Spot gold price prints on October 1, 2026 were above $4,100. Reuters put the spot price at $4,159 an ounce late in the New York morning, with December futures near $4,189. Later quotes on the same day sat around $4,170 to $4,177. The round number held. The month that just ended did not. Gold prices fell over 6 percent in September. One late-month close, on September 28, was $4,115 after a drop of nearly 4 percent in a single session. Intraday trade that day reached about $4,110.
A closing high of $5,405 was set on January 29. The October 1 area is about 23 percent under that close. Some reports put the January peak above $5,500. Either way, the metal is no longer in the runaway phase. It is in a gold price correction that has already lasted long enough to scare people who bought the high, and short enough that a gold bull market can still be the larger story. Those two facts can both be true. They are why October feels like a verdict and is not one.
Gold mining stocks did worse than the bar. Through September 25, one comparison showed the big-miner fund, GDX, down about 9.3 percent over a month while the bullion fund, GLD, was down about 6.6 percent. GDX finished September near $88 a share. It had traded above $101 early in the month. The fund’s own figures show a year-to-date gain of only about 2 percent by September 30, after a much larger run earlier. Leverage is the reason. It is also the risk.
This piece has one theme. A start above $4,100 is not a gold price recovery, and it is not a reason to assume gold mining stocks rebound now. The metal is leaning on a round number while bond yields and gold are still in a fight, and the dollar is still firm. Miners rebound when the metal stops making lower highs and the mines can keep their margins. September showed the gear works down. October has not shown that the gear has switched. Until yields stop rising, or the dollar does, a bounce is a pause inside a gold market correction.
Nothing here is advice to buy or sell gold, a miner, or a fund. A level can break. A bank forecast can be wrong. A stock can fall after the metal rises, because costs and politics do not trade as a bar.
What September actually did
The path inside the month matters more than the monthly total. In the last week of September the tape went like this. A close near $4,287 on September 23. A close near $4,274 the next day. Then $4,285. Then, on September 28, $4,115. Then a bounce to about $4,182 on September 29. Then a close near $4,157 on September 30. October 1 opened in that same band and stayed above $4,100. The low of the washout is about $4,110. The high of the prior shelf is about $4,200 to $4,300. That is the box.
A 6 percent month is a gold price correction, not a crash back to an old world. Gold was under $4,000 as recently as late June in some accounts of the year’s path, after a record in January and a war that started in February. It had a strong August. September gave a large piece of that August back. People who say the gold bull market is over are using one month to cancel a year that still left the price more than double the levels of a few years ago. People who say the dip was nothing are ignoring a 23 percent fall from the January close and a week in which $4,200 failed.
Gold price volatility is the fee on that argument. The bullion fund’s volatility has been about 21 percent. The miner fund’s has been about 40 percent. The miner fund’s beta to bullion, on a three-year count, is about 1.6. A down month in gold of 6 percent can be a down month in miners of 9 or more, and a single bad week can be worse. That is not a reason to avoid the shares forever. It is a reason not to call a one-day bounce a gold stock rebound. The shares need more than the metal needs. They need the metal to keep a gain.
The news that should have worked, and did not
The cleanest fact of the last week is a failed rally. Core inflation in the PCE report came in at 3.0 percent from a year earlier. The forecast had been about 3.3 percent. Fed rate hike expectations for October collapsed. Reuters said traders cut the odds of an October hike to 31 percent, from 69 percent a week earlier. A softer print and a smaller chance of a hike are the sort of Federal Reserve gold headline that used to lift the price and hold it.
It lifted the price for a morning. One account put a spike near $4,251 on the release. The gain did not hold. Yields came back. The dollar came back. Futures settled only slightly higher, and the spot gold price finished the day lower, near $4,157. A market that cannot keep a rally on the news it wanted is telling you the buyers are tired. Gold market sentiment is not a poll. It is that failure.
The next morning did not repair it. Jobless claims were reported near 197,000, a low number. A prices-paid gauge in a factory survey jumped. The dollar index pushed toward 102, a high for the year in that account. The 30-year Treasury yield was reported near 5.66 percent, a high in decades. The 10-year has been trading around 5.2 to 5.3 percent, the highest zone since 2007. Gold interest rates are this simple. A bond that pays more than 5 percent, in nominal terms, raises the cost of holding a bar that pays nothing. Gold and Treasury yields moved against the metal even as the front end of the curve, the part tied to the next Fed meeting, got a little easier.
