A weaker dollar can rent gold a round number before lunch. It cannot rent the year. The year is still under the January high, and the next hike has only moved from October toward December.
This is not a recommendation to buy, sell, or hold gold, a gold miner, a gold fund, or any other security. It is not investment advice and not a solicitation. Gold moves on the dollar, on real yields, and on a policy rate the buyer does not set. You can lose money. The question in the headline gets a test on this page. It does not get a price target you should trade.
Here is the only idea. A gold price rebound to 4,200 dollars is a place on a Friday screen. It is not a rally, and it is not a gold price forecast. On the morning of October 16, 2026, December gold futures were quoted near 4,204 dollars an ounce as the dollar eased off an 18-month high. Spot quotes through the same morning sat in a band from the mid-4,100s to just over 4,200, depending on the desk and the hour. A later afternoon reading was back under 4,170. The round number happened. The round number did not stay put long enough to be called a new regime. A weakening U.S. dollar can lift the metal for a session. It cannot, by itself, answer whether the gold price recovery continues next week, or next month, or through a December meeting the market still treats as live.
What the screen actually showed
Start with the contract people mean when they say gold “hit 4,200.” A Yahoo Finance note on Friday morning said December futures opened at 4,192.10 dollars, up 0.5 percent from Thursday’s close, and were near 4,204.40 dollars at 6:45 a.m. New York time. The same page later showed the contract near 4,195.90, up about 26 dollars on the day. So the 4,200 print was a morning trade in the futures, not a settlement, and not a number that every later quote repeated.
Spot gold, the cash price, did not march in lockstep with that futures print. Early Friday notes put spot near 4,174 to 4,179 dollars. One desk later said physical gold was holding near 4,200.10. An afternoon aggregator had the price near 4,160 at 1:31 p.m. New York time, down from the prior London fix. These are not errors to be averaged into one true price. They are clocks. A gold price rebound that exists at 6:45 a.m. and is gone by early afternoon is a trade. It is not a trend you can paste into a note and call done.
Thursday is the day the bounce actually started. One daily table put Thursday’s close at 4,170.70 dollars, up 1.61 percent from 4,104.80 the day before. That table had the week up about 0.9 percent, the past month down, and the year 2026 down about 4.5 percent. Reuters put Wednesday’s spot close near 4,096.58, down 1.45 percent, on the day the inflation report hit. The path into Friday was three soft sessions, then a Thursday lift of a bit more than 1 percent, then a Friday morning push that kissed a round number. Calling that whole path a rally skips the part where the metal was falling on hot prices two days earlier.
The year is the part a headline will not fit. The same daily table put the all-time closing high at 5,405 dollars on January 29, 2026. Thursday’s close was about 23 percent under that high. The 52-week range on that table ran from about 3,948.50 to 5,405. Gold in October 2026 is not cheap against a long history. It is cheap against its own January. A gold bullish outlook that starts at 4,200 and never mentions 5,405 is an outlook with the top cut off. The investor who needs the top should write it down before he writes the word rebound.
The dollar eased. It did not reverse.
The headline’s second half is the weakening U.S. dollar. That half is real, and it is smaller than the verb suggests. On Wednesday, October 14, Reuters said the dollar index hit an 18-month high after the inflation report, trading as high as about 102.94 before settling near 102.53, up about 0.34 percent on the day. One other screen had the index near 102.90 at the hot part of that session. By Friday morning, vendor readings were back near 101.6 to 101.8. A widely used daily series had Thursday at 102.00 and Friday at 101.635. From Wednesday’s peak to Friday morning is a slip of roughly one index point. That is enough to help gold for a day. It is not a new dollar regime.
Look at the year on the dollar, not only the week. That same daily series had the index up about 3.3 percent from a year earlier. A currency that is still up on the year, and that just printed an 18-month high, can fall for two sessions and still be the strong currency in the room. Gold likes a falling dollar because a cheaper dollar makes an ounce less expensive for a buyer who pays in euros or yen. The mechanism is arithmetic. It is not a vote of confidence in the metal. If the dollar stops falling, the arithmetic stops. The gold price outlook that treats two soft dollar sessions as a trend has already spent a move that may be finished.
