1. Bottom Line & Directional Bias
Call: Bearish silver (SI=F). Invalidation: a settle above pivot P 60.95 that is held on the following session; hard invalidation at R1 61.92.
Three reasons drive the call. First, the trend structure is broken on every horizon that matters: settle 60.42 (2026-10-02) is -6.77% over five sessions and -10.77% over twenty, and sits at the 4.8% position of the 59.99–68.98 twenty-day channel — one of the weakest readings possible without having already broken the floor. Second, positioning has already de-risked rather than built a base: managed-money net length dropped to 7,614 lots on 2026-09-29, a weekly change of -5,695, with netPct at 7.11% and crowding at the 39th percentile of three years. There is no crowded long to squeeze; there is a trend-following community (CTA proxy +62) still positioned to sell rallies. Third, the curve charges longs to wait: M1–M2 at -0.227 (-0.38%) and roll yield -4.51% mean a long position bleeds carry while the market searches for a floor.
The house daily brief is also short (entry 61.92, stop 63.42, target 59.45). This deep dive agrees with the direction but not with the entry logic: selling strength at R1 assumes a bounce that may not come, while the data argue for selling into the 60.95–61.92 pivot band or on a confirmed break of 59.99. Invalidation is a settle above 60.95 that holds; a settle above 61.92 would force a full re-think because it would put price back inside the prior range.
2. Price Action & Technical Analysis
The settle of record is 60.42 on 2026-10-02, down 1.24% on the day (settle). The five-day change is -6.77% and the twenty-day change is -10.77% (settle). ATR14 is 2.25, which is 3.72% of price as a full daily range — this is a high-volatility tape where a single session can travel more than two dollars. RV20 is 33.1% annualized, confirming that realized movement is running hot.
The twenty-day channel is 59.99 to 68.98, and the settle sits at the 4.8% position of that range (settle). The 52-week range is 45.39 to 121.3, so silver is trading in the lower third of its annual envelope but far above the 45.39 extreme. Pivot P is 60.95, R1 61.92, R2 63.42, S1 59.45, S2 58.48. Note the arithmetic: the settle at 60.42 is below P 60.95, so the market is trading on the bearish side of the pivot, and the first resistance is only 1.5 dollars above the settle. S1 at 59.45 sits just below the 20-day low of 59.99, which makes 59.45 the first real downside objective and 58.48 the second.
The Asia snapshot for 2026-10-05 07:00 shows last 61.1, +1.13% versus the settle, with a high of 61.23 and a low of 60.64 (Asia). This is an early-Asian bounce, and it is important not to over-read it: 61.1 is still below R1 61.92 and only marginally above P 60.95. A bounce that cannot reclaim the pivot is a selling opportunity, not a reversal.
The last completed weekly bar, 2026-09-28 to 2026-10-02, opened at 64.66, high 64.66, low 59.99, and closed at 60.42, down 6.77% w/w. That is a weekly bar that closed on its low with no lower wick of consequence — a completed bearish candle. The current week has no settled bar yet, so no weekly conclusion can be drawn from it; the only weekly statement the data support is that the last completed week was decisively weak.
View: bearish while below P 60.95; the 59.99 floor is the pivot of the whole setup, and a settle below it opens 59.45 then 58.48.
3. Supply-Demand Balance & Fundamental Drivers
COMEX registered silver stood at 101.1 Moz (3,144,564 kg) on 2026-10-01, down 8,709 (-0.3%) d/d. The prior two readings were 101.38 Moz (3,153,273 kg) on 2026-09-30 and 101.56 Moz (3,158,871 kg) on 2026-09-29. The direction is consistent: registered metal has been drawn down modestly across the three sessions shown. That is a mildly constructive physical signal, but the magnitude — roughly 0.45% over three days — is small relative to a market that has lost 10.77% in twenty sessions. Inventory draws of this size do not arrest a momentum decline.
