1. Price Action & Technical Analysis
Silver (SI=F) closed at 63.63 on 2026-09-25, down 1.17% on the session, extending a sequence of lower closes from the 65.93 print on 2026-09-22 (close, +0.16% that day). The five-day change is -2.29 as of 2026-09-24 and the 20-day change is -5.91, confirming a corrective phase rather than a trend reversal at this stage. The daily sequence 65.83 (2026-09-21, -1.1%), 65.93 (2026-09-22, +0.16%), 64.38 (2026-09-23, -2.35%), 63.46 (2026-09-24, -1.44%), 63.63 (2026-09-25, -1.17%) shows two failed attempts to hold above 65.9 and a decisive break of the 64.24 pivot (P: 64.24 on 2026-09-24).
On the pivot framework, the 2026-09-24 reference set gives P 64.24, R1 64.73, S1 63.48. Price at 63.63 sits between S1 and P, i.e. below the pivot but above the first support — a classic weak-neutral zone. The 2026-09-23 pivot set was higher (P 65.95, R1 67.03, S1 63.84), and the 2026-09-22 set was P 66.12, R1 67.2, S1 64.92. The systematic decline in pivot levels across the week (66.98 on 2026-09-21 to 64.24 on 2026-09-24) is a clean mechanical confirmation of the downswing. For 2026-09-25, the relevant intraday reference is the 63.48 S1 from the prior session; a sustained break opens the 63 psychological handle, while reclaiming 64.24 would neutralize the immediate bearish bias.
A 51.4% drawdown from the 52-week high implies silver is trading at roughly half its trailing-year peak, which is consistent with a market that has already absorbed a major liquidation. The 20-day drawdown is 6.93%, meaning the current pullback is modest relative to the annual range — this is a shallow correction within a larger, still-damaged structure. The 30-day Sharpe of 0.071 is essentially zero, indicating no persistent risk-adjusted edge in either direction over the past month.
Momentum and volatility: ATR readings have compressed from 1.702 (2026-09-21) to 1.625 (2026-09-22), 1.608 (2026-09-23), and 1.417 (2026-09-24). Falling ATR during a decline typically signals exhaustion of the impulsive leg rather than acceleration, and the 2026-09-25 move of -1.17% is inside one ATR, consistent with a controlled grind lower. Vol20 is 32.94%, elevated in absolute terms but below the panic regime implied by VXSLV at 37.8. The absence of an oversold oscillator reading means we cannot claim a washed-out condition; the honest read is that momentum is negative but not yet stretched.
Weekly and monthly context: with a 20-day change of -5.91 and a 5-day change of -2.29, the weekly candle is on track for a second consecutive down week. The monthly picture is dominated by the 51.4% drawdown from the 52-week high, which places silver in the lower half of its annual range. The key monthly support zone is the 63–63.48 band (S1 and round number); the key monthly resistance is the 65.93–66.12 shelf (recent closes and pivots). Until either edge breaks on a closing basis, the market is range-bound with a downward tilt.
2. Fundamental Drivers
The dominant fundamental driver on 2026-09-25 is the real-yield shock. The US 10-year Treasury yield (^TNX) printed 5.162 on 2026-09-24, up 0.94%, while the dollar index (DX-Y.NYB) was 101.25, up 0.15%. The headline flow explicitly attributes the silver pullback to “real yields surge amid rising oil prices and strong US PMIs” (ForexLive_Commodities, 2026-09-24 12:05:10). This is the classic macro bind for silver: strong growth data lifts nominal yields and the dollar, raising the opportunity cost of holding a zero-coupon asset, and silver — with roughly half its demand base industrial — is doubly exposed because strong PMIs also raise the prospect of tighter policy for longer.
