Gold (GC=F) Deep Dive | 2026-10-02 — Bearish Below Pivot 4198, GCZ26 | 2026-10-02

Call: Bearish gold into year-end, with COMEX December gold (GCZ26) — invalidation on a settle back above the 20-day channel top at 4558.5, or a decisive reclaim of pivot P 4198.2 that holds for two consecutive settles. Three reasons. First, trend and momentum: settle 4202.3 sits at the 14th percentile of the 20-day 4143.1–4558.5 channel, 5D -2.23% and 20D -4.81%, with the last completed weekly bar (2026-09-21/25) closing at 4321.2, -2.34% w/w. Second, positioning: managed-money net is 127,389 lots, netPct 30.86% of open interest, a 92.46 three-year crowding percentile, and it has been cut four straight weeks — a long base still unwinding, not a washed-out low. Third, the macro transmission channel: DXY 102.04 (+0.58%) with 10-year yields at 5.24% keeps the carry cost of a non-yielding asset elevated, and the M1-M2 contango of -15.5 (-0.37%) is a roll cost for longs, not a floor. GVZ at 23.32 (16th 1Y percentile) versus RV20 20.4% means optionality is cheap relative to realized — favouring defined-risk shorts over outright size.

1. Bottom Line & Directional Bias

Call: Bearish gold (GC=F, referencing COMEX December gold GCZ26) into year-end. Invalidation: a settle above the 20-day channel top at 4558.5, or two consecutive settles back above pivot P 4198.2.

Three reasons drive the call. First, price structure: the 2026-10-01 settle of 4202.3 sits at the 14th percentile of the 20-day 4143.1–4558.5 channel, with 5D -2.23% and 20D -4.81% — a market that has lost its August impulse and is grinding along the lower band. Second, positioning: managed-money net length of 127,389 lots (netPct 30.86% of OI) still registers a 92.46 three-year crowding percentile even after four consecutive weekly reductions totalling roughly 9,400 lots; crowded longs that are still being reduced are a supply of sellers, not a contrarian buy signal. Third, the macro transmission: DXY at 102.04 (+0.58%) and the 10-year at 5.24% raise the opportunity cost of holding metal, while the curve's -15.5 M1-M2 contango is a roll cost for longs. Invalidation is a settle above 4558.5, which would restore the August range; a two-settle reclaim of 4198.2 would neutralise the tactical short.

2. Price Action & Technical Analysis

The 2026-10-01 CME final daily settlement was 4202.3, +0.37% on the day (settle). Over five sessions the contract is -2.23% and over twenty sessions -4.81% (settle). The 20-day channel runs 4143.1–4558.5, placing the settle at the 14th percentile — the lower quarter of the recent range. The 52-week range is 3823.7–5586.2, so the market is roughly 25% below the 52-week high and about 10% above the 52-week low.

ATR14 is 89.7, equal to 2.13% of price as a full daily range — not a half-range. RV20 is 20.4% annualised. In early Asian trade on 2026-10-02 (07:00), gold last printed 4208.5, +0.15% versus the prior settle, with an Asian session high of 4208.5 and low of 4201.1 (Asia) — a narrow, low-conviction tape ahead of the US payrolls print.

Pivots from the settle-based snapshot: P 4198.2, R1 4226.9, S1 4173.5, R2 4251.6, S2 4144.8. The settle at 4202.3 is only 4.1 points above P, so the pivot is the immediate decision line. A settle below S1 4173.5 opens S2 4144.8, which coincides almost exactly with the 20-day channel floor at 4143.1 — a confluence that makes 4143–4145 the first real downside objective. Conversely, R1 4226.9 and R2 4251.6 cap the topside; the 20-day channel top at 4558.5 is the structural invalidation.

On the weekly timeframe, the last completed bar (2026-09-21 to 2026-09-25) opened 4413, high 4422.1, low 4278.3 and closed 4321.2, -2.34% w/w. The current week, running from 2026-09-28, is not closed; its last mark is 4202.3, -2.75% week-to-date. No weekly-close conclusion can be drawn from an unfinished bar, but the direction of travel from the completed 4321.2 close to the current 4202.3 is unambiguously lower. Risk metrics reinforce the point: 20-day drawdown 8.18%, 52-week drawdown 25.06%, 30-day Sharpe -3.19, and VaR95 at -2.92% of a single day's loss.

View: bearish while below 4226.9; the first target zone is 4143–4173.

3. Supply-Demand Balance & Fundamental Drivers

COMEX registered gold stood at 15.14 Moz (470,907 kg) on 2026-09-29, up 1,244 kg day-on-day (+0.3%) from 15.1 Moz (469,663 kg) on 2026-09-28. That is a marginal build in deliverable inventory, not a drain — mildly negative for the prompt, though the daily change is small relative to the stock. SHFE warrants were 116,028 kg on 2026-09-30, down 3 kg day-on-day (-0.0%), essentially flat after 116,031 kg on 2026-09-29. The Chinese exchange stock is not signalling physical tightness; it is a static pool.