That split is the story of the week. Fed rate hike expectations fell, and the long bond did not care. A softer Federal Reserve gold impulse met a harder bond market. The US dollar and gold pulled in opposite directions, with the dollar firm enough to make the ounce dearer for buyers who do not earn dollars. You can have a kinder meeting and a harsher hurdle. October started on the hurdle’s side.
Can gold recover after the September selloff
Yes, it can. The question is what the word recover means. A move from $4,110 back to $4,170 is a bounce. It has already happened. A gold price recovery that repairs September would need to reclaim the shelf the month broke. Analysts watching the tape have marked $4,110 as nearby support, $4,000 as the larger floor, and $4,200 then $4,300 as the first gold price resistance. Kathleen Brooks at XTB put it in that order. The $4,000 area has held. She also said the short and medium trend is rough, and that a strong dollar and high yields are the weight.
HSBC cut its average gold price forecast for 2026 to $4,490 an ounce, and its 2027 average to $4,825. Read the word average. A year that opened above $5,000 and spent time above $4,500 can still average $4,490 even if the autumn is ugly. The bank said near-term pressure can continue, and that the price may be nearing a bottom. It also said central bank gold buying could return if the price falls toward $4,000 or through it. That is a gold price prediction about the year’s average and about official demand. It is not a promise that October is the low.
A recover-or-not test that an investor can actually use has three lines. First, does $4,000 hold if Friday’s jobs report is strong. A hot payroll number is the risk several desks named, because it would revive Fed rate hike expectations and push gold and Treasury yields the wrong way again. Second, does $4,200 get taken back and kept. A gold breakout that matters is a close back through the level September lost, not a wick. Third, does the dollar stop making highs. US dollar and gold cannot both be the safety bid every day. In September and into October, the dollar was winning that bid. A gold price outlook that ignores the dollar is a speech.
Global gold demand did not vanish because the price fell 6 percent. Jewelry and investment demand slow when the price is still near records in most currencies. Official buying is a different wallet. Central bank gold demand has been the quiet bid of this cycle. It does not have to buy every dip on the day of the dip. HSBC’s point is that the dip, if it deepens toward $4,000, is where that bid has shown up before. Korea’s central bank, which has not added physical gold since 2013, has told parliament it aims to buy about one tonne of domestic gold in December, paid in won. One tonne will not set the world price. It is a sign that a new buyer still wants a bar after a large decline. Central bank gold buying is a floor argument, not a timing tool.
Is the gold selloff over
Not on the evidence of one session above $4,100. A selloff is over when the price stops falling after good news and starts holding breaks. This week the good news was the inflation miss. The price gave the rally back. The short-term trend, in the words of more than one desk, is still down. A retest of $4,000 is still on the table if jobs surprise to the upside. Calling the gold selloff over because the calendar turned is how people buy the middle of a gold market correction and then blame the metal.
It is also wrong to say the gold bull market is finished because September was red. A bull market is a sequence of higher major lows, not a sequence of green months. The major low that would break the long story is not $4,150. It is a loss of $4,000 that does not come back, and then a loss of the summer floor under that. We are not there. We are in the zone where both sentences are cheap. “It is over” is cheap. “The low is in” is cheap. Gold technical analysis earns its name when it names the level that would retire each sentence.
Gold support and resistance, stated plainly, is this. Support sits near $4,110, then $4,000, then the mid-June area some charts put near $3,985. Resistance sits near $4,200, then $4,300, which was support earlier in September. A gold technical outlook that is honest will say the price is below the broken shelf and above the round number. That is a range. Ranges end. They do not end because a writer needs a headline on the first of the month.
Gold market sentiment will flip faster than the fundamentals if $4,200 is reclaimed on a weak jobs print and a softer dollar. It will flip the other way if $4,110 breaks on a strong jobs print. The jobs report is Friday. Until it prints, October 1 is a holding day. Holding days feel like answers. They are not.
Why the miners are not the metal
Gold mining companies are businesses that happen to sell gold. The sale price went down in September. Many of their other costs did not. Fuel is the clean example, because two large producers have already published the math. Barrick’s 2026 outlook assumes about $70 West Texas oil. Its own table says a $10 move in oil changes cash costs by about $12 an ounce when that move reaches the diesel. Agnico Eagle has bought most of the diesel for its Nunavut mines through the 2027 sealift, at a price about 30 percent above its 2026 budget after hedges. Oil can fall on a screen and that diesel does not get cheaper. A miner that locked in fuel does not get September’s gold bounce until the gold price clears the fuel, the royalty, and the grade.