There is a second dollar fact, and it is about oil and fear, not only about rates. The dollar’s Wednesday jump came with stocks down and with crude still high. Brent settled near 106 dollars a barrel that day, on the Reuters account. A strong dollar and expensive oil can hit gold together, because the dollar raises the metal’s foreign-currency price and the oil can lift the inflation that keeps rates up. Friday’s gold bounce arrived as both the dollar and, on one bank’s note, Brent eased. OCBC’s strategists said gold rebounded more than 1 percent as Treasury yields, the dollar, and Brent all gave ground. They also said the metal is trading the bond market’s reaction to the data, not the data itself. That sentence is the cleanest description of this week. The inflation did not cool. The market’s first fear of an immediate hike cooled. Gold rose on the fear, not on a better report.
The chair refused to pre-judge. He did not cut.
The dollar slipped after Federal Reserve Chair Kevin Warsh spoke. Friday notes said he told an audience, in Bangkok and at an IMF setting depending on the write-up, that he does not pre-judge meetings. He did not deliver a plan to raise rates in October. He also did not deliver a plan to stand down for the year. FXEmpire’s James Hyerczyk put the October hike probability near 27 percent before the remarks and near 15 percent after. Another Friday note put the October 28 odds near 16 percent, down about 12 points from Thursday’s close. A third note said the October hike was close to priced out. The direction is clear. The exact percentage is not, because the feeds do not match. Use the direction. Do not marry a single odds screenshot.
October is not the only meeting. The same Friday notes that marked October down still had a December hike as the likely case. One of them put the chance of a hike by the December 9 meeting near 89 percent. Hyerczyk’s forecast kept “December risk” in the headline on purpose. Warsh took one meeting off the immediate table. He left the next one on it. A gold price prediction that says “the Fed blinked, so gold can run” has blended two dates into one. The blink, if it was a blink, was about October 27 and 28. December is still a live argument. A live argument is a yield. A yield is a cost of holding a metal that pays none.
The policy rate itself did not move this week. Reuters said the central bank raised the target range by a quarter point in September, to 3.75 to 4.00 percent. That hike is three weeks old in the memory of the bond market and one month old on the calendar. Officials have spent the weeks since sounding patient and then watching inflation come in hot. New York Fed President John Williams had leaned the dollar softer earlier in the week. Vice Chair for Supervision Michelle Bowman, on Thursday, signaled a wait-and-see stance before another increase, on the Reuters account. Patience from officials plus a hot price index is how you get a dollar that spikes and then slips. It is also how you get a gold market that cannot decide if 4,200 is a ceiling or a step.
The jobs side of the mandate is not a boom, and it should be named so the patience has a source. The Bureau of Labor Statistics listed the September unemployment rate at 4.2 percent and preliminary payroll growth at 29,000. That is a labor market adding people slowly. It is one reason officials can say they are in no rush. It is not a reason to ignore a 3.7 percent inflation rate. The Fed’s job is both numbers. Gold’s job, in a week like this, is to trade whichever number the bond market is afraid of that morning. On Wednesday the bond market was afraid of inflation. On Friday it was afraid of being early. The metal followed.
The inflation report did not get rewritten
The Bureau of Labor Statistics said the consumer price index rose 0.6 percent in September, seasonally adjusted, and 3.7 percent over the year, not seasonally adjusted. The core index, which leaves out food and energy, rose 0.3 percent on the month and 2.6 percent on the year. Reuters said economists had looked for a core monthly gain of 0.2 percent and a headline yearly rate of 3.6 percent. Both came in a tenth hotter. The agency’s own release title pointed at gasoline. Over the year, the same release noted fuel oil up sharply and airfares up sharply. This was not a quiet miss. It was an energy-led jump with a core that still quickened.
Producer prices need the same care, because some market notes this morning quoted a 6.3 percent producer-price figure that the official monthly line does not match. The Bureau of Labor Statistics home page lists the producer price index for final demand at plus 1.3 percent in September, and it marks that figure preliminary. A 1.3 percent monthly move in final demand is a large monthly move. It is not the same sentence as 6.3 percent. A gold note that borrows the larger figure without saying which series and which horizon has already left the release. Use the release. If a year-over-year producer-price rate matters to the trade, wait until you can name it from the same table. Do not let a round inflation scare do the work of a footnote.