The 5.16% 10-year yield is the single most important number in this report. At that level, the real yield (assuming inflation expectations near the revised UoM series due 2026-09-25) is restrictive, and silver's carry cost is punitive. Historically, silver struggles to sustain rallies when the 10-year is above 5% unless inflation expectations are rising faster than nominal yields. The 2026-09-25 calendar includes Revised UoM Inflation Expectations, which is the key swing factor: if expectations rise, real yields fall and silver gets relief; if they fall, real yields rise further and 63 is at risk.
Inventories are a secondary but relevant bearish input. SHFE warrants rose from 1,437,649 kg (2026-09-22, weekly change +14,333 kg) to 1,442,143 kg (2026-09-23, +4,494 kg) to 1,443,896 kg (2026-09-24, +1,753 kg). The cumulative weekly build of roughly 17,533 kg is modest in absolute terms but directionally negative — Chinese exchange inventories are rising into a price decline, which argues against a physical squeeze narrative in the near term. The COMEX number is the larger pool and its stability (no reported draw) removes a potential bullish catalyst.
On the supply side, the headline “Silver Mountain's Reliquias Mine Achieves First Concentrate Production in Peru” (SMM_EN, 2026-09-24 16:47:07) is a marginal bearish supply signal. First concentrate production at a new operation adds incremental mine supply to the medium-term balance. It is not a game-changer at the global scale, but in a market already contangoed and with rising visible inventories, it reinforces the absence of a near-term deficit shock.
We flag this as a gap: without ETF tonnage changes, we cannot confirm whether the 20-day decline is being driven by paper liquidation or by physical accumulation on weakness. The COT data (Section 3) suggests paper length is being reduced, which is consistent with ETF outflows, but this is inference, not observation.
Geopolitics: no specific geopolitical headline appears in the 48-hour window. The oil-price strength referenced in the ForexLive headline is the closest proxy — rising oil typically carries an inflation-hedge bid for precious metals, but in this instance the yield response is dominating. The net read is that the macro channel (real yields, USD) is currently overwhelming the geopolitical and inflation-hedge channels.
3. Positioning & Fund Flows
COT data through 2026-09-15 shows open interest of 103,745 contracts, with longs at 20,205, shorts at 7,081, and net length at 13,124, a change of -1,262 from the prior week. The four-week sequence is: 2026-08-25 net 14,073 (Δ +2,378), 2026-09-01 net 12,598 (Δ -1,475), 2026-09-08 net 14,386 (Δ +1,788), 2026-09-15 net 13,124 (Δ -1,262). The pattern is choppy but the latest print is a reduction, and the 2026-09-15 reading is below the 2026-08-25 level. Net length is being trimmed, not aggressively liquidated.
Crowding metrics: netPct was 12.65% on 2026-09-15, down from 13.93% on 2026-09-08 and 12.07% on 2026-09-01. The crowding score was 55.04 on 2026-09-15 versus 58.85 on 2026-09-08, 53.27 on 2026-09-01, and 54.54 on 2026-08-25. A crowding score in the mid-50s is moderate — not a crowded long, not a washed-out short. CTA positioning is static at 62 across all four weeks, which is notable: the trend-following community has not flipped, implying that systematic selling pressure is contained for now. Hedge pressure is 25.81% (2026-09-15), essentially flat versus 25.91%, 25.88%, and 25.12% in the prior three weeks — producer hedging is stable, not accelerating into the decline.
The combination of moderate crowding, static CTA exposure, and stable hedging suggests the current selloff is discretionary/macro-driven rather than a positioning cascade. That is a double-edged read: it means there is no forced-selling overhang that would produce a violent capitulation low, but it also means there is no short-squeeze fuel for a sharp rebound. The market can grind lower without mechanical interruption.
Options and volatility: VXSLV (CBOE Silver Volatility Index) was 37.8 on 2026-09-24, unchanged on the day, versus VIX at 15.67 (+3.23%). The silver-specific vol premium over equity vol is roughly 22 points, an elevated ratio that reflects genuine two-way uncertainty in the metals complex. A VXSLV near 38 with spot falling suggests put demand is being met by call overwriting rather than panic buying of downside. The unchanged VXSLV print on a -1.44% day is mildly encouraging for bulls — it implies the decline is orderly and not triggering a vol-of-vol spiral. If VXSLV breaks above 40 while spot loses 63, the risk of an accelerated move increases materially.