That is why gold mining stocks can keep falling after the spot gold price stabilizes. The shares are a claim on the margin, not on the ounce. In a rising market the margin widens faster than the ounce, which is why miners beat gold on the way up earlier in this cycle. GDX compounded far faster than GLD over three years in one late-September comparison, about 49 percent a year against about 30 percent. The same gear took more back in the latest month. Investors who want the gear only on the way up want a product that does not exist.
Canadian gold mining stocks are a large part of that gear, not a side bet. Agnico and Barrick sit near the top of the big-miner fund, with Kinross, Wheaton, and Franco-Nevada also heavy. Canadian gold stocks are therefore a way many global portfolios own the gold price outlook without meaning to own Canada. They also own Canadian costs, Canadian diesel, and Canadian royalties. A rebound in the ounce helps them. It does not erase a fuel bill already signed, or a yield that pulls generalist money back toward bonds.
Junior gold stocks are the same gear with less ballast. A junior has fewer mines, thinner cash, and a share price that is a vote on a story. Gold exploration stocks often have no production at all. Gold development companies have a project and a bill. When the gold price correction is only 6 percent, a producer can still be profitable. An explorer can still be a promise that the market refuses to fund until the gold price recovery is obvious. The order is cruel and regular. The metal stabilizes. The big miners stop making new lows. Only then do the juniors get a bid that lasts more than a day. Buying the junior because the senior bounced is how the last rally’s late money gets trapped.
What could trigger a gold stock rebound
A gold stock rebound has a trigger, and the trigger is not the first of the month. It is a change in the thing that made September red. Three changes would qualify. None of them is guaranteed. All of them can be watched.
The first is bond yields and gold moving back into the old relationship. Yields do not have to collapse. They have to stop rising. A 10-year that churns near 5.3 percent instead of pushing to 5.5 is a different market from a 10-year that makes a new high every week. Gold mining investment is a long-duration bet. A higher real yield discounts the future ounces. When that discount stops getting worse, the shares can rise even if gold is only flat, because the multiple stops shrinking. If yields keep rising, flat gold is not enough. The multiple keeps shrinking, and the gold mining stocks lag.
The second is the dollar. US dollar and gold have been a one-way trade into October, with the dollar taking the safety bid. A dollar that fails at the year’s high, or that falls after the jobs report, gives the metal room. A metal that reclaims $4,200 gives the miners room. The sequence is dollar, then metal, then shares. People invert it. They buy the shares and hope the dollar cooperates. Sometimes it does. The times it does not are the drawdowns of 10 percent in a fund that was supposed to be the rebound.
The third is the mine itself. Costs that stop rising, a fuel market that finally delivers cheaper diesel to the site, a quarter that shows margins held even at $4,100. That is slower than a jobs print. It is also more durable. Gold mining companies that can show all-in costs well under the spot price do not need a new high in gold to be solvent. They need the price not to fall through those costs. At $4,100, most large producers are still far above their costs. The equity damage in September was about the multiple and the fear of a further fall, not about a sudden loss. That distinction is why a rebound can be violent if the fear breaks, and why it can also be a trap if the fear was early rather than wrong.
What would not be a trigger is a headline that restates the level. “Gold starts October above $4,100” is a description. It is not a gold price forecast. It is not a gold investment outlook. A description can be true and still be the middle of the range. The range is $4,000 to $4,200 until one side breaks. Gold support and resistance are the trigger lines. The calendar is not.
The bull case that survives a red month
The long case was never “every month is green.” It was that official buyers, fiscal strain, and a choppy trust in currencies would keep a bid under the metal that private traders could not fully cancel. Central bank gold demand is the part of that case that does not mark to market every afternoon. HSBC still thinks those buyers step in on weakness. A Korean tonne, a Chinese addition in earlier months, and a reserve manager who would rather hold a bar than add more long bonds at the top of a yield spike are not the same trade as a futures short covering into the jobs report. They are slower. They are why a gold market outlook can be ugly for a month and intact for a decade.
The fiscal side is the yield itself. A 10-year near 5.3 percent and a 30-year near 5.6 percent are not only bad for gold. They are a price the bond market is charging governments. If that price is about deficits and not only about the next Fed move, then gold’s competition with the bond is real, and so is gold’s role as the asset that is nobody’s liability. Both can be true, which is why the metal can fall while the long case remains. Gold market sentiment in October is the fall. The case is the liability. Do not mix them and then claim the case failed because the sentiment was bad.
There is a version of the next rally that does not need a new record. A climb from $4,100 back to $4,300 would be a large move for the miners, given a beta near 1.6, if it stuck. GDX does not need $5,400 gold to rise 15 percent. It needs the fall to stop and the margin to be believed. That is a smaller claim than “the high is back.” It is also the only claim the October 1 tape can support. Anything louder is a gold price prediction wearing the clothes of a fact.