Hot inflation is supposed to be bad for gold, in the textbook, because it pushes interest rates up and rates are the cost of holding a bar. This week followed the textbook on Wednesday and broke it on Thursday and Friday. Wednesday: prices in, dollar up, gold down about 1.5 percent. Thursday and Friday: officials sound unwilling to hike this month, yields ease, dollar slips, gold bounces. The textbook did not fail. The timing of the policy response did the work. OCBC said the latest move is a reminder that gold is trading the bond market’s reaction to U.S. data rather than the data itself. Write that on the gold market outlook and a lot of noise falls away. The data were hot. The reaction, for two sessions, was that October might not be the meeting that answers them.
Real yields are the part of the reaction that gold actually cares about. One Friday note put the 10-year Treasury inflation-protected yield at 2.78 percent at Thursday’s close, down 0.06 percentage point on the day, citing the Treasury’s own figure. A real yield near 2.8 percent is not low by the standard of the last decade. It is a little less high than it was on Wednesday. Gold can rise on “less high.” It struggles when “less high” turns back into “higher again.” OCBC’s condition for a further gold price recovery was sustained relief in real yields, and an avoidance of a fresh jump in yields. That condition is a forecast you can check. A round number is not.
Nominal yields tell the same story with more noise. Wednesday’s Reuters note had the 2-year yield, the one most tied to near-term policy, up 1.5 basis points to 4.81 percent. A Friday morning note had the 2-year near 4.75 percent and the 10-year near 5.09 percent, up a little overnight after Thursday’s drop. Another note said the 10-year had fallen to a three-week low on Thursday, by about a tenth of a percentage point. The week is a down-and-up, not a collapse. Five percent on the 10-year is still a high number next to a metal with no coupon. Anyone who says yields “do not matter because gold is at 4,200” is describing a morning, not a model.
The levels are a map, not a promise
Traders will talk about 4,200 as if the number has a memory. Sometimes it does, because many orders sit on a round figure. Sometimes it does not. OCBC, on Friday, marked 4,200 dollars as resistance and tied it to the 21-day moving average. The next marks on that note were 4,240 at the 100-day average and 4,330 at the 50-day. Support was 4,100 and 4,060. A different Friday note, from FXStreet, had the metal near 4,179 with a first cap at the round 4,200, then the October 2 high near 4,230, and beyond that the September 25 high near 4,315. On the downside that note had an intraday low near 4,150 and a range floor near 4,070. One more desk used 4,000 as the lower reference and 4,330 as the upper one. The maps rhyme. They are not identical. A gold price forecast that pretends they are one line is selling certainty the screens do not have.
Read what a moving average is. It is the average of recent closes. It is not a floor the market owes you. If Friday’s push fails under 4,200, the 21-day average did its job as a place where sellers showed up. If the price closes above it and holds, the same line becomes a place where buyers showed up. The line did not predict either outcome. It labeled the argument. The gold price prediction worth keeping is the one that says what must stay true for the line to matter. Here, what must stay true is the slip in real yields and the slip in the dollar. If both reverse after the weekend, the line is just a line again.
The lower end of the map matters as much as 4,200, and the headline will not mention it. A range floor near 4,070 to 4,100 is only about 100 to 130 dollars under Friday morning’s spot cluster. That is roughly 3 percent. A market that can travel 3 percent in a few sessions, and that just did, is not a market you describe with the word “holding.” Thursday’s bounce started from just over 4,100. Wednesday’s Reuters print was just under 4,100. The gold price rebound is a trip back across a band the metal has already crossed twice this week. Crossing a band is not leaving it.
There is a larger band, and it is the year. From the January close near 5,405 to a Friday morning near 4,180 is a drawdown of about 1,200 dollars an ounce. From the 52-week low near 3,949 to Friday morning is a climb of a bit more than 200 dollars. The metal is closer to the bottom of its own year than a “rebound to 4,200” headline sounds, and it is still far above the lows of the prior cycle. Both distances belong in the gold investment outlook. The first keeps you from buying a bounce as if it were a bottom. The second keeps you from talking as if 4,200 were a distressed price. It is a high price that has fallen. Those are different sentences.
The slow bid is not this morning’s bid
Official buying is the part of the gold story that does not care about a Friday morning print, and it should be kept in its own box. The People’s Bank of China said its gold holdings were 77.47 million ounces at the end of September, up from 76.73 million ounces in August. The monthly add was 740,000 ounces. It was the 23rd straight month of buying, on the official count carried by the Global Times. August’s add had been 650,000 ounces. Converted at about 32,151 troy ounces to the tonne, 740,000 ounces is about 23 tonnes, and 77.47 million ounces is about 2,410 tonnes. A Kitco note the same week said 21 tonnes and a stock of about 2,196 tonnes. The ounce figures are the ones the central bank’s data release supports. Use the ounces when the tonne counts disagree. A 23-tonne month is a large month for that buyer. It is a small month next to the world’s daily gold trade. It does not explain a 100-dollar swing between Wednesday and Friday. It explains why dips have a bidder who does not mark his book to a headline.
