessions, over the last 15 years. The mean move is +0.67% but the median is -1.15%, with an up hit rate of only 7 of 15 years — a coin-flip record in which the average is flattered by outliers rather than confirmed by the typical year. The best sample is 2016 at +19.73% and the worst is 2015 at -8.41%, a spread that underscores how wide the distribution is; the sample is small and is context only, not a standalone signal. Read against the current setup, the tape is already 20D +4.38% (settle) into this window, so the historical median argues for expecting chop rather than a straight continuation, while the mean and the 2016 tail support keeping upside optionality. Practically, this argues for the disciplined invalidation at S1 6.6997 (settle basis) rather than a wider stop: in a 7-of-15 regime, the edge comes from cutting the -1.15% median-type years quickly and letting the +0.67% mean-type years run. The seasonal evidence neither overrides the supply-driven bullish call nor justifies adding on strength alone; it simply sets expectations that the next 20 sessions are as likely to retest as to extend, so entries should be staged and stops mechanical.
7. Scenario Analysis (Base / Bull / Bear)
Base case (55%) — Bullish grind to R1/R2. Trigger: HG=F holds above the S1 pivot at 6.6997 on a daily settle basis and clears the pivot P at 6.7413, with the Centinela strike unresolved and LME/COMEX draws continuing. Target: R1 6.7887 first, then R2 6.8303; a sustained break of R2 opens the 52-week high zone near 6.96. Action: hold the core long per the validated strategy, add only on a settle above 6.7413, and trail stops up under S1 as R1 is taken. This agrees with the section 1 call.
Bull case (25%) — Supply shock repricing. Trigger: confirmation that the Centinela strike is cutting November output materially (unions say by half), or a fresh exchange-stock drawdown beyond the LME 20-report -11,625 MT trend, pushing price through R2 6.8303. Target: the 52-week high at 6.96, then the 20-day channel top at 6.9285 is already exceeded en route. Action: scale into the long on the break of 6.8303, keep the stop at 6.6114/6.5999 per the validated strategy, and let the position run; do not chase beyond the 52-week high without a new catalyst.
Bear case (20%) — Strike resolution and macro miss. Trigger: a confirmed Centinela resolution without output loss (negates the supply-premium leg), or a daily close below S1 6.6997, or a China/US data surprise that lifts DXY from 102.2. Target: S2 6.6523, then the 20-day channel bottom at 6.4215. Action: exit the long on the 6.6997 daily close, stand aside, and re-enter only on a reclaim of 6.7413 with inventories still falling.
8. Trading Strategies & Risk Management
The validated strategy, quoted verbatim: “LONG entry 6.726 / stop 6.6114 / target 6.8979 / horizon 1-2w / R:R 1:1.5” and “LONG entry 6.726 / stop 6.5999 / target 6.9152 / horizon 1-5d / R:R 1:1.5.” Both entries reference the 2026-10-13 settle of 6.726; the first is the swing expression, the second the short-horizon expression, each at 1:1.5 reward-to-risk.
Stop placement: the 6.6114 stop sits 0.1146 below entry, exactly 1×ATR (ATR14 0.1146), i.e., one full expected daily range beneath the settle and below the S1 invalidation at 6.6997; the 6.5999 stop sits 0.1261 below entry, 1.1×ATR, giving the short-horizon trade slightly more room. Per the risk math, with a contract multiplier of 25,000, risk per contract is 2,895 USD (1×ATR variant) and 3,183 USD (1.1×ATR variant), including fees/slippage; a 10,000 USD risk budget caps size at 3 contracts in either variant. Do not exceed that cap.
Execution: enter at 6.726 or better on the next session; if price gaps through the entry, stand aside rather than chase. Scale out half at R1 6.7887 and the balance at the strategy target (6.8979 or 6.9152), moving the stop to breakeven once R1 is taken. Hard invalidation remains a daily close below 6.6997 (settle basis) or a confirmed Centinela resolution without output loss — exit regardless of the stop level. Stand-aside conditions: a settle below 6.6997, or SHFE warrant builds extending beyond the 2026-10-13 +622 MT tick alongside flat LME/COMEX. Re-entry trigger: a daily settle back above P 6.7413 with LME warehouse stock still falling.
9. This Week's Data Calendar
All times BJT | ET; a print beyond forecast ± threshold counts as a surprise.
Wed 2026-10-14: BJT 09:30 | ET 10-13 21:30 — China CPI y/y (F 1.0%, P 0.8%, surprise if outside ±0.2%) and PPI y/y (F 4.4%, P 3.8%, ±0.6%); BJT 10:03–11:04 | ET 22:03–23:04 — China trade balance (F 790B, P 809B, ±19B) and USD-denominated trade balance (F 116.8B, P 119.1B, ±2.3B), repeated across four timestamps; BJT 16:03 | ET 04:03 — M2 money supply y/y (F 7.4%, P 7.5%, ±0.1%) and new loans (F 1025B, P 60B, ±965B); BJT 20:30 | ET 08:30 — US core CPI m/m (F 0.2%, P 0.3%, ±0.1%), core CPI y/y (F 2.5%, P 2.4%, ±0.1%), CPI m/m (F 0.6%, P 0.4%, ±0.2%), CPI y/y (F 3.6%, P 3.4%, ±0.2%).
Thu 2026-10-15: BJT 20:30 | ET 08:30 — US core PPI m/m (F 0.3%, P 0.2%, ±0.1%), core retail sales m/m (F 0.6%, P 1.4%, ±0.8%), PPI m/m (F 0.6%, P 0.4%, ±0.2%), retail sales m/m (F 0.3%, P 1.2%, ±0.9%). China prints map to HG; US prints to GC, SI, DXY.
This report is generated automatically from public quantitative and macro data for research and market tracking only. It does not constitute investment advice or a recommendation to trade.