The term structure is in contango: M1-M2 at -15.5 (-0.37%), with roll yield of -4.44% and a slope of 22.17. Contango here reflects the cost of carry — with the 10-year Treasury at 5.24%, financing and storage make deferred contracts more expensive than spot. It is a roll cost for long holders and is not a price cap; it does not by itself force the market lower. But it does mean a long position bleeds roughly 4.4% annualised in roll, which raises the bar for holding length in a market already losing momentum.

The macro transmission is straightforward. DXY at 102.04, +0.58% on 2026-10-01, is the primary channel: a firmer dollar mechanically pressures dollar-denominated gold. The 10-year yield at 5.24%, down 1.06% on the day but still elevated in absolute terms, keeps the real-cost argument against non-yielding metal intact. The week-ahead calendar puts the heaviest weight on the 2026-10-02 US employment report (Non-Farm Employment Change forecast 89K versus prior 162K, surprise threshold ±73K; Average Hourly Earnings forecast 0.3% versus 0.3%, threshold ±0.1%; Unemployment Rate forecast 4.1% versus 4.1%, threshold ±0.1%). A hot payrolls print would reinforce the dollar and yields and validate the short; a very weak print is the main near-term risk to the call. ISM Services PMI on 2026-10-05 (forecast 54 versus prior 55.4, threshold ±1.4) and the FOMC minutes on 2026-10-08 are the secondary event risks.

View: inventory marginally additive to supply, contango a carry cost, dollar and yields the dominant driver — net bearish.

4. Positioning & Fund Flows

CFTC managed-money positioning has been shrinking for four consecutive weeks. Net length fell from 136,771 lots on 2026-09-01 to 134,972 on 2026-09-08, 133,116 on 2026-09-15 and 127,389 on 2026-09-22 — a cumulative reduction of roughly 9,382 lots. Open interest over the same window moved from 415,196 to 412,800. The weekly deltas were -7,976, -1,799, -1,856 and -5,727 lots respectively. The critical point is that this de-risking has coincided with falling price: the settle has declined from the 4413 weekly open area toward 4202.3. Longs are being reduced into weakness, which is trend-confirming rather than capitulative.

Crowding remains elevated. NetPct — managed-money net divided by open interest — was 30.86% on 2026-09-22, with a three-year crowding percentile of 92.46. That is a high reading on the multi-year window, so the trade is still crowded long even after four weeks of trimming. The CTA trend-following proxy sits at 62, and the hedge ratio at 15.07%. A crowded long base that is still being liquidated is a persistent source of sell pressure; it is not yet the washed-out positioning that typically marks a durable low.

On the volatility side, GVZ at 23.32 on 2026-10-01 (-0.42 points on the day) sits at the 16th percentile of its one-year range, against RV20 of 20.4%. That leaves IV minus RV at +2.9 vol points, an IV/RV ratio of 1.14. Implied vol is only modestly above realized and cheap in absolute percentile terms, which favours expressing the bearish view through defined-risk structures rather than paying up for outright downside.

View: crowded but still-unwinding longs, cheap optionality — bearish, expressed with defined risk.

5. Cross-Asset Relative Value

The copper/gold ratio stands at 1.58 (ratio ×1000), at the 96th percentile of its one-year range and the 53rd percentile of its three-year range. A high and elevated copper/gold ratio is a pro-growth signal — industrial demand is being priced more favourably than defensive metal — which is a relative headwind for gold.

The gold/silver ratio is 69.13, at the 73rd percentile of its one-year range but only the 24th percentile of its three-year range. The one-year reading is high, meaning silver has lagged gold over the past twelve months; the three-year reading is low, meaning silver has been structurally strong versus gold over the longer window. The two horizons disagree, so the ratio is not a clean directional signal for gold on its own — but the recent one-year elevation is consistent with gold having been the favoured leg, and any rotation out of that preference would weigh on gold specifically.

The oil/gold ratio is 0.0216, at the 87th one-year percentile and 45th three-year percentile. A high oil/gold ratio means energy has been relatively strong versus gold, again a pro-cyclical tilt rather than a defensive-metals bid. WTI implied vol (OVX) at 51.69 (51st one-year percentile) is far above gold's GVZ at 23.32 (16th percentile), underscoring that the macro volatility premium is in energy, not in gold. VIX at 16.39 (35th percentile) shows no broad risk-off impulse that would typically drive haven demand for bullion.

View: cross-asset configuration is pro-growth and anti-haven — a relative-value headwind for gold.

