1. Bottom Line & Directional Bias
We are LONG COMEX December copper (HGZ26), invalidated on a daily settle below the 20-day low of 6.509 [settle 2026-10-15]. Three reasons support the call. First, the sell-off is a channel-bottom retracement, not a trend break: the 6.5565 settle sits at the 11.3% position of the 20-day 6.509–6.9285 range, and the last completed weekly bar (2026-10-05 to 2026-10-09) closed at 6.692, up 2.18% w/w — the current week's -2.02% is an unfinished pullback inside that up-week. Second, physical balances are tightening at the exchanges that set the marginal price: LME warehouse stock is down 26,150 MT over 20 sessions to 228,000 MT as of 2026-10-14, and COMEX registered stock slipped 100 short tons to 471,200 as of 2026-10-13; the SHFE warrant surge to 34,933 MT as of 2026-10-15 is a small, single-exchange counter-signal. Third, positioning has de-risked: managed-money net length dropped 16,924 contracts in the week to 2026-10-06 to 61,134, netPct to 20.11%, crowding to the 62nd percentile — the long side is no longer crowded, and CTA trend signals remain at 98. The copper/gold ratio at 1.57 (92.46% 1Y percentile) shows copper still leading gold on a pro-growth basis. Near-term catalysts are China's 2026-10-19 IP/retail/FAI prints, 2026-10-20 GDP and the 1-y/5-y LPR decision. We add on holds above S1 6.5767 and cut on a settle under 6.509.
2. Price Action & Technical Analysis
COMEX December copper (HGZ26) settled at 6.5565 on 2026-10-15, down 1.56% on the day (settle), with the 5-day change at -0.18% and the 20-day change at -1.58% (settle-based snapshot). The metal trades in the lower half of its 20-day channel of 6.509–6.9285, at the 11.3% position — i.e., within roughly 0.7% of the 20-day low and about 5.4% below the 20-day high. ATR14 is 0.127, equal to 1.94% of price as the full expected daily range, and RV20 is 16.8% annualized from settles. The classic pivot ladder from the 2026-10-14 high/low/close governs the 2026-10-15 session: P 6.6828, R1 6.7667, R2 6.8728, S1 6.5767, S2 6.4928. The 2026-10-15 settle broke below S1 6.5767, leaving S2 6.4928 as the next structural support; a move below S2 would open the 20-day low at 6.509's neighborhood and then the 52-week low of 4.9045 as the deeper reference. The last completed weekly bar (2026-10-05 to 2026-10-09) opened 6.573, high 6.763, low 6.5115 and closed 6.692, up 2.18% w/w — a strong up-week whose high at 6.763 is the level to reclaim. The current week (from 2026-10-12, four sessions) is unfinished and last printed 6.5565, -2.02%; no weekly-close conclusion can be drawn from it. The last five settled bars show a clean lower-high sequence: 10-09 C 6.692, 10-12 C 6.747, 10-13 C 6.726, 10-14 C 6.6605, 10-15 C 6.5565, with the 10-15 low at 6.5245 already probing toward the 20-day low. In early Asian trade on 2026-10-16 as of 06:55, HG=F last printed 6.5601, up 0.05% versus the settle, with an intraday high of 6.5705 and low of 6.5601 — a stabilization, not yet a reversal. The 52-week range is 4.9045–6.96, so price remains in the upper third of the annual range. View: the tape is testing the bottom of the 20-day channel with momentum still negative; the long is valid only while the 20-day low at 6.509 holds on a daily settle basis, with R1 6.7667 as the first upside objective.
3. Supply-Demand Balance & Fundamental Drivers
The physical balance is tightening on the two exchanges that set the marginal price, with one small counter-signal. LME warehouse stock stood at 228,000 MT as of 2026-10-14, down 1,650 MT d/d (-0.72%) and down 26,150 MT over the 20-report window — a sustained draw that signals firm off-warrant demand and limited immediate supply. COMEX registered stock was 471,200 short tons as of 2026-10-13, down 100 short tons d/d (-0.02%) versus 2026-10-12 — essentially flat, but at a level that still reflects the large US-bound arbitrage stocks built during the tariff-threat period; SMM reported on 2026-10-15 that the tariff threat built over 1 million tonnes of US copper stocks but the arbitrage is fading, which is consistent with the flat-to-down registered count and removes a marginal source of visible supply pressure. The counter-signal is SHFE warrants at 34,933 MT as of 2026-10-15, up 11,476 MT d/d (+48.92%) from 23,457 MT on 2026-10-14 — a sharp single-day build, but from a very low base (9,407 MT on 2026-10-13), so it reads as post-holiday warrant registration rather than a structural surplus. On the demand side, SMM's 2026-10-15 weekly survey noted post-holiday restarts lifting processor operating rates while demand recovery remains uncertain, and India's base metals exports surged 29% in April–August with copper doubling (SMM, 2026-10-15) — a sign of growing non-Chinese supply that caps, but does not reverse, the drawdown narrative. On scarcity, Anglo American's Wanblad told the FT (2026-10-15) that tech giants are seeking copper exposure as scarcity bites, a medium-term demand tailwind from electrification and AI infrastructure. Macro transmission: the 10-year Treasury yield at 5.09% and DXY at 102 (both 2026-10-15, DXY -0.53% on the day) are a mixed backdrop — a softer dollar supports dollar-priced copper, while a 5% handle on yields raises the carry cost of inventory. The term structure is in contango with M1-M2 at -0.032 (-0.48%) and a roll yield of -5.77%, a roll cost for longs rather than a price signal. Seasonality over the next 20 sessions (15 years) shows a mean of +0.35% and median +0.71%, up 8 of 15 years — mildly supportive but a small sample. View: LME and COMEX draws plus fading US arbitrage outweigh the SHFE warrant build; the fundamental skew supports the long while the 20-day low holds.