That buyer accelerated as the price fell from the January high. The central bank’s own logic, as outside economists described it, is reserves, not a trade. Foreign-exchange reserves were about 3.40 trillion dollars at the end of September, down 38.1 billion dollars, or 1.11 percent, on the month, according to the same official release. Gold as a share of those reserves is still a slice, not the pie. A slice that grows for 23 months is a habit. A habit is not a floor under 4,200 dollars. If the price runs, the habit can pause. If the price falls, the habit can speed up, which is what the larger September add suggests. The gold market outlook should say “a steady buyer on weakness,” not “a buyer who will defend Friday’s high.”
Funds are the other slow bid, and they are lumpier. A Friday research roundup said gold-backed funds took in about 31 billion dollars in the third quarter, a record for that window, and that holdings stood near 4,256 tonnes. Treat that as the roundup’s figure, not as a line you have audited from each fund’s file. Record inflows in a quarter when the price was well off its January high tell you that some investors bought the dip in size. They do not tell you those investors will buy the next dip, or that they will hold through a December hike. Flows follow performance until they do not. A gold investment outlook that quotes the inflow and skips the condition is an advertisement for the last quarter.
Put the two slow bids next to the fast one. The fast one is the dollar and the real yield, and it moved the price this week. The slow ones are the central bank ounce count and the fund tonne count, and they move the floor over months. Mixing them is how a morning rally gets narrated as a structural turn. They can point the same way for a while. They are still different clocks. If you own gold because of the slow clock, a failure at 4,200 is noise. If you own it because of the fast clock, a failure at 4,200 is the trade. Know which clock you set.
Silver moved more, and it is not the same bet
Silver is the noisy cousin, and it belongs here only as a check on the story. Thursday’s table had silver at 57.42 dollars, up 3.20 percent on the day, with the gold-to-silver ratio at 72.6. The year-to-date figure on that table was about minus 20 percent, against gold’s minus 4.5 percent. Friday morning quotes ran from the high 57s to about 58.23 depending on the desk. Silver’s bounce on Thursday was larger than gold’s, in percent. Silver’s hole for the year is also larger. A weakening U.S. dollar helps both, because both are priced in dollars. It does not make them one metal. Silver still has an industrial buyer who can step away. Gold’s extra buyer, this year, has been the official sector and the fear trade.
The ratio near 72 says one ounce of gold bought about 72 ounces of silver at Thursday’s close. That is a number, not a signal to switch. Ratios wander. They do not owe you a return to some old average. Use the ratio as a way to see which metal did the week’s work. This week, into Thursday, silver did more of the bouncing and had more of the year’s damage left to repair. A gold bullish outlook that quietly assumes silver will “confirm” the move is borrowing a second market. If silver gives back the Thursday pop and gold holds, the dollar story is mostly a gold story. If both give it back together, the dollar slip was the whole trade, and it is over.
Will gold prices continue to rise?
That is the question people are asking under the headline, and it deserves a straight answer that is still not a tip. Gold prices can continue to rise from here if the thing that lifted them on Thursday and Friday stays in place. That thing is not the round number. It is a dollar that does not climb back to Wednesday’s 18-month high, and a real yield that does not give back Thursday’s drop, while the market keeps October off the table. If those three hold into the October 27 and 28 meeting, a push through 4,200 toward the early-October high near 4,230 is a plausible path, not a promise. The September 25 area near 4,315 is the next old high on one map. It is not a target this page is setting. It is a place the price has already been, and failed to hold, inside this same season.
Gold prices can stop rising, or fall, on a shorter list than the bulls are using. The list is a dollar bounce, a yield bounce, or a December hike that the 89 percent crowd is right about and the gold market has not fully digested. Hot inflation is already in the file. It does not need to get hotter to hurt the metal. It only needs to convince the bond market that October’s patience was a delay, not a change. One more tenth on the next price index can do that. So can a chair who, having refused to pre-judge, later judges. The gold price prediction that says “the rally continues because the dollar weakened” has named a cause that is two sessions old. Causes that young die young.