The bear case that a bounce does not kill
The bear case is not silly, and it fits the week. Real and nominal yields are high. The dollar is firm. Oil has been expensive enough, in a war year, to keep an inflation scare alive even after one soft PCE print. Gold failed to hold a rally on bullish news. Positioning that was built for $5,000 does not become a fresh bid just because the price is $4,160. If jobs are strong, $4,000 is a magnet. If $4,000 breaks, the mid-June lows come into view, and the gold price correction becomes something people will stop calling a dip.
For the shares, the bear case is the gear plus the cost. A further 6 percent on the metal can be 10 percent or more on GDX, and more than that on junior gold stocks. Canadian gold stocks with northern diesel bills feel a second cut if energy stays high while gold falls. Explorers feel a third cut if financings dry up. None of that requires the long-term case to be false. It requires the next hundred dollars to be down instead of up. Gold price volatility is how that hundred dollars arrives, usually faster than a note can be rewritten.
A sensible reader holds both cases in one hand. The long case is about who buys a bar when trust thins. The short case is about the hurdle rate this month. October 1 did not settle the argument. It opened above a number that both sides can live with. Bulls call $4,100 proof of demand. Bears call it a pause under $4,200. Gold technical analysis that picks one label today is guessing the jobs report.
How to read a rebound if one starts
If the shares jump, ask three questions before you call it the turn. Did gold close back above $4,200, or did the shares jump alone. A share jump alone is a squeeze. Did yields fall, or did they rise and the shares rose anyway. The second one can happen for a day and then reverse. Did the dollar fall. If the dollar made a high and the shares made a high, one of those highs is lying.
Then ask a fourth question that is about the business. Is the move in the large gold mining companies, which have cash and mines, or only in gold exploration stocks, which have neither. A healthy gold price recovery shows up first in the metal, then in the biggest producers, then in gold development companies that have a path to a mine, and last in the explorers. A move that starts at the bottom of that list and never reaches the metal is a story rally. Story rallies end when the next macro number hits.
Gold mining investment is a choice about that order. Some investors want the bar, because the bar has no diesel bill and no government that can reopen a contract. Some want the big miners, because the beta pays them if the recovery is real and the balance sheet lets them survive if it is not. Some want the juniors, because the percentage gain is larger if they are early. Early is the word that hides the graveyard. The September tape is the reminder. The juniors and the seniors both fell. The seniors had revenue. The juniors had a beta. Match the tool to the claim you are actually making about the next three months, not to the claim you wish you had made in January.
What this week can still change
Friday’s payrolls report is the next hard fact. A weak number would fit the softer inflation print and could push Fed rate hike expectations down again. If the long bond follows, bond yields and gold can relax together, and a test of $4,200 becomes more than a wish. If the long bond does not follow, the week’s lesson repeats. The front end eases. The hurdle stays. The metal pops and fades. Gold market sentiment would look healed for a morning and be the same market by the close.
A strong number does the opposite. Hike odds rise. The dollar gets another reason. $4,110 is the first line. $4,000 is the line HSBC and several chart readers already treat as the area where official demand might answer. Central bank gold buying is not a stop-loss you can see. It is a behavior that shows up after the price has already scared the private money. Waiting for the central bank to announce the low is how the low gets missed. Assuming the central bank will defend $4,000 before it is tested is how support becomes a story.
Between those outcomes, the gold investment outlook for a long holder and a trader are different jobs. The long holder cares whether $4,000 still looks like a place official demand appears, and whether miners can earn at that price. The trader cares whether $4,200 breaks this week. Mixing the jobs is how a long holder sells the low because a trader’s level failed, and how a trader holds a loss because a long case sounded noble. October does not require you to pick a side forever. It requires you to know which side you are on before the jobs number.
An average is not a target
One more trap sits in the gold price forecast itself. An average for a year is not a destination for October. HSBC’s $4,490 figure for 2026 includes the months when gold was far above today’s spot gold price. Cutting the average is a way of saying the autumn is softer than the bank once modeled. It can coexist with a view that the low is near. It can also coexist with another leg down that still leaves the year’s average intact, because January did so much of the work. A reader who treats $4,490 as “where gold is going next” has misread a calendar. A reader who treats the cut as “the bull market is over” has misread it the other way.
