This article is for information only. It is not investment advice. It is not a recommendation to buy or sell gold, futures, ETFs, or any mining stock. Gold is volatile. Do your own work. Speak with a licensed adviser if you need one.
Gold does not begin the week at last Friday’s bounce. It begins lower.
Spot XAU/USD traded near $4,198 early on Monday, September 28, 2026, on several public feeds. That was about 2% under late-Friday quotes in the $4,280–$4,320 area. The drop matches silver’s open. Both metals are answering the same bond market.
The U.S. 10-year yield touched about 5.22% last week. That was the highest in 19 years. It then eased toward 5.15%. The Federal Reserve raised rates on September 16 to a 3.75%–4.00% range. Futures still price a high chance of another hike on October 28. Gold pays no coupon. A five-handle 10-year yield makes that fact expensive.
Last week still left a map. Buyers defended a cluster near $4,235–$4,255 more than once. The 50-day moving average near $4,318 capped the rebound. The daily swing high that would flip the short-term trend sits near $4,400. Monday’s print puts those lines back in play from below.
The one theme for next week is levels plus data. The gold price forecast for the next five sessions is not a single number. It is a set of doors. PCE on Wednesday and payrolls on Friday are the handles.
The levels that matter more than adjectives
Write these numbers down. They will do more work than any headline.
First support is the zone that held last week: $4,235 to $4,255. Friday’s low near $4,255 and the September swing low near $4,235 were the floor under a three-touch defense. Monday’s open under $4,200 means that floor is already being tested from the wrong side. A failed retest that cannot recapture $4,235 would confirm a break.
Second support is $4,203. That was a lower swing high and the top of an earlier base in some daily maps. Third support is $4,162–$4,157, a daily pivot support on Monday boards. Under that sits $4,114, then the psychological $4,000–$4,075 band. The World Gold Council earlier this year flagged $4,075 as a 38.2% retracement of the long 2022–2026 advance. That line is not next Tuesday’s trade. It is the backstop if this week becomes a washout.
First resistance is former support. $4,255–$4,280 is now a ceiling until it is reclaimed on a closing basis. $4,300 is the round number that rejected several September probes. $4,318–$4,320 is the 50-day average from late last week. That average is the near-term trend filter. $4,369 was Wednesday last week’s high. $4,384–$4,406 is the next resistance pocket. $4,400–$4,430 is the line several swing charts need to call the daily trend up again. The 200-day average remains far higher, near the mid-$4,500s on some late-September reads. It is a later problem.
A simple week’s box is this. Bears win if gold closes under $4,162. Bulls get a hearing if gold closes back above $4,300 and then $4,320. Everything between $4,162 and $4,320 is noise until PCE speaks.
Range math from one volatility board put a weekly band near $4,089 to $4,432 around the Monday print. That is wide. It is also honest. September already printed $2 days. Assume another one.
How last week built this week’s test
Gold spent September stuck between roughly $4,230 and $4,410 after failing to hold the mid-$4,300s and the higher summer peaks. The 61.8% retracement of the late-June to late-August rally sat near $4,231. The 38.2% retracement sat near $4,409. Price lived in the middle. Moving averages bunched around $4,326–$4,344. Neither side owned the tape.
The September 16 Fed hike did not crash the metal. It also did not launch it. The market had priced a 25-basis-point move. What it had not fully priced was the follow-through in yields. When the 10-year and the 30-year tagged multi-decade highs last week, gold finally gave ground. Thursday broke $4,299 and the short rising trend line. Friday found buyers near $4,235–$4,255 and bounced into the $4,300s. That bounce died at the 50-day average.
Monday opened as if the bounce was rented. That is normal after a yield shock. It is not proof the bull market is over. It is proof the next five days are a rate tape, not a jewelry tape.
RSI readings last week sat in the high 30s to low 50s depending on the cut. That is not a crash oscillator. It is a tired-but-not-capitulated oscillator. MACD on some shorter charts had started to curl up into Friday. Monday’s gap can unwind that curl before it matters.
The calendar: two prints and a choir of Fed speakers
Monday already features Fed talk. Michelle Bowman and Thomas Barkin were on the slate. Dallas Fed manufacturing data is noise unless it is extreme.
Tuesday brings JOLTS job openings for August and Conference Board confidence. John Williams, Austan Goolsbee, and other officials speak. Australia’s RBA decision hits in Asia. A hike there adds to the global tightening story. Gold can ignore a foreign hike if U.S. yields fall. It cannot ignore a foreign hike if U.S. yields rise with it.
