A reader messaged me last week. "Where does gold go if the Fed hikes again in December?"
Wrong question.
Not because the Fed doesn't matter. Because asking the Fed first is how you end up permanently surprised by gold. Trust me on this one — that was me from 2013 through 2018. Every single time.
Here's my read heading into late 2026. Gold is not a pure real-rates trade anymore. Most macro guys still model it that way. Tighter policy, stronger dollar, gold down. Looser policy, weaker dollar, gold up. Fine. That worked for about thirty years. It's been leaking for five. And honestly? The last two years it's been actively misleading.
The long-term trajectory of XAUUSD isn't set by the FOMC. It's set by the slow erosion of dollar credibility. Real rates didn't stop mattering — no, that's not what I'm saying. They became cyclical accelerants on top of a structural story that keeps getting stronger.
Let me show you what I mean on the tape.
The Real Rates Correlation Broke. Almost Nobody Repriced.
The real-rates correlation broke. That's point one. Almost nobody repriced.
I've traded gold since the days when an NFP print would throw $30 into the spread and a Fed statement set the tone for a month. In that regime, DXY was a reliable inverse for XAUUSD on a D1 close basis. You could almost trade gold by trading the dollar. I did. For years. Expensive tuition, that — because I kept applying the model long after the model stopped working.
Then something shifted.
Watch the price action around $4,000. Gold has held structure there through dollar strength that would've sent it to $3,400 in the old regime. That sequence matters more than any single Fed statement. I've watched traders dismiss it as noise. Is it noise? A market that stops responding to its usual driver is telling you the driver changed.
When I pull up D1 and mark the recent swing lows, the retracement behavior on the pullbacks convinced me. The 38.2% and 50% levels have been bought aggressively. Not the deep 61.8% washouts the old real-rates model liked to hunt. That's accumulation behavior, not distribution. I've traded both long enough to tell them apart by feel now. This is the accumulation kind.
Ask yourself something simple. If the real-rates model still ran the show, why did gold fail to break down through the $4,000 handle on the biggest dollar squeeze of the last eighteen months? Either the model is broken or the buyers are different. Both answers point the same direction.
Central Banks Are Buying Like the Dollar's Future Is an Open Question
The second leg of this thesis is the part most retail traders ignore. It doesn't show up on an intraday chart.
Sovereign reserve managers have been adding bullion at a pace we haven't seen in decades. I don't have exact 2026 figures in front of me, and I won't invent them. What I will tell you is what I track: World Gold Council reserve data has shown sustained net central bank buying for multiple years running, and that trend did not reverse when the dollar rallied. That's the part that matters. It held through a strong-dollar phase. That's exactly when the old model said it should've stopped.
That buying isn't price sensitive the way retail flow is. It's policy sensitive. These aren't people watching a 4H chart at the London open. These are reserve managers asking a strategic question about what happens to a dollar-denominated reserve system when the issuer runs deficits that no longer look temporary.
This is the de-dollarization piece the headlines oversimplify. I'm not telling you the dollar is dying tomorrow. I'm not telling you some BRICS currency replaces it next quarter. I'm telling you reserve diversification is a slow drip. Slow drips move long-term price floors. Gold at $4,000 in 2026 is not the same trade as gold at $1,300 in 2019, even if the Fed funds rate is higher now. The buyer base changed underneath it.
I could be wrong about the pace. I've been wrong about the pace before. Back in 2021 I thought the structural bid would show up faster than it did, and I sat through a year and a half of chop that tested my patience hard. The direction of the flow was right. My timing was bad. Those are two different mistakes, and I've paid for both.
The D1 Structure Is Saying the Same Sentence
The chart says the same thing: $4,000 is the floor.
Back to my chart. I keep it simple now. After a decade of running more indicators than I care to admit, I work off Fibonacci levels plus D1 support and resistance. That's the whole kit.
Here's where I sit as of this writing. Spot gold is trading around $4,075 to $4,080. The swing high that matters for my structure sits well above, and I've been using the $4,445 zone as a reference for the extension area when buyers regain control. On the downside, the $4,000 handle has been the structural floor. It has survived multiple tests on a closing basis. That is the level I'd need to see broken on a weekly close before I abandon the long