There is a third path, and it is the dull one, and it may be the most likely until the meeting. OCBC said daily momentum was not showing a clear bias and that two-way trade was the base. Support near 4,060 to 4,100. Resistance near 4,200, then 4,240. A band of about 150 dollars is a wide band for a quiet week and a narrow band for a year that has already traveled from 5,405 to under 4,000 at the low. Consolidation is not a defeat for a holder who bought the slow bid. It is a defeat for a holder who bought Friday morning and needs a straight line. Most of the damage in gold this year has come from people who needed the straight line.
So the honest gold price forecast is a fork, not a number. Fork one: the dollar stays soft, real yields stay below this week’s peak, and the price spends time above 4,200. Fork two: the dollar reclaims the mid-102s, the 10-year real yield turns up from 2.78 percent, and the price is back toward 4,100 before the October meeting. Fork three: the price chops inside the band and the meeting decides. A forecast that picks a fork without saying what would prove it wrong is a mood. This page picks no fork. It says the morning print is not evidence for the first one. The morning print is what the first one looks like for a few hours. Hours are cheap.
What a holder is actually holding
A bar of gold is a way to own something that is not a promise from a government. That is the old case, and it is still the case. It is not, this week, the thing that moved the price from 4,097 to a morning print near 4,204. That move was a rates move and a currency move. The old case matters for the person who can sit through both forks. It does not matter, as a trading reason, for the person who needs Friday’s bounce to keep going on Monday. Mixing the old case with the new tick is how gold letters get people hurt. The letter says “the system.” The tick says “the dollar slipped.” Only one of those showed up in the tape.
The cost of the old case is the yield you do not earn. A 10-year note near 5 percent, and a real 10-year near 2.8 percent, is the bill for holding the bar instead of the bond. When that bill shrinks, the bar gets easier to hold and the price often rises. When the bill grows, the bar gets harder to hold and the price often falls. This week the bill shrank a little after it had grown a little. The gold price recovery is the price of that shrink. If you cannot say the size of the bill, you do not have a gold investment outlook. You have a hope that the line goes up. Hope is not a position. The bill is.
Miners are a third object, and this page will not name one as a buy. A miner is a business with costs, a mine life, a country, and a share count. The share can fall while the metal rises, and it can rise while the metal falls, because the share also trades the cost, the news, and the mood of the stock market. Anyone who treats a 4,200 print as a reason to buy a miner has skipped the only question that matters for the equity. Does the mine still make money if the metal is at the bank’s dull price, not at Friday’s exciting one? This week’s dull neighborhood is the 4,100s, which is where the metal was on Wednesday. A mine that only works at 4,200 is a mine that only works on a morning. That is not a mine plan. It is a wish.
Funds that hold bars are closer to the metal and still not the metal. They charge a fee. They can trade at a premium or a discount. They can be forced to buy or sell because of flows, which then moves the very price the holder thought he was tracking. The third-quarter inflow figure, if it holds up under a look at the source files, says a lot of money chose the fund wrapper while the price was off the high. The next quarter can choose the other way if the December meeting goes against the bar. A wrapper is a convenience. It is not a shield against the fork.
What you can write down
You can write the morning without promoting it. December futures were quoted near 4,204 dollars at 6:45 a.m. New York time on October 16, after opening near 4,192. Spot quotes the same morning were mostly in the 4,170s and 4,180s, with at least one desk printing just over 4,200, and a later afternoon figure back under 4,170. Thursday’s close on one table was 4,170.70, up 1.61 percent. Wednesday’s Reuters spot figure was about 4,096.58, down 1.45 percent. The gold price rebound is that sequence. It is not a close at 4,200.
You can write the dollar without calling two sessions a bear market. The index hit an 18-month high on Wednesday, near 102.9 at the session extreme, and was near 101.6 to 101.8 by Friday morning depending on the vendor. It is still up on the year on a commonly cited daily series. A weakening U.S. dollar explains the lift. It does not explain a new cycle. The lift lasts as long as the weakening lasts.
You can write the Fed without a cartoon. The target range is 3.75 to 4.00 percent after a September quarter-point hike. October hike odds fell after the chair refused to pre-judge the meeting. Published readings this week ran from the low 40s after the inflation report to the mid-teens by Friday morning. December, on Friday notes, was still the meeting the market treated as live, with one reading near 89 percent. Warsh cooled October. He did not cancel the hike debate.
