Saturday Night Thoughts: Up on the year, far below the peak

Gold is higher than it was a year ago. Silver is higher by considerably more. That sounds like a straightforward story—until you look at the road between the two dates.

In the stored references used for this review, gold went from $3,886.84 to $4,140.93 per troy ounce, a 6.54% increase. Silver went from $47.98 to $60.39, up 25.86%. But both finished this review below their end-of-2025 references, and a long way below their January peaks.

That is the central distinction: an anniversary gain describes the destination, not the experience of getting there.

These figures compare MetalpriceAPI observations on October 3, 2025 at 4:00 p.m. CDT with October 3, 2026 at 4:59:50 p.m. CDT. They occupy the same calendar-hour bucket, not the same second. Saturday's endpoint is a stored provider reference—not an exchange settlement or a claim that the market was trading then. This historical series is separate from the Aurora board used in our regular briefs.

One year, three very different answers

All four rows can be true at once. Someone remembering last October, someone remembering New Year's Eve, and someone remembering late January are anchoring to three different markets. The arithmetic does not contradict itself; the starting points differ.

“End of 2025” here means the final stored Central-Time calendar-hour reference: December 31 at 11 p.m. CST—$4,318.26 for gold and $71.58 for silver. It is not an official year-end closing price. The peak and drawdown rows use the highest and subsequent lowest observations in our hourly sample, not every traded tick.

Both lines begin at 100. The chart uses the stored hourly representatives, with separate line segments across the missing September 25 hours. It does not interpolate missing prices. January peak annotations are sampled reference highs.

The dated record behind the chart

The headlines below establish what was announced or subsequently measured. They do not prove that a particular announcement caused a particular hourly price move. Publication dates matter: a report written after the event is evidence for a retrospective, not information investors necessarily had beforehand.

September 17, 2025 — The easing backdrop. Just before our price window begins, the Federal Reserve's Federal Reserve issues FOMC statement announced a quarter-point cut, taking its target range to 4.00–4.25%. This gives the late-2025 policy story a concrete starting point. Federal Reserve, September 17.

October 29 and December 10, 2025 — Two further cuts. The Fed lowered the range to 3.75–4.00% in October and 3.50–3.75% in December. Thus the rise in our late-2025 metal references took place alongside a documented sequence of easing decisions. “Alongside” is important: these releases establish policy changes, not a controlled explanation of the metals' entire advance. October statement, December statement.

January 29, 2026 — The 2025 demand story gets its accounting. The World Gold Council's Gold Demand Trends: Q4 and Full Year 2025 reported total demand including over-the-counter activity above 5,000 tonnes, and 53 new highs in the LBMA PM gold-price benchmark during 2025. It also recorded an 801-tonne increase in ETF holdings and 863 tonnes of central-bank purchases: investment vehicles and reserve buyers both belonged in the story. These are retrospective industry figures published in January, not conclusions reconstructed from our incomplete January-to-June 2025 archive. World Gold Council, January 29.

April 15, 2026 — Silver's supply story survives the volatility. The Silver Institute's release carried the headline Elevated Lease Rates, Regional Liquidity Tightness, and Robust Investor Interest Resulted in Record Silver Prices in 2025. Its survey estimated a 40.3-million-ounce deficit in 2025, the fifth consecutive deficit year. The April release forecast a 46.3-million-ounce deficit for 2026—a forecast at that date, not a completed-year result. This is important context for silver, but an annual supply-and-demand estimate is not a promise of uninterrupted price appreciation. Our subsequent sampled drawdown makes that distinction tangible. Silver Institute's April 15 release, original survey, PDF page 11.

July 1, 2026 — Looking back at the surge and reversal. The World Gold Council's Gold Mid-Year Outlook 2026: Point break describes gold exceeding $5,500 intraday in late January and falling below $4,000 in late June. Those outside observations independently support the shape of the journey visible in our data. They are not substituted for our hourly sampled high or trough, and the July report is clearly hindsight—not a headline supposedly published at the January peak. World Gold Council, July 1.