Wednesday is the first hinge. August PCE, income, and spending print at 8:30 a.m. Eastern. Consensus late last week clustered near 3.7% year-over-year for headline PCE and 3.3% for core. Month-over-month forecasts sat near 0.4% headline and 0.3% core. ADP payrolls and the final Q2 GDP revision land the same morning.
PCE is the Fed’s preferred inflation gauge. Officials already nudged their 2026 PCE view higher at the September meeting. Chair Kevin Warsh described summer readings as offering no meaningful relief. A core print at 3.4% or hotter would feed October hike odds. Those odds sat near two-thirds last week, with December odds even higher on some futures snapshots. A core print at 3.2% or softer would be the first clean gift for XAU/USD this week.
Watch the 10-year yield in the first five minutes after PCE. If yields drop and gold does not rise, the metal is still digesting Monday. If yields rise and gold holds $4,162, official and physical bids may be real.
Thursday brings claims and ISM manufacturing. A hot ISM prices-paid index would rhyme with hawkish PCE. A soft ISM would rhyme with growth fear. Gold likes growth fear only when yields fall. If growth fear arrives with sticky inflation, you get the worst mix: weak risk assets and high real rates.
Friday is the second hinge. September nonfarm payrolls are due at 8:30 a.m. Eastern. One consensus sat near 100,000 jobs after a 162,000 August print. Unemployment was seen at 4.1%. Wages were seen up 0.3% on the month. Revisions matter. A 70,000 print with soft wages is a different gold market than a 150,000 print with 0.4% wages. October 28 remains the next FOMC date. Friday’s jobs number is the last major labor print before that meeting.
Geopolitics can still interrupt. It has lifted gold more than silver this year when oil and freight spiked. Treat it as a tail, not a base case, unless bonds vote with it.
Interest rates, real yields, and why $4,200 is a fight
Nominal yields near 5.2% are the headline. Real yields are the mechanism.
If headline PCE holds at 3.7% and the 10-year sits at 5.2%, the simple gap is still positive. TIPS yields near 2.8% in some late-September notes make the opportunity cost clear. Gold has lived with high real yields before when official buying and debasement fear were stronger than the coupon. It has not lived with them quietly. Pullbacks of $100 to $200 have been the rent.
The dollar index sat near 101 late last week, a two-month high in several notes. A firm dollar is a second headwind. Gold is priced in dollars. Foreign buyers feel the markup. The dollar and the 10-year often travel together when the Fed is hiking. Next week one of them may blink. Gold will follow the blink, not the press conference.
Rate-cut hopes are not the 2026 story they were in 2024. The market is pricing hikes. That flips the usual “Fed put” script. Safe-haven demand can still appear if equities crack. It will compete with the yield wall rather than ride a cutting cycle.
The bid that does not show up in one PCE print
Central bank gold buying remains the structural floor under the multi-year bull case. It does not set Monday’s tick. It does set how far a liquidation can run before real-money buyers appear.
World Gold Council flow notes through 2026 have shown official demand staying positive even when Western ETFs wobbled. August ETF inflows were large in some tallies, including a figure near $18 billion and more than 100 tonnes added in one monthly cut. That inflow and last week’s price drop can live in the same year. Patient money and fast money are not the same crowd.
Mine supply is inelastic in a week. Jewelry demand in Asia can dip when local prices spike and rebound when they fall. That physical dip-buying showed up near $4,235 last week in the bounce. It may show up again near $4,160 if that level prints. It will not offset a 20-basis-point yield spike in an hour.
Bank price targets for late 2026 and 2027 still sit well above spot at several houses. Those targets assume the official bid and fiscal worry survive a tight Fed. They are not a gold price target for next week. Do not confuse a 12-month slide with a five-day map.
Three paths for XAU/USD through Friday
Path one is repair. PCE undershoots. Payrolls cool. The 10-year yield falls toward 5.00%. Gold recaptures $4,255, then $4,300, then the 50-day near $4,320. $4,369 and $4,400 become live. That path answers “will gold continue to rise next week” with a cautious yes from a $4,200 start. It does not deliver a new high. It delivers a range reclaim.
Path two is the grind. Data is mixed. Yields stay between 5.10% and 5.25%. Gold chops between $4,160 and $4,300. Mining stocks whip. That path is the most common after a failed summer high. It bores everyone. It is how bases form.