The same care applies to any gold price prediction that names a round number without a condition. “Back to $4,300” is a condition: $4,200 must break and hold, and the dollar must stop helping the bear. “Down to $4,000” is a condition: $4,110 must break, probably on a hot jobs print or a fresh high in yields. Neither number is a personality. Both are places on a map that already has the September damage drawn in. Gold technical outlook language earns trust when it states the condition in the same sentence as the number. It loses trust when the number is left to do the selling.
Global gold demand is the other average people abuse. A strong year for official buying does not mean every week was strong. A weak September for investment flows does not mean the official sector changed its mind. The wallets move on different clocks. Private futures can dump a month of inflows in two sessions. A central bank board meets on a quarter. Korea’s December plan, if it happens, will not show up in this Friday’s payroll reaction. Putting those clocks on one chart and calling it one bid is how a gold market correction gets narrated as a rejection of the whole cycle. The cycle can be fine and the week can still hurt.
The share is a claim on a margin
Return to the miner one last time, because the headline asks about stocks and not only about the ounce. A gold mining company turns an ounce into earnings only after cash costs, sustaining capital, royalties, and tax. At $4,100, that stack is still comfortable for the largest producers in most published cost curves. Comfortable is not the same as immune. A slide toward $4,000 narrows the stack. A slide that also lifts diesel, power, and labor narrows it from both ends. September did not do the second thing in full. It did enough of the first thing to reprice the multiple. Gold price volatility hits the multiple before it hits the cash cost, which is why the share can look broken while the mine is still printing cash.
That is the rebound people will feel first if it comes. Not a miracle quarter. A multiple that stops falling. Multiples stop falling when the fear of the next hundred dollars down fades. The fear fades when gold support and resistance resolve upward, and when gold interest rates stop tightening the story every afternoon. If those two happen together, gold mining stocks can move more than the metal, including Canadian gold stocks that the world already owns through the big funds. If only one happens, the move is a trade. If neither happens, the bounce above $4,100 remains what it was on the morning of October 1. A number. Not a turn.
Gold development companies sit one step further out on that fear. They spend now for ounces later. Later is exactly what a high long-term yield discounts. A project with a permit and a road is still a set of future ounces. Those ounces got less valuable in a month when the 10-year rose, even if the spot price only fell 6 percent. A rebound in development shares that is not matched by a calmer bond market is a squeeze in a story. A rebound that is matched by a calmer bond market is the start of a financing window. Windows close. The September window closed on anyone who needed fresh equity at the old multiple. The next window opens when the metal and the yield agree, not when a headline says October.
The close
Gold started October above $4,100 after a September fall of more than 6 percent. The spot gold price was near $4,160 to $4,180 on the first day, well under a January closing high of $5,405, and above the late-September washout near $4,110. The news that should have helped, a soft inflation print and October hike odds near 31 percent, produced a rally that did not last. Yields near multi-decade highs and a firm dollar took it back. That is the gold market outlook as the month opens. Not a collapse. Not an all-clear.
Gold mining stocks fell harder than the metal, which is what a beta near 1.6 means. Canadian gold mining stocks and the big producers can rebound if $4,200 is reclaimed and yields stop rising. Junior gold stocks and gold exploration stocks need that to happen first, and then some. A gold price forecast that skips those steps is a wish. HSBC’s lower average for the year still sits above the spot price, and still allows for more near-term pain. Central bank gold demand is the slow bid. It is not this week’s trigger.
The theme is the filter. Do not treat a hold above $4,100 as a gold price recovery, and do not treat a red September as the end of the larger bull case. The selloff is over only if broken levels are taken back and the yield stops being the boss. A gold stock rebound is a consequence of that, not a calendar event. Watch $4,000, $4,110, and $4,200. Watch the dollar. Watch the long bond. The first of the month is not a level.
Important information
This article is for information and education only. It is not investment advice and not a recommendation to buy, sell, or hold gold, silver, shares, or any fund. Prices move. Forecasts miss. Mining shares can fall even when gold rises. Past rallies do not predict the next one.
Spot and futures levels for October 1, 2026, the September decline of more than 6 percent, the October hike odds of about 31 percent versus about 69 percent a week earlier, and HSBC’s average forecasts of $4,490 for 2026 and $4,825 for 2027 are drawn from same-day wire reporting, including Reuters. The January 29 closing high of $5,405 and nearby daily closes are from published daily bullion series and can differ from intraday peaks. GDX figures are from the fund’s September 30 data and from published performance comparisons. Yield levels differ slightly across sources and are described as a range. Company cost comments are from those companies’ own disclosures, not from this article’s model. This article does not consider any reader’s finances.