September 16, 2026 — The policy direction changes. The Fed announced a quarter-point increase to 3.75–4.00%. Compare that decision with the three cuts cited above: by the end of this review window, simply repeating the late-2025 easing narrative would miss an actual policy turn. Federal Reserve, September 16.

The sequence is more useful than a collection of bullish or bearish headlines. It keeps the late-2025 backdrop, the January acceleration, the subsequent drawdown and the later policy setting in their proper order.

The course of the year, in numbers

The final stored October 2025 calendar-hour references were $4,002.73 gold and $48.68 silver. By the end of December they were $4,318.26 and $71.58. Silver's move was becoming much larger than gold's—not merely the same dollar change on a smaller price tag.

Then came January. Our highest sampled gold reference was $5,562.73 on January 29 at 1 a.m. CST. Silver's was $119.51 that day at 9 a.m. CST. From the opening references of this review, those represented gains of 43.12% and 149.08%, respectively. They were not simultaneous observations, and neither should be mistaken for the highest transaction price recorded anywhere in the market.

Month-end figures alone conceal some of that drama. January ended in our sample at $4,892.81 gold and $85.20 silver; February ended higher, at $5,278.59 and $93.77. This was not a single peak followed by a smooth decline. There were rebounds strong enough to change how the story felt along the way.

These are calendar references, not settlements: 11 p.m. CT for the backfill rows; September 2026 uses the last stored observation at 11:59:45 p.m. CDT. January 29's within-month peaks are deliberately described separately.

The largest sampled gold decline ran from that January high to $3,962.90 on June 29 at 10 p.m. CDT. Silver's ran to $54.91 on July 16 at 10 p.m. CDT. A nearly 29% gold drawdown and a roughly 54% silver drawdown are substantial even in a period that ends positive against its opening reference.

August's recovery did not erase the earlier damage, and the September references weakened again. That is why “still higher than last year” is useful context but an incomplete description of the market.

What silver's outperformance does—and does not—tell us

The gold/silver ratio fell from 81.01 to 68.57 across the two October 3 reference points. Put plainly, one ounce of gold corresponded to fewer ounces of silver at the end than at the beginning. Silver outperformed gold between those endpoints.

Yet silver also experienced the much deeper sampled drawdown. Outperformance and stability are different questions. A metal can lead over twelve months while being far harder to hold through the intervening path.

That distinction matters for interpreting a headline about a supply deficit, a rate decision or a new record. Such facts can help explain the environment. They do not provide a floor under every subsequent quotation, and they do not tell us what price a particular person paid, received or could have transacted at.

The Saturday-night perspective

The lesson I take from these numbers is not that one metal “won,” or that a fall from a peak automatically makes something cheap. It is that every price claim needs a reference point.

Ask three questions together: Compared with when? What happened between the endpoints? And what evidence explains that period?

For this review, the answers are unusually revealing: positive annual changes, negative changes from the end of 2025, and large retreats from January's sampled highs. The same chart contains all three stories. Keeping them together gives a more honest picture than selecting whichever comparison best suits today's mood.

This is a historical review, not a new trading forecast. It does not replace or score any existing Market Desk outlook.

Data and calculation notes

All price calculations use full-precision stored MetalpriceAPI USD-per-troy-ounce observations, rounded only for display. Percentage change is (later / earlier − 1) × 100. The gold/silver ratio uses simultaneous observations. News-source quotations, futures prices and Aurora dealer quotes are not spliced into this series. These are reference-price changes, not an investor's net return or a customer's settlement value.

The requested year contains 8,744 of 8,761 hourly buckets for each metal: 8,547 backfilled references and 197 hourly representatives of live collection. There is one 17-hour gap, September 25, 2026 from midnight through 4 p.m. UTC inclusive (September 24 at 7 p.m. through September 25 at 11 a.m. CDT). The sources remain identified, and the gap stays unfilled. Hourly sampling and missing data can miss other highs and lows; bucket presence does not prove complete tick coverage.

The newly imported 2025 history begins in June, not January. This review therefore does not claim a locally reconstructed full-calendar-2025 return. Any full-year demand or market statistics cited from industry reports are separately attributed to those reports.

Public series: Gold reference observations and silver reference observations, retrieved October 3, 2026 at approximately 8:15 p.m. CDT.