1. Bottom Line & Directional Bias
We are long COMEX December gold (GCZ26), with invalidation on a daily close below 4091.2 — the 20-day low [settle 10-13]. Three reasons support the call. First, the correction looks exhausted: the settle of 4177.3 is only -0.23% over five sessions after a -3.59% 20-day drawdown, and the last completed weekly bar (2026-10-05 to 2026-10-09) closed at 4216.3, up 1.3% w/w, with the 4091.2 low holding on 10-07. Second, physical demand is absorbing the dip: GLD holdings rose 0.65% over 20 sessions to 33,893,188 oz [2026-10-13] even as managed-money futures length was cut for four straight weeks — ETF buying against speculative selling is a constructive divergence. Third, the entry is discounted: price sits at the 25th percentile of the 20-day channel (4091.2–4439.8), with pivot P at 4203.67 overhead and S1 at 4165.03 just below the settle. The risk is crowding: net-length percentile is 90.8, so a squeeze lower is possible, which is why the stop sits at 4000, roughly 2.3 ATR below the settle and beyond the 52-week low of 3901.3. Tonight's US CPI (Core y/y forecast 2.5%, surprise threshold ±0.1%) is the immediate event; a hot print is the fastest path to invalidation. Upside targets are R2 at 4267.87 and the 20-day high at 4439.8. View: long, invalidation 4091.2 daily close.
2. Price Action & Technical Analysis
Gold settled at 4107.9 on 2026-10-13, down 0.32% on the day (settle), with the 5-day change at -0.23% and the 20-day change at -3.59% (settle). The 20-day channel spans 4091.2 to 4439.8, and the settle sits at the 25th percentile of that range — a lower-third entry that keeps the long asymmetric while the 20-day high at 4439.8 defines the recovery target. ATR14 is 77.21, or 1.85% of price as the full expected daily range, and RV20 is 17.5%, so realized moves remain contained relative to the channel width. The pivot ladder from the 2026-10-12 H/L/C governs the 2026-10-13 session: P at 4203.67, R1 at 4229.23, S1 at 4165.03, R2 at 4267.87, S2 at 4139.47. The settle of 4107.9 sits between S1 and P, and the 10-13 session traded 4169.7 to 4224.8 — it probed above P intraday but failed to hold it, leaving the pivot as the first resistance to clear. In early Asian trade on 2026-10-14 (as of 08:00), the last price was 4177.7, up 0.01% versus the settle, with an intraday high of 4186.1 and low of 4177.7 (Asia) — a flat, narrow bar that does not alter the technical picture. The last five settled bars show the structure: 10-07 carved the 4091.2 low and closed 4140.7; 10-08 closed 4157; 10-09 rallied to 4216.3; 10-12 reached a high of 4242.3 but closed 4190.6; 10-13 closed 4107.9. Two sessions of lower closes after the 10-09 rebound is a mild negative, but the 4091.2 floor held and the 5-day change has flattened. The last completed weekly bar (2026-10-05 to 2026-10-09) closed at 4216.3, up 1.3% w/w, with a 4091.2 low; the current week is unfinished and no weekly-close conclusion is drawn from it. Seasonality over the next 20 sessions is mildly negative (mean -0.56%, median -1.38%, up 5 of 15 years), a headwind to size against rather than trade. View: long from the lower third of the channel; a break above P at 4203.67 targets R2 at 4267.87, while a daily close below 4091.2 invalidates.