Path three is the breakdown. Core PCE re-accelerates. Jobs stay hot. October hike odds jump. Gold closes under $4,162. Then $4,114 and $4,075 come into view. $4,000 is the headline if liquidation hits leveraged futures. That path does not cancel central-bank demand. It does punish anyone who treated $4,235 as a permanent floor.
From Monday’s $4,198, path two with a risk skew to path three is the honest base until Wednesday. Path one needs a gift from the data.
What gold stocks will do while the metal argues
Gold mining stocks are stocks first this week. They will trade the S&P and the metal.
Canadian names give a research list. Agnico Eagle Mines is the senior quality compounder. Kinross Gold is a producer with torque. Wheaton Precious Metals is a streaming name with a different cost shape. Alamos Gold and OceanaGold sit in the mid-tier. Junior gold stocks and exploration names on the TSX and TSXV will move more than the metal in either direction. They can also go quiet if financing fear returns.
A repair path helps producers first. All-in sustaining costs that looked comfortable at $4,400 still work at $4,200 for many seniors. They look worse for high-cost or heavily hedged names. Read the last quarterly AISC before you read the chart.
A breakdown path hits developers hardest. The market stops paying up for ounces in the ground when real yields are rising. That is not a comment on geology. It is a comment on the cost of capital.
Watch relative strength. If gold falls 2% and the seniors fall 6% on light volume, that can be forced selling. If they fall 6% on heavy volume with no bounce into PCE, that is a de-rating. Those tapes need different patience. None of these tickers is a recommendation. They are how Canadian investors usually express a gold view after they already have one.
How to use the two prints without guessing the close
Before PCE, write two sentences.
Sentence one: If core PCE is 3.2% or lower and the 10-year drops, I will treat $4,300 as the first test, not $4,160.
Sentence two: If core PCE is 3.4% or higher and the 10-year rises, I will treat $4,162 as the line that decides whether this is a dip or a trend.
Before payrolls, write two more.
If jobs come in under 80,000 with soft wages, the October hike can be repriced. Gold can then try $4,320 even if PCE was firm.
If jobs come in over 140,000 with firm wages, the hike stays alive. Gold can then tag $4,114 even if Wednesday looked fine.
Fed speakers can overwrite both sentences for a session. Count voting members. Color commentary from non-voters is cheaper.
Position size is the control. Average true range near $38 on one late-week note is enough to wreck a tight stop. Wider risk or no leverage is how most readers should meet this calendar.
People also asked
Will gold continue to rise next week?
Not from Monday’s open unless data help. Gold is starting near $4,200 after a failed hold of $4,300. A rise next week means a reclaim of $4,255 and $4,300. That is possible on soft PCE and soft jobs. It is not the default while yields sit above 5%.
What will gold price do next week?
The honest range is roughly $4,160 to $4,370 if volatility stays like September. A close under $4,162 opens $4,114–$4,075. A close above $4,320 opens $4,369–$4,406. The metal will follow the 10-year yield more than it will follow any single forecast.
A checklist for the gold outlook next week
Each morning, check the 10-year yield first. Then DXY. Then XAU/USD. That order matches this tape.
Mark $4,162, $4,203, $4,235, $4,300, $4,320, and $4,400. Those six levels describe the week.
Read PCE as a rates event. Read payrolls as an October FOMC event. Read mining stocks as leveraged versions of both.
Remember the official bid when the tape feels like the end of the bull market. Remember the 5% yield when the tape feels like a bargain that cannot fall. Both memories are useful. Neither is a trade by itself.
The week in one paragraph
Gold price prediction for next week is a level problem. Spot starts near $4,200. Support sits at $4,235, then $4,162, then $4,075. Resistance sits at $4,300, $4,320, and $4,400. Wednesday’s PCE and Friday’s jobs report will move the 10-year yield. The yield will move the metal. Central-bank demand and long-term forecasts still argue for a bid measured in months. They do not argue for a quiet Tuesday. Watch the lines. Respect the calendar. Treat every gold stock as a research file, not a conclusion.
Disclaimer. Canadian Mining Report and the author are not advising you to purchase or sell gold or any security. Companies named are for illustration only. Spot prices are snapshots from public feeds on or around September 28, 2026, and will be stale quickly. Verify PCE, payrolls, ETF flow, and company filings against primary sources before you act. Past prices do not predict future prices.