SHFE warrant stocks were 1,490,865 kg on 2026-09-30, up 30,504 (+2.1%) d/d. This is the more interesting divergence: while COMEX registered metal is slipping, Shanghai warrants are building. A build in Asian deliverable stocks against a falling price is consistent with physical metal moving into the exchange system rather than being consumed, which is a soft signal for the near term.
The curve tells the same story from the carry side. M1–M2 is -0.227, or -0.38%, with roll yield at -4.51% and slope 0.2521. Silver is in contango, which is normal for a metal with storage and financing costs, but the practical implication is that a long position pays to hold. Contango is not a cap on price — it is a roll cost — but in a market with no visible prompt tightness, it removes one of the arguments for being long.
Macro transmits through two channels here. The US ten-year yield is 5.277, up 0.76% (2026-10-02), and DXY is 101.86, down 0.07% (2026-10-04). A 5.28% ten-year is a high real-cost backdrop for a zero-carry asset, and it is the single most important macro headwind for silver. The dollar is essentially flat, so it is not providing a directional offset. The gold/silver ratio at 68.9 is discussed in section 5.
View: physical draws at COMEX are too small to matter, Shanghai warrants are building, and the curve charges longs. Fundamentals do not offset the technical breakdown; they reinforce patience on the short side.
4. Positioning & Fund Flows
The CFTC managed-money series is the cleanest read on who is left in this market. Net length was 7,614 lots on 2026-09-29, down 5,695 from 13,309 on 2026-09-22. The two weeks before that were 13,124 (2026-09-15) and 14,386 (2026-09-08). So net length has fallen from 14,386 to 7,250 to 107,047. Longs fell from 21,148 to 16,886 and shorts rose from 6,762 to 9,272. This is not a market where longs are trapped; it is a market where longs have been leaving and new shorts have been arriving.
Crowding confirms it. netPct was 7.11% on 2026-09-29 with a crowding percentile of 38.93, down from 12.5% and 55.01 on 2026-09-22. A 39th percentile is not crowded on the three-year window, so the standard “crowded long squeeze” argument does not apply. The CTA trend-following proxy is +62 and has been unchanged across all four weeks — trend followers remain net short-biased in this market, and their mechanical behavior is to sell into strength and add on breaks. Hedge pressure is 24.83%, essentially stable.
The volatility surface is the one caution. VXSLV is 36.9 (2026-10-02), down 0.33 points on the day, against RV20 of 33.1%. That is IV minus RV of +3.8 vol points, an IV/RV ratio of 1.11. Options are paying up modestly for event risk, which means short-dated downside protection is not cheap. For a short position this argues for defined-risk structures rather than naked exposure into the FOMC minutes.
View: positioning is clean rather than crowded, trend followers are still sellers, and the only friction is a modestly rich implied-vol surface. Bearish.
5. Cross-Asset Relative Value
The gold/silver ratio is 68.9, with a one-year percentile of 71.43% and a three-year percentile of 23.81%. Read this carefully. A low three-year percentile means the ratio has spent most of the last three years higher than it is now — silver has been structurally strong against gold over that window. The one-year percentile at 71.43% says the ratio has risen over the past year, i.e. silver has recently been losing ground to gold. The combination describes a market that had a strong multi-year silver leg and is now in the early phase of giving some of it back. That is consistent with the price action: silver is down 10.77% over twenty sessions while gold implied vol (GVZ 23.23, 15th percentile) is subdued. Silver is the high-beta expression of the precious complex, and when the complex is under pressure from a 5.28% ten-year yield, silver falls harder.
There is no copper/gold ratio in the feed, so no pro-growth read is available from that channel. The VIX at 15.31 (15th percentile) and OVX at 51 (49th percentile) tell a broader story: equity vol is low and oil vol is mid-range, so this is not a systemic risk-off event. Silver's decline is idiosyncratic to the precious complex and to its own positioning unwind, not a broad liquidation. That matters because it means the downside is not being cushioned by a flight-to-quality bid.