3. Supply-Demand Balance & Fundamental Drivers
The physical balance is quiet but supportive at the margin. COMEX registered gold stood at 15.07 Moz (468,730 kg) as of 2026-10-09, unchanged d/d (0%) from 2026-10-08 — no delivery-side stress, but also no inventory build to pressure the front of the curve. SHFE warrant inventory was 116,028 kg as of 2026-10-13, also unchanged d/d, and flat across 10-09, 10-12 and 10-13, indicating stable Chinese warehouse stocks into the correction. The demand signal comes from ETFs: GLD holdings were 33,893,188 oz on 2026-10-13, up 0.65% from 33,675,496 oz twenty sessions earlier — roughly 217,692 oz of accumulation while the futures price fell 3.59% over the same 20-day window. ETF buying into weakness is the classic dip-demand signature and it offsets the speculative length reduction documented in the COT data. The term structure is in contango: M1-M2 at -19 (-0.45%) with a roll yield of -5.42% and a slope of 17.641. This contango reflects carry — the cost of financing and storing metal against a 5.22% 10-year yield — and is a roll cost for longs, not a price cap; it does not alter the directional call but argues for futures over leveraged ETF exposure for the carry-sensitive. Macro transmission is direct: the 10-year yield at 5.22% fell 0.44% and DXY at 102.22 fell 0.36% on 2026-10-13, both modest tailwinds for a non-yielding asset, and the cross-asset snapshot shows DXY at 102.2 and the 10Y at 5.22% [2026-10-13]. The week-ahead calendar is the dominant fundamental risk: US Core CPI y/y is forecast at 2.5% (prior 2.4%, surprise threshold ±0.1%) and CPI y/y at 3.6% (prior 3.4%) tonight, followed by Core PPI and retail sales on 10-15; a hot inflation print would push real yields and the dollar higher against gold. Chinese data this morning (CPI, PPI, trade, M2, new loans) carries a lower direct transmission to GC. View: stable inventories plus ETF accumulation support the long; the CPI print is the swing factor.
4. Positioning & Fund Flows
Positioning is the main risk to the long, though it is improving. Managed-money net length fell to 110,987 contracts as of 2026-10-06, down 9,331 w/w, after declines of 7,071 (09-29), 5,727 (09-22) and 1,856 (09-15) — four consecutive weekly cuts totaling roughly 22,129 contracts from the 133,116 peak on 09-15. Open interest also declined to 396,109 from 412,800 over the same window, so the length reduction is occurring alongside falling OI: price down and OI down is liquidation, not fresh shorting, which is a healthier correction profile than active selling. The crowding metric, however, remains elevated: netPct (managed-money net ÷ OI) was 28.02% on 2026-10-06, at the 91st percentile of its three-year range, down from 92.58 on 09-15 but still firmly in crowded territory — this is a crowded long, and the trade must be sized for squeeze risk. The CTA trend proxy is unchanged at 62 across all four weeks, indicating trend followers have not yet flipped; hedge fund net share is stable near 14.4%. On volatility, ^GVZ (gold implied vol) was 22.48 on 2026-10-12, down 0.31 pts and at the 10th percentile of its one-year range, versus RV20 of 17.5% — an IV−RV spread of +5.0 vol points (IV/RV 1.28), meaning options are pricing in event risk above what has been realized, so outright futures are preferred over premium-heavy option structures for the long. The four-week pattern of falling net length against a falling price is a mild positive divergence: weak hands are exiting while ETF holdings rise, and the 91st-percentile crowding is decaying rather than building. View: long, but respect the crowded tape; a further sharp OI drop with price below 4091.2 would signal the squeeze has begun.
5. Cross-Asset Relative Value
Cross-asset ratios as of 2026-10-13 frame gold as neither rich nor cheap in relative terms. The gold/silver ratio is 70.18, at the 83rd percentile over one year but the 28th percentile over three years — a high one-year reading means silver has lagged gold recently, while the lower three-year percentile reflects silver's longer-run structural strength; neither reading is a direct return forecast, but the elevated short-term ratio argues against expressing the bullish view through silver. The copper/gold ratio is 1.61, at the 100th percentile over one year and 61.64% over three years — a very high ratio means copper has outperformed gold sharply, consistent with a pro-growth tape that is currently a headwind for gold's relative appeal; it also means the ratio is extended, so a growth scare would reverse it in gold's favor. The oil/gold ratio is 0.0225, at the 95th percentile over one year and 50.53% over three years, indicating oil has been rich versus gold on a one-year view. The crack 3:2:1 spread at 65.94 USD/bbl is a refinery margin and not directly relevant to gold. The macro backdrop in the same snapshot — DXY at 102.2, 10Y at 5.22%, VIX at 15.27 — shows a firm dollar and yield with low equity vol, a neutral-to-negative mix for gold that the CPI print tonight can shift quickly. View: long gold outright rather than via silver; a hot CPI that lifts DXY above recent ranges is the key relative-value risk.