4. Cross-Asset Relative Value
The gold-silver ratio is 66.81, sitting at the 55th percentile over one year and the 18th percentile over three years. The one-year percentile says silver is roughly fairly valued versus gold relative to the past 12 months; the three-year percentile says silver is cheap versus gold relative to the past three years. The divergence is the key insight: the past year has seen the ratio normalize higher (silver underperform), but the three-year window still shows silver near the cheap end of its historical band. For a relative-value allocator, this argues for patience rather than chasing the ratio higher — the mean-reversion case for silver versus gold strengthens as the ratio approaches the upper end of its three-year range.
We note this as a limitation: without the copper-gold ratio, we cannot assess whether the industrial-metals complex is confirming or diverging from silver's industrial demand signal. The oil-gold ratio would inform the inflation-hedge channel. Both are material gaps in a complete cross-asset read.
The contango structure is a direct relative-value input. M1-M2 is -0.51 (-0.79%), and the roll yield is -9.5% with a slope of 0.2842. A -9.5% annualized roll yield is a significant carry cost for long holders of front-month exposure. In a contangoed market, physical holders are incentivized to sell forward, which caps rallies and reinforces the range-bound-to-lower bias. The slope of 0.2842 indicates the contango is not uniformly steep across the curve — the front is more penalizing than the back, which is typical of a market with near-term surplus and longer-term deficit expectations. For relative-value traders, the contango argues for calendar spreads (short front, long back) over outright length.
5. Sentiment & News Monitor
The 48-hour headline set contains two items. The first, “Silver Mountain's Reliquias Mine Achieves First Concentrate Production in Peru” (SMM_EN, 2026-09-24 16:47:07), is a supply-side headline with a mildly bearish tilt. The second, “Silver falls back to the key $63 support as real yields surge amid rising oil prices and strong US PMIs” (ForexLive_Commodities, 2026-09-24 12:05:10), is the dominant narrative and is explicitly bearish, framing the move as macro-driven and identifying $63 as the key level. The headline bias over the 48-hour window is therefore negative, with a 2:0 bearish-to-bullish skew in the available sample.
Sentiment score: we would characterize the tape as cautious-to-bearish, roughly 3.5/10 on a bull-bear scale, driven by the yield narrative and the failure to hold 65.9. The absence of any bullish counter-narrative (no ETF inflow story, no industrial demand shock, no geopolitical escalation) means sentiment has no offset. The one mitigating factor is that the decline is orderly — VXSLV unchanged, ATR compressing — which prevents a capitulation read. Net: negative bias, but not panic.
6. Historical & Seasonal Patterns
We cannot state with quantitative support that late September is historically weak or strong for silver, nor can we map the current setup to a specific historical analogue. What we can say from the supplied data is structural: the 51.43% 52-week drawdown places silver in a post-liquidation regime, and the 0.071 30-day Sharpe indicates no persistent momentum edge. Historically, post-liquidation regimes in silver have been characterized by extended base-building rather than V-shaped recoveries, but we flag this as a general observation, not a data-derived claim. The honest position is that seasonal and analogue analysis is pending, and any strategy relying on it should be sized accordingly.
7. Bull/Bear Scenario Analysis
Bull case (≥4 points):
- Real-yield peak: if the 2026-09-25 Revised UoM Inflation Expectations show an uptick while the 10-year holds near 5.16%, real yields compress and silver gets its primary macro relief valve. A reclaim of the 64.24 pivot would confirm.
- Gold-silver ratio mean reversion: at the 18th percentile over three years, the ratio has room to fall, and silver outperformance versus gold is the highest-probability relative-value trade if the metals complex stabilizes.