4. Positioning & Fund Flows
Positioning has de-risked meaningfully, which is the core of the long case. In the CFTC week to 2026-10-06, open interest was 304,033 contracts, managed-money longs 76,836, shorts 15,702, net length 61,134 — a weekly change of -16,924 contracts, the second consecutive large cut after -4,464 in the week to 2026-09-29 (net 78,058) and following a +17,416 build in the week to 2026-09-22 (net 82,522). NetPct (managed-money net ÷ OI) fell to 20.11% from 25.92% and 27.36% in the prior two weeks, and the crowding percentile eased to 61.66 from 68.51 and 69.05 — the long side is no longer extended on the three-year window, so the trade is not crowded. The CTA trend-following proxy has held at 98 across all four weeks, meaning systematic trend capital is still fully long and any stabilization in price should find trend support; commercial hedging share was 56.9%, down from 58.27% on 2026-09-15, a modest reduction in producer hedging that is consistent with the LME draw. The price-versus-positioning read is not a divergence: price fell from the 6.747 settle on 2026-10-12 to 6.5565 on 2026-10-15 while the latest COT week shows net length also falling — both moving down together is liquidation, not a contrarian setup, and it clears room for fresh longs. On volatility, RV20 is 16.8% annualized from settles; the feed carries no copper implied-vol index, so the IV-versus-RV read is unavailable for HG specifically — the cross-asset vol block shows ^GVZ at 22.78 (1Y percentile 13%) and ^OVX at 48.62 (41%) as of 2026-10-14, and ^VIX at 16.46 (40%) on 2026-10-15, a calm macro vol backdrop. Retrospective risk metrics: 52-week max drawdown 13.49%, 20-day drawdown 4.36%, Sharpe30 -0.1775, VaR95 -2.75% — the negative 30-day Sharpe reflects the recent pullback and argues for sizing, not avoidance. View: positioning is clean, trend signals are intact, and the liquidation of the past two weeks creates entry room for the long above the 20-day low.
5. Cross-Asset Relative Value
The cross-asset snapshot as of 2026-10-15 keeps copper in the pro-growth leadership seat. The copper/gold ratio stands at 1.57, at the 92.46% 1-year percentile and 54.1% 3-year percentile — copper is rich relative to gold on a one-year view, which is a pro-growth signal and consistent with the long, though the 3-year percentile near the mid-range says the move is not yet a multi-year extreme. The related-spreads row shows HG_GC_RATIO at 1.62 (ratio×1000) with a 1-year percentile of 98.81% and 3-year percentile of 61.51%, reinforcing that copper's relative strength versus gold is at the top of its recent range. Gold/silver at 72.96 (1Y percentile 87.3%, 3Y percentile 29.23%) means the ratio is high — gold is relatively rich and silver is lagging — a late-cycle, risk-selective backdrop rather than a broad reflation, which argues for copper over silver in the metals sleeve. Oil/gold at 0.0235 (1Y percentile 98.02%, 3Y percentile 53.84%) shows crude also rich versus gold, and the 3:2:1 crack spread at 58.91 USD/bbl confirms healthy refining demand — both supportive of the global growth pulse that copper prices. DXY at 102.01 and the 10-year at 5.09% (both 2026-10-15) round out the macro frame: a slightly softer dollar on the day (-0.53%) is a tailwind for dollar-priced copper. View: the ratio complex is pro-growth and copper-led; the long is the right side of the relative-value trade while the copper/gold ratio holds above its 3-year mid-range and the 20-day low at 6.509 remains intact on a settle basis.
6. Historical & Seasonal Patterns
The seasonality block covers the same calendar start as the current window — the next 20 sessions — over the last 15 years. The mean move is +0.35% and the median is +0.71%, with the window up in 8 of 15 years, a 53% hit rate. The best year was 2016 at +19.17% and the worst was 2015 at -9.85%; the spread between those tails is wide, so the sample is small and the signal is directional rather than decisive. Read against the current setup, the median +0.71% applied to the 6.5565 settle [2026-10-15] points to roughly the 6.6 handle over the next month, which sits below the first pivot resistance at R1 6.7667 and well inside the 20-day channel of 6.509–6.9285. That is consistent with a retracement that recovers part of the recent -1.58% 20-day decline without immediately challenging the channel top. The positive skew of the median over the mean also indicates the typical year is better than the average year, with a few large negative prints dragging the mean down. We treat this as a mild tailwind, not a standalone trigger: the long is governed by the 20-day low at 6.509 on a daily settle basis, and the seasonal edge only adds conviction to entries taken above S1 6.5767. Positioning context reinforces the modest read — managed-money netPct at 20.11% and crowding at the 62nd percentile leave room for the seasonal pattern to play out without a crowded-long unwind.