You can write inflation from the agency, not from a scare line. Consumer prices rose 0.6 percent in September and 3.7 percent on the year. Core prices rose 0.3 percent and 2.6 percent on the year. Final-demand producer prices were listed at plus 1.3 percent for the month, preliminary. The 10-year inflation-protected yield was 2.78 percent at Thursday’s close on one note that cited the Treasury, down 0.06 point on the day. The data were hot. The real-yield reaction, for a day, was not.
You can write the slow bid in ounces. China held 77.47 million ounces of gold at the end of September, up 740,000 ounces on the month, the 23rd straight monthly add. That is about 23 tonnes and a stock near 2,410 tonnes if you convert at the standard ounce count. A research roundup put third-quarter fund inflows near 31 billion dollars and holdings near 4,256 tonnes. Those figures are a floor story. They are not a Friday story. Keep them in the month, not in the minute.
You can write the map as a map. Resistance near 4,200, then about 4,230 to 4,240, then the mid-September area near 4,315. Support near 4,150, then 4,100 to 4,060, with 4,070 as one desk’s range floor. The January closing high near 5,405 is the number that makes this year a down year of about 4.5 percent at Thursday’s table. A gold price outlook that omits the January high will sound braver than the position is.
What would make this reading wrong
The reading is wrong if Friday’s futures did not trade near 4,200 dollars, or if the 4,204 figure was a bad tick that no one could have dealt on. Use the exchange settlement when it prints. A morning quote and a settlement can differ by more than a round number. If the settlement is 4,160, the headline “rebounds to 4,200” was a moment, and this page already treats it as a moment. If the settlement is well through 4,230, the moment became a break and the resistance map needs to be redrawn. Say so. Do not keep a ceiling the market has closed above.
The reading is wrong if the dollar did not ease, or if Friday’s index is a different dollar from Wednesday’s. Vendor indexes differ by tenths. A tenth is not the argument. A Friday index that is still at Wednesday’s high would kill the “weaker dollar” half of the headline, and gold’s bounce would need another cause. Find that cause or drop the bounce’s explanation. Do not invent a dovish turn the chair did not give.
The reading is wrong if December hike odds have collapsed, not merely October’s. The split between the two meetings is the heart of the piece. If a later odds sheet shows December back with October, in the teens rather than near 90, then the market has priced a longer pause and the gold price forecast can be less grudging. Until that sheet exists, a one-meeting delay is not a cycle of cuts. There are no cuts in this week’s tape. There is a debate about one more hike, and about which month wears it.
The reading is wrong if the inflation figures used here are not the agency’s. The consumer-price lines are the Bureau of Labor Statistics release of October 14: 0.6 and 3.7, core 0.3 and 2.6. The producer-price monthly line used here is the agency’s listed plus 1.3 percent for September final demand, preliminary. A revision, or a different index, can change the heat. It is unlikely to change the sign. If the next report is soft and yields fall with it, the “trading the reaction, not the data” line still holds, and the reaction will finally agree with a cooler report. That would be a different week. Write the different week when it arrives.
The reading is wrong if you use it as a reason to buy because the dollar weakened and a round number printed. A weakening U.S. dollar is a reason the price rose this morning. It is not a reason the price must rise next month. The year is down. The real yield is still high. The chair has not left December alone. A buyer who can sit through a return to 4,100, and who is really holding the slow bid rather than the morning, is having a different conversation from the one the headline offers. This page is about the headline’s conversation. In that conversation, the rally has not been shown to continue. It has been shown to start, for a day and a half, off a three-day drop.
The idea, once
Gold’s Friday morning push toward 4,200 dollars was a rent payment from a softer dollar and a milder fear of an October hike. The dollar had just made an 18-month high. Inflation had just printed 3.7 percent. The chair refused to pre-judge one meeting and left the next one in play. Real yields eased for a day and did not become cheap. The year is still a down year against a January close near 5,405. China’s central bank added ounces in September, and funds added tonnes in the third quarter, and neither of those clocks is the clock that struck 4,200 before lunch.
Can the move continue? Only if the rent keeps being paid. That means the dollar stays down from Wednesday’s high, the real yield stays down from this week’s peak, and December does not reprice into a shock. None of those is known on a Friday morning. A gold price rebound to a round number is a place. A place is not a rally. The rally, if it comes, will be visible in the yields and in the currency first, and only then in the metal. Watch those. Do not watch the adjective.