- Positioning is not crowded: netPct 12.65% and crowding 55.04 mean there is no long overhang to liquidate; static CTA at 62 removes systematic selling pressure.
- Inventory build is modest: the SHFE weekly build of ~17,533 kg is small relative to the 1.44 million kg base, and COMEX registered at 2.97 million kg shows no reported draw — but also no reported surge.
Bear case (≥4 points):
- Real yields at 5.16% with DXY at 101.25 are a persistent headwind; strong PMIs reinforce the higher-for-longer narrative.
- Contango with -9.5% roll yield penalizes long holders and incentivizes forward selling, capping rallies.
- Price is below the 64.24 pivot and the 63.48 S1 is the only support before 63; a close below 63 opens the 52-week drawdown channel.
- Supply headlines (Reliquias first concentrate) and rising SHFE warrants add incremental bearish physical signals.
- VXSLV at 37.8 with spot falling means downside protection is expensive but available; a break above 40 would signal acceleration.
Near-term balance (1-2 weeks): bearish-to-neutral. The path of least resistance is a test of 63, with 63.48 S1 as the first gate. A close below 63 targets the lower end of the recent range; a reclaim of 64.24 shifts to neutral.
Medium-term balance (1-3 months): neutral-to-bullish, contingent on a real-yield peak. The three-year gold-silver percentile and moderate positioning are the structural supports. If the 10-year peaks and rolls over, silver's high beta to the metals complex makes it the preferred expression on the long side.
8. Trading Strategies & Risk Management
Strategy 1 — Tactical short into support break. Direction: SHORT. Entry: 63.45 (on a close below the 63.48 S1). Stop: 64.25 (above the 64.24 pivot). Target: 61.8. Timeframe: 1-5 days. Size: 0.5x normal risk unit, given moderate crowding and the risk of a headline-driven reversal. Conviction: 6/10. Rationale: price is below pivot, ATR is compressing but the trend is down, and the macro narrative (real yields, USD) supports continuation. Risk: a dovish inflation-expectations print on 2026-09-25 could squeeze this position.
Strategy 2 — Relative-value long silver versus gold. Direction: LONG (silver/gold ratio). Entry: ratio 66.81. Stop: ratio 69.5 (approximately 4% adverse). Target: ratio 62. Timeframe: 1-3 months. Size: 0.75x normal risk unit. Conviction: 7/10. Rationale: the three-year percentile (18.25%) shows silver historically cheap versus gold, and the one-year percentile (54.76%) shows the ratio is not yet stretched. This trade is insulated from outright direction and benefits from silver's higher beta if the complex rallies. Risk: a continued real-yield grind higher would hurt silver more than gold, widening the ratio.
Risk management overlay: total portfolio risk on silver-linked exposure should be capped at 1.5x normal given Vol20 at 32.94% and VaR95 at -6.28%. The -6.28% daily VaR at 95% confidence means a single bad session can erase a week of carry. Position sizing should respect the 1.417 ATR — stops tighter than one ATR are likely to be noise-stopped. The 2026-09-25 UoM inflation expectations release is the key event risk; consider reducing size into the print.
9. This Week's Data Calendar
| Date | Time | Event | Importance |
|---|
| 2026-09-25 | 05:15 | BOE Gov Bailey Speaks | HIGH |
| 2026-09-25 | 10:00 | Revised UoM Consumer Sentiment | MEDIUM |
| 2026-09-25 | 10:00 | Revised UoM Inflation Expectations | MEDIUM |
| 2026-09-27 | 21:30 | China Industrial Profits ytd/y | MEDIUM |
| 2026-09-29 | 21:30 | China Manufacturing PMI | HIGH |
The UoM inflation expectations print is the highest-signal event for silver this week, given the real-yield narrative. The China Manufacturing PMI on 2026-09-29 is the key industrial-demand read for the metal's cyclical component.
This report is generated automatically from public quantitative and macro data for research and market tracking only. It does not constitute investment advice or a recommendation to trade.