Saturday review — start with the hardest fact: on Friday (October 23), spot gold settled at $3,995.38/oz, down 0.69% on the day. This is the first time since July 29 that bullion has closed below the $4,000 handle. The three-round battle is worth replaying: a vertical Asian-session flush to $3,986, a dip-buyer rally that reached $4,032.56, a fresh New York-session leg down toward $3,984, and a final hour in which the longs failed to lift price back above the line. Front-month COMEX futures settled at $3,991.10, down another 1.28%. The “$4,000 defense” lasted all day — and at the closing bell, the bears were standing inside the gate.
1. Why a Confirmed Close Is More Dangerous Than an Intraday Break
When Friday’s Asian session broke the level, we stressed in our breakdown of the triple hit and the round-number battle that an intraday pierce is only a “probe” — a daily/weekly close below the level is the “confirmation.” Berenberg framed it identically before the bell: a confirmed close below $4,000 would likely trigger further gold-ETF outflows and technical selling across the entire mining sector. That trigger condition is now met.
The weekly statistics capture the severity:
- COMEX gold fell 4.05% ($168.40) this week — the largest weekly decline since the week ending June 5. Two consecutive weekly losses, -4.77% over the fortnight, and seven down weeks out of the last nine. The settlement price now sits roughly 25% below the January 29 peak of $5,318.40.
- Mining stocks just printed their worst week in seven months, with BHP and Rio Tinto both down more than 8%. Equity markets are beginning to price a systematically lower gold center of gravity — a signal that historically leads the metal itself.
- The haven divergence is the most extreme since December 2008. Berenberg notes that the last time gold performed this poorly while Treasury yields fell this much was the depths of the global financial crisis. Thursday’s “haven capital choosing bonds over bullion” structure — detailed in our safe-haven failure transmission chain — repeated on Friday: the Dow dropped 1.23%, and zero-yield gold once again lost to a coupon-bearing Treasury.
The transmission reached domestic China in real time: the main SHFE gold contract touched an intraday low of 858.08 yuan/gram — a year-to-date low — before settling at 864; major jewelry brands including Chow Tai Fook and Lao Miao cut retail prices by 24-32 yuan/gram overnight, back toward 1,208 yuan/gram.
2. The Bull Case Institutions Are Still Holding
A confirmed break is not a one-way short signal — the divergence among major houses is precisely the fuel for next week’s volatility:
- Goldman Sachs maintains its end-2027 target of $5,400, while warning of trading-structure risk: prices are approaching key call-option strikes where derivatives positioning can amplify two-way swings.
- UBS Wealth Management expects a short-term retest of the $4,000 area, but sees “structural dip-buying demand forming a bottom,” with gold regaining momentum toward year-end and into 2027 as the market reprices Fed hikes lower — consistent with the allocator absorption we tracked in ETF flows.
- Nanhua Futures defines $4,000 as the “core defense line” but argues the market has now fully priced its hawkish scenario (three or more additional hikes by the end of next year), making this pullback a dip-buying opportunity.
- CME Group chief economist Erik Norland offers the coolest summary: the biggest driver of this year’s precious-metals decline is the systematic repricing from “rate cuts” to “rate hikes” — yet gold remains well above its early-2025 levels, leaving long-term holders’ cushion intact.
3. The Super-Week Calendar: Three Nodes, Beijing Time
Next week carries the highest information density of this entire correction. Three nodes, in order:
- Thursday, October 29, 02:00 — the FOMC decision. An October hold is the base case (~84% priced); the real focus is the statement language. If it stresses “higher for longer” without cooling December hike expectations, any “sell-the-news exhaustion” bounce in gold will be short-lived.
- The PCE inflation report. The Fed’s preferred gauge directly determines whether the 86%-91% December hike pricing hardens or unwinds. Friday’s flash composite PMI of 57.6 — still elevated — gives the Fed plenty of cover to stay hawkish.
- Friday, October 30 — month-end rebalancing. Gold will almost certainly print a long red monthly candle, and month-end pension/index rebalancing tends to amplify late-session one-way moves — historically, trend continuation in the final two hours of month-end runs well above that of an ordinary session.
Do not lose sight of the background variable threading through the whole week: Brent crude rose another 3.14% on Friday to settle at $116.72, with Hormuz-strait tensions lifting the supply premium. As long as the oil-to-inflation-to-hike-expectations chain stays tight, the “rate ceiling” above gold stays in place.
4. The Quant Playbook: Four Rules for Event Week
The cardinal sin of a super week is running normal-week position sizing and stop parameters into it. My system enforces four rules:
- Calendar-based sizing. Monday and Tuesday: probe rallies and flushes with no more than 30% of normal size. Wednesday into the New York close (pre-decision): cut to 0-20% and hand directional choice back to the market. After Thursday’s decision prints, scale back in according to the scenario plan. Carrying full size into the statement is the number-one cause of event-week blowups.
- Mechanical levels. Supports: $3,960 (wedge measured-move extension) and $3,900 (the August platform). First overhead supply: $4,030-$4,040 — Friday’s rebound high of $4,032.56 landed precisely in this breakdown-retest zone, validating it. The trend-level bull/bear line is $4,135; until it is reclaimed, every rally is classified as a “post-breakdown technical correction.”
- Volatility adaptation. ATR always expands in event weeks. Widen stops to 1.5x ATR while cutting order size in half, keeping total risk exposure constant. An EA whose parameters do not adapt to the volatility regime will get stop-hunted on both sides on FOMC night.
- Execution quality first. Spreads and slippage are the largest hidden costs on decision night. I have run event-driven strategies on my own ECMarkets ECN account for years — fill quality during headline pulses and limit-order protection are its structural edge. In a super week, your broker’s execution capability is booked directly into your equity curve. That is also why my copy-trading stack is permanently bound to ECMarkets as the execution venue.
💡 Key Takeaways
- Confirmed break: Spot gold settled at $3,995.38 on Friday — the first close below $4,000 since July 29; COMEX futures settled at $3,991.10.
- Weekly alarms: -4.05% on the week, the worst since the week ending June 5; seven down weeks of nine; ~25% off the peak. Miners had their worst week in seven months, and Berenberg warns of ETF outflows plus technical selling.
- The bull case is alive: Goldman holds its $5,400 end-2027 target, UBS sees a structural bottom near $4,000, and Nanhua calls the core defense fully priced.
- Super-week calendar: FOMC decision early October 29 (watch December guidance) → PCE inflation → month-end rebalancing on October 30; Brent at $116.72 is the biggest inflation-side variable.
- Four rules: calendar-based sizing (≤20% into the decision), mechanical levels ($3,960/$3,900 supports, $4,030-$4,040 retest, $4,135 watershed), volatility adaptation, and execution quality first — never carry heavy one-way risk into the statement.
About the Author: Dongyi, professional quantitative trader focused on XAUUSD gold trading strategies, running DongyiTrade (dongyitrade.com), sharing quantitative trading strategies and in-depth market analysis.
Contact: WeChat: DongyiTrade | Telegram: @DongyiTrade