6. Historical & Seasonal Patterns

Using the same calendar start and the next 20 sessions over the last 15 years, gold has averaged +1.15% with a median of +0.39%, rising in 9 of 15 years. The best instance was 2023 at +9.08% and the worst was 2022 at -3.37%. The sample is small and the dispersion is wide — the gap between the best and worst outcomes is more than 12 percentage points — so the seasonal edge is weak and should be treated as context only, not as a driver.

Notably, the median of +0.39% is far below the mean of +1.15%, indicating that the average is pulled up by a small number of strong years rather than a broad, reliable pattern. A 9-of-15 hit rate is barely better than a coin flip. Against a backdrop of negative 5D and 20D momentum, elevated crowding and a firm dollar, the seasonal tailwind is not sufficient to override the tactical bearish case. Seasonality is a mild counterweight to the short, not a reason to abandon it.

View: seasonality is a weak, low-conviction positive — it does not change the bearish call.

7. Scenario Analysis (Base / Bull / Bear)

Base case — 55% probability: grind lower toward the 20-day floor. Trigger: the 2026-10-02 US employment report lands near or above the 89K forecast, keeping the dollar firm and the 10-year near 5.24%. Path: gold settles below pivot P 4198.2 and tests S1 4173.5, then the 4143–4145 confluence of S2 4144.8 and the 20-day channel floor 4143.1. Action: hold tactical shorts with a target at 4145, trail stops to the 4226.9 R1 area. This base case agrees with the section 1 call.

Bull case — 25% probability: short squeeze back into the range. Trigger: a materially weak payrolls print (well below the 89K forecast, beyond the ±73K surprise threshold) or a dovish FOMC minutes read on 2026-10-08, combined with a softer dollar. Path: a reclaim of P 4198.2 that holds, then R1 4226.9 and R2 4251.6, with an extension toward the 4321.2 completed weekly close. Action: cover shorts on a two-settle reclaim of 4198.2; do not add length until 4251.6 is settled above. Invalidation of the bearish call proper requires a settle above 4558.5.

Bear case — 20% probability: acceleration through the floor. Trigger: a hot payrolls print plus a firm ISM Services PMI on 2026-10-05 above 55.4, pushing DXY through 102.5 and the 10-year higher. Path: a settle below S2 4144.8 and the 20-day floor 4143.1 opens a vacuum toward the 52-week low at 3823.7, with the 4000 round number as an intermediate marker. Action: add to shorts on a confirmed settle below 4143.1, with stops back above 4198.2; size up only if GVZ remains below its one-year median, keeping optionality cheap.

View: the probability-weighted path is lower; the base case is a grind to 4145, with the bull case gated by the 4198.2 pivot.

8. Trading Strategies & Risk Management

Strategy 1 — Tactical short GCZ26 (COMEX December gold). Entry 4208–4215 on rallies into the Asian high, stop 4290 (above R2 4251.6 and roughly one ATR14 of 89.7 beyond entry), target 4145 (the S2 4144.8 / 20-day floor 4143.1 confluence), horizon 1–5 days, conviction 7 of 10. Size at half normal risk budget given the 2026-10-02 payrolls event inside the holding window; the stop sits beyond a real level and outside normal daily noise.

Strategy 2 — Defined-risk bearish structure into the FOMC minutes. With GVZ at 23.32 (16th one-year percentile) and IV/RV at 1.14, buy GCZ26 put spreads struck 4200/4100 expiring after 2026-10-08, risking no more than 0.5% of book. This caps loss at premium paid while retaining downside participation to the 4143 floor and beyond. Horizon 5–8 days, conviction 6 of 10. Do not add outright length in either strategy; both are aligned with the bearish call and are invalidated on a two-settle reclaim of 4198.2.

9. This Week's Data Calendar

BJT 10-02 20:30 | ET 10-02 08:30 — US Average Hourly Earnings m/m (F 0.3%, P 0.3%), Non-Farm Employment Change (F 89K, P 162K), Unemployment Rate (F 4.1%, P 4.1%); all HIGH impact on GC, SI, DXY. BJT 10-05 22:00 | ET 10-05 10:00 — ISM Services PMI SEP (F 54, P 55.4), HIGH. BJT 10-08 02:00 | ET 10-07 14:00 — FOMC Minutes, HIGH on GC, SI, DXY. Crude stock data on BJT 10-07 is MEDIUM and not a gold driver.

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.

单品种早报深度分析

COMEX黄金GC=F 2026-10-02

Gold (GC=F) Deep Dive | 2026-10-02 — Bearish Below Pivot 4198, GCZ26 | 2026-10-02

Gold (GC=F) Deep Dive | 2026-10-02 — Bearish Below Pivot 4198, GCZ26 | 2026-10-02

今日风险地图

单品种早报 深度分析

往期早报

⟠ 本月策略回测表现

单品种早报深度分析