7. Scenario Analysis (Base / Bull / Bear)
Base case — Bullish continuation (probability 55%). Trigger: price holds above the 20-day low of 6.509 on a daily settle basis and reclaims S1 6.5767, with the Asia 2026-10-16 06:55 print of 6.5601 already stabilizing above the invalidation. Target: R1 6.7667 first, then the 2026-10-12 settle of 6.747 as confirmation of the recovery leg. Action: hold the core long and add on settles above S1 6.5767, per section 1; the LME 20-report draw of 26,150 MT and the de-risked COT net of 61,134 support the skew.
Bull case — channel reclaim (probability 25%). Trigger: a daily settle back above the pivot P 6.6828, which would erase the 2026-10-15 break of S1 and retake the mid-channel. Target: R2 6.8728, then the 20-day high of 6.9285; a push through 6.9285 reopens the 52-week high at 6.96. Action: add to the long on the P 6.6828 reclaim, trail stops to S1 6.5767, and let the position run while the CTA trend proxy holds at 98.
Bear case — invalidation (probability 20%). Trigger: a daily settle below the 20-day low of 6.509, with a sustained break of S2 6.4928 confirming the failure. Target: S2 6.4928 first, then the 52-week low of 4.9045 as the deeper reference if the channel breaks. Action: exit the long on the 6.509 settle, do not re-enter until price reclaims S1 6.5767 on a daily settle; the SHFE warrant build of 11,476 MT on 2026-10-15 and the negative 30-day Sharpe of -0.1775 argue for discipline rather than averaging into a break.
Probabilities sum to 100%. The base case agrees with the section 1 call: Bullish, with the long valid only while 6.509 holds on a daily settle.
8. Trading Strategies & Risk Management
The validated strategy, quoted verbatim: LONG entry 6.5565 / stop 6.4167 / target 6.747 / horizon 1-5d / R:R 1:1.4 and LONG entry 6.58 / stop 6.45 / target 6.77 / horizon 1-2w / R:R 1:5. The first leg is a same-session entry at the 2026-10-15 settle of 6.5565; the second is a confirmation entry at 6.58, above the Asia 2026-10-16 06:55 intraday high of 6.5705, requiring the stabilization to hold. Stop placement sits below the 20-day low of 6.509 on both legs — 6.4167 and 6.45 respectively — so the stops clear the invalidation level and the S2 6.4928 structural support, avoiding a stop-run on a marginal settle. Per the risk math, the 6.5565 leg carries a stop distance of 0.1398, equal to 1.1×ATR (ATR14 0.127), with a contract multiplier of 25,000 and risk per contract of 3,525 USD including fees and slippage; a 10,000 USD risk budget caps size at 2 contracts. The 6.58 leg carries a stop distance of 0.13, equal to 1.02×ATR, risk per contract of 3,280 USD, and a 10,000 USD budget caps size at 3 contracts. Sizing rule: never exceed the per-leg contract cap; if both legs are live, aggregate risk stays within the 10,000 USD budget by prioritizing the 6.5565 leg first. Scaling: add only on daily settles above S1 6.5767, and trail the stop to entry once price reaches the first target. Stand-aside conditions: a daily settle below 6.509 invalidates the long and no new entries are taken until S1 6.5767 is reclaimed on a daily settle.
9. This Week's Data Calendar
All times BJT | ET. BJT 10-16 11:30 | ET 10-15 23:30 [USD/HIGH] Fed Chairman Warsh Speaks (F:-, P:-) → GC, SI, DXY. BJT 10-19 10:00 | ET 10-18 22:00 [CNY/HIGH] Fixed Asset Investment ytd/y (F:-, P:-) → HG, CL, ZS. BJT 10-19 10:00 | ET 10-18 22:00 [CNY/HIGH] Industrial Production y/y (F:-, P:-) → HG, CL, ZS. BJT 10-19 10:00 | ET 10-18 22:00 [CNY/HIGH] Retail Sales y/y (F:-, P:-) → HG, CL, ZS. BJT 10-19 10:00 | ET 10-18 22:00 [CNY/HIGH] Services Production Index y/y (F:-, P:-) → HG, CL, ZS. BJT 10-20 09:15 | ET 10-19 21:15 [CNY/HIGH] 1-y Loan Prime Rate (F:-, P:3%) → HG, CL, ZS. BJT 10-20 09:15 | ET 10-19 21:15 [CNY/HIGH] 5-y Loan Prime Rate (F:-, P:3.5%) → HG, CL, ZS. BJT 10-20 09:30 | ET 10-19 21:30 [CNY/HIGH] GDP y/y (F:-, P:-) → HG, CL, ZS. Surprise threshold is |forecast − previous|; a print beyond forecast ± threshold counts as a surprise and is the trigger to reassess the 6.509 invalidation level.
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.