6. Historical & Seasonal Patterns
The seasonal window ahead — the next 20 sessions from the same calendar start — has been mildly negative over the last 15 years: mean -0.56%, median -1.38%, with a positive return in only 5 of 15 years (a 33% hit rate). The best year was 2022 at +7.57% and the worst was 2015 at -8.08%, a 15.65-point spread that underscores how wide the distribution is around a negative median. This is a small sample, so it is context rather than a signal: the median drag of roughly -1.4% is smaller than one ATR14 (77.21 points, 1.85% of price), meaning a single volatile session can erase the typical seasonal headwind. The pattern argues for sizing discipline, not for fading the bullish call — the 20-day channel (4091.2–4439.8) and the 4091.2 invalidation floor remain the operative risk framework, with seasonality treated as a modest statistical drag on the base case's expected path.
7. Scenario Analysis (Base / Bull / Bear)
Base case (60%) — Bullish, grind to R2. Trigger: Core CPI y/y prints within the 2.4%–2.6% band (forecast 2.5%, surprise threshold ±0.1%) and the settle holds above S1 at 4165.03, then clears pivot P at 4203.67. Target: R2 at 4267.87, with the 20-day high at 4439.8 as the extended objective. Action: hold the validated long from 4177.3; add only on a confirmed daily close above 4203.67, never into the CPI release itself.
Bull case (20%) — Inflation surprise to the downside. Trigger: Core CPI y/y at or below 2.4% (a downside surprise beyond the ±0.1% threshold) with Core CPI m/m at or below 0.1%, pressuring DXY below 102.2 and the 10Y below 5.22%. Target: a sustained break above R2 at 4267.87 opens the 20-day high at 4439.8, roughly 0.5×ATR beyond R2. Action: hold the full position and trail the stop from 4100 toward breakeven once 4267.87 is taken out.
Bear case (20%) — Hot CPI triggers invalidation. Trigger: Core CPI y/y above 2.6%, or a daily close below 4091.2 (the 20-day low, settle basis). Target: S2 at 4139.47 first, then the stop at 4100, with the 52-week low at 3901.3 the tail risk. Action: exit the long at the 4100 stop without discretion; do not re-enter until price reclaims 4203.67 on a daily close.
Probabilities sum to 100%. The base case agrees with the section 1 call: long, invalidation 4091.2 daily close.
8. Trading Strategies & Risk Management
The validated strategy, quoted verbatim: LONG entry 4177.3 / stop 4100 / target 4331.7 / horizon 1-2w / R:R 1:2.0, and LONG entry 4177.3 / stop 4100 / target 4293.3 / horizon 1-5d / R:R 1:1.5. The stop at 4100 sits 77.3 points below the 4177.3 entry — exactly 1×ATR (ATR14 77.207) — and is placed just under the 4091.2 invalidation floor, so the technical thesis and the risk stop fail together rather than the stop being clipped by ordinary range noise (ATR14 is 1.85% of price as a full daily range). Per the program's risk math, with a contract multiplier of 100 the risk per contract is 7,755 USD including fees and slippage, and a 10,000 USD risk budget caps size at 1 contract. Scaling: the 1-5d leg targets 4293.3 (R:R 1:1.5) and can be closed at that level, leaving the runner for the 1-2w target of 4331.7 (R:R 2.0); do not average down, and do not add ahead of the 20:30 BJT CPI print. If price closes below 4091.2, exit immediately regardless of the stop fill. If the CPI print is hot (Core y/y above 2.6%), stand aside entirely; re-entry requires a daily close back above pivot P at 4203.67, with the stop reset at 4100.
9. This Week's Data Calendar
All times BJT | ET. Tonight is the swing event: BJT 10-14 20:30 | ET 10-14 08:30 [USD/HIGH] — Core CPI m/m F:0.2% P:0.3% (surprise outside F±0.1%), Core CPI y/y F:2.5% P:2.4% (surprise outside F±0.1%), CPI m/m F:0.6% P:0.4% (±0.2%), CPI y/y F:3.6% P:3.4% (±0.2%) — all tagged GC, SI, DXY. Thursday BJT 10-15 20:30 | ET 10-15 08:30 [USD/HIGH]: 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%). Earlier today, China CPI y/y F:1.0% P:0.8%, PPI y/y F:4.4% P:3.8%, trade balance F:790B P:809B, M2 F:7.4% P:7.5% and new loans F:1025B P:60B (BJT 09:30–16:03) carry lower direct transmission to GC.
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.