View: the ratio's three-year percentile says silver has been the strong leg and is now mean-reverting; with no systemic risk-off bid, silver remains the weaker expression of the complex. Bearish.
6. Historical & Seasonal Patterns
The seasonality block covers the same calendar start over the next twenty sessions for the last fifteen years: mean +0.95%, median +0.95%, up in 8 of 15 years. Best case 2011 at +11.91%, worst 2012 at -7.23%. The sample is small and the distribution is wide — the best and worst outcomes are both far larger than the mean. A 53% hit rate with a +0.95% median is a mild positive seasonal tilt, but it is not a directional edge, and it is dwarfed by the current ATR of 2.25 per day (3.72% of price). Twenty sessions of seasonal drift at +0.95% is less than half of one day's expected range.
View: seasonality is a mild tailwind for longs and therefore a mild headwind for this short call, but it is statistically too weak to override the trend, positioning and carry evidence. It is context, not a reason to trade.
7. Scenario Analysis (Base / Bull / Bear)
Base case — 55% — floor breaks, grind lower. Trigger: a settle below the 20-day low of 59.99, ideally with the Asia bounce failing below P 60.95 first. Target: S1 59.45, then S2 58.48. Action: hold or add to shorts on the break, trail stops above the pivot. This is the path most consistent with section 1: the settle is already at the 4.8% position of the channel, the last completed weekly bar closed on its low, and trend followers remain sellers.
Bull case — 25% — bounce reclaims the pivot. Trigger: a settle above P 60.95 that holds, with the Asia strength extending through R1 61.92. Target: R2 63.42. Action: stand aside on shorts, do not chase; a reclaim of 61.92 would invalidate the bearish structure and put the 64.66 weekly high back in play. The probability is not trivial because IV is running 3.8 points above RV, meaning the options market is pricing real two-way risk, and the seasonal window is mildly positive.
Bear case — 20% — acceleration. Trigger: a decisive break of 59.99 with rising volume and a further drop in managed-money net length below the 7,614 low. Target: 58.48, with an extension toward the lower 50s if the ten-year yield pushes further above 5.28%. Action: press shorts, but respect that ATR is 2.25 — a two-dollar adverse session is normal and stops must sit beyond real levels, not inside noise.
Probabilities sum to 100%. The base case agrees with the section 1 call: bearish, with 59.99 as the pivot.
8. Trading Strategies & Risk Management
Strategy 1 — Short into the pivot band (primary). Entry 60.95 (pivot P), stop 63.42 (R2, beyond the 61.92 R1 and roughly one ATR above entry), target 59.45 (S1), horizon 1-5 days, conviction 7. Size at half normal risk because the entry is a limit into a bounce that may not fill; if unfilled, do not chase.
Strategy 2 — Short on a confirmed break (secondary). Entry 59.9 on a settle below the 59.99 twenty-day low, stop 61.95 (above R1 61.92), target 58.48 (S2), horizon 1-5 days, conviction 6. This is the higher-probability trigger but the worse price; it should be sized smaller than Strategy 1.
Risk management: ATR14 is 2.25 (3.72% of price), so both stops are placed beyond real levels rather than at round numbers. The FOMC minutes on BJT 10-08 02:00 is the event that can gap this market; consider reducing size into it or expressing the view with defined-risk options given IV is 3.8 points above RV. Do not add to shorts if the settle reclaims 60.95.
9. This Week's Data Calendar
BJT 10-05 22:00 | ET 10-05 10:00 — ISM Services PMI (SEP), forecast 54, previous 55.4; surprise if outside forecast ±1.4. Affects GC, SI, DXY. BJT 10-08 02:00 | ET 10-07 14:00 — FOMC Meeting Minutes; affects GC, SI, DXY. Crude inventory releases on BJT 10-07 affect CL and BZ only and are not silver drivers.
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.