📅 2026-10-09 · No.2 Soybeans · Single-name brief Legacy coverage

This report is provided by MacroTerms.com

Data as of: 2026-10-08 · Settlement: Exchange settlement (15:00 Beijing close, vs prior settlement)

DCE.B Soybean Meal In-depth Report | Position-cut rebound meets 20-day channel lower edge, neutral watch | 2026-10-09

DCE.B Soybean Meal directional conclusion is neutral watch: the market shows a position-cut rise (short covering) near the 20-day channel lower edge, but 5-day and 20-day changes remain negative, lacking confirmation of active long position increases. Three core reasons: first, [10-08 settlement] 4086 yuan/ton, 5-day -1.02%, 20-day -1.48%, 20-day channel 3996–4222 position only 28.8%, the rebound is a weak repair; second, main holdings 142005 lots down 8855 lots, main nominal holdings down 360 million yuan on the day, the position-cut rise is characterized as short covering rather than trend reversal; third, US soybean crush margin 2.5174 USD/bushel (2026-10-07) at 1-year percentile 60.71%, international market ZS=F settlement 1287.5 cents, 20-day -3.36%, cost side has no upward drive. Direction invalidation conditions: if settlement price effectively stands above pivot P 4075.33 with main position increases, the short-covering logic is void and turns bullish; if S1 4032.67 is breached with volume, look down to the 20-day low 3996.

1. Bottom Line & Directional Bias

Directional conclusion: Neutral, stay on the sidelines.

Reason 1: The price structure remains weak. [10-08 settlement] CNY 4086/t, +0.17% versus the prior settlement, but the 5-day change is -1.02% and the 20-day change is -1.48%. The position within the 20-day channel of 3996–4222 is only 28.8%, in the lower third of the channel; the rebound has not yet altered the medium-term downward slope.

Reason 2: Positioning does not support a trend reversal. Main contract open interest stands at 142,005 lots, down 8,855 lots on the day; main contract notional open interest is CNY 5.77bn, down CNY 360m on the day. Prices edged higher while open interest fell, which we characterize as short covering rather than active long building. The top-20 seats show a net long of 7,870 lots (+8,389 lots d/d); net longs have recovered but the absolute level remains small, insufficient to confirm long dominance.

Reason 3: No upward driver from international markets or the cost side. ZS=F soybeans [10-08 settlement] 1,287.5 cents/bu, -3.36% over 20 days, at 23% of the 20-day channel; US soybean crush margin is USD 2.5174/bu (2026-10-07), at the 60.71st percentile over one year—mid-to-upper range but with no expansion momentum.

Directional invalidation conditions: If the settlement price effectively holds above pivot P 4075.33 and main contract open interest turns to increase, the short-covering thesis is voided and the bias shifts to bullish. If the settlement price breaks below S1 4032.67 with rising open interest, the bias shifts to bearish, with downside to the 20-day low of 3996.

2. Price Action & Technicals

[10-08 settlement] CNY 4086/t, +0.17% versus the prior settlement; intraday full-range ATR14 is 74.21 (1.83%), RV20 is 20.1%. Realized volatility exceeds the daily move implied by ATR, indicating that single-day swings have widened recently. The 5-day change is -1.02% and the 20-day change is -1.48%; both windows are negative, so the medium-term direction is unchanged.

The 20-day channel is 3996–4222, with the current settlement at 28.8% of the channel, near the lower band. Pivot system: P 4075.33, R1 4103.67, S1 4032.67. The settlement of 4086 is about 10.7 points above P, less than half an ATR (37.1 points); therefore “price near the pivot” does not constitute a trading signal. R1 is 17.7 points from settlement and S1 is 53.3 points away; neither has opened up meaningful space, so short-term pivot levels have limited reference value.

Last five settlement days: 09-24 closed 4143, 09-28 closed 4021, 09-29 closed 4036, 09-30 closed 4069, 10-08 closed 4061. On 09-28 the price fell from the 4135 high to 4021 in a single session, the key breakdown candle of the recent move. Since then, the lows of the following three sessions were 3996, 4019 and 4047—successively higher lows—but the highs of 4091 and 4118 also failed to break the 09-24 high of 4144, forming a converging pattern.

Weekly view: the last completed weekly candle covers 2026-09-28 to 2026-09-30, opening 4134, high 4135, low 3996, close 4069, a weekly change of -1.79%—a bearish candle. This week (from 2026-10-08, only one session so far) has not closed; it is tentatively marked at 4061 (-0.2%), and no weekly conclusion should be drawn from it.

Night session: DCE.B Soybean No.2 [10-08] 21:05–23:00 opened 4069, high 4101, low 4068, closed 4072, +0.27% versus the day session close and -0.34% versus settlement. The night session spiked to 4101 before fading, failing to hold above 4100, consistent with the daily converging pattern.

3. Supply-Demand Balance & Fundamental Drivers

On inventories and balance sheets: WASDE, ICSG and other balance sheets are not connected, so no related figures are cited in this section. The core fundamental anchor available is the crush margin: US soybean crush margin of USD 2.5174/bu (2026-10-07), versus 2.3985 twenty trading days ago—an expansion on the period—with a 1-year percentile of 60.71% and a 3-year percentile of 86.77%. A crush margin in the high 3-year percentile implies that Chinese crush demand for US soybeans provides some floor support to bean prices, but the 1-year percentile is only mid-to-upper, so marginal momentum for further near-term expansion is limited.

International comparison: ZS=F soybeans [10-08 settlement] 1,287.5 cents/bu, -0.77% over 1 day, +0.27% over 5 days, -3.36% over 20 days; ATR14 is 21.13 (1.64%), RV20 is 18%, the 20-day channel is 1273.25–1335.25 with a position of 23%, and the 52-week range is 1001–1335.25. International markets are likewise near the lower band of the 20-day channel—marginally positive over 5 days but clearly negative over 20 days—consistent with the 5-day -1.02% and 20-day -1.48% in China's Soybean No.2. The synchronized weakness across international and Chinese markets indicates this correction is a cost-driven, broad-based decline rather than a China-specific move.

Term structure (international comparison): CONTANGO, with M1-M2 at -17.25 (-1.32%), roll yield (RY) of -7.91% and a slope of 8.625. The near month trades at a discount to the far month; under a contango structure the roll cost of holding the near month is negative, reflecting that the spot end is not tight and offering no support for a large near-month premium.

Macro transmission: DXY [10-08] at 102.11 (-0.12%), US 10-year Treasury yield at 5.23% (-0.87%). A slightly weaker dollar and lower rates are theoretically a mild positive for USD-denominated US soybeans, but the magnitude is limited, and international markets remain -3.36% over 20 days, indicating macro factors are not yet the dominant variable. China will release CPI and PPI on 10-14 (Beijing time 09:30), affecting the ferrous, nonferrous and chemicals sectors; the direct transmission to Soybean No.2 is limited, operating only indirectly through overall commodity risk appetite.

Taken together, mid-to-high crush margins, a contango structure in international markets, and synchronized weakness across international and Chinese markets all point to a lack of upward fundamental drivers; the price rebound stems more from short covering than from demand improvement.

4. Positioning & Fund Flows

Top-20 seats (aggregate for the instrument, 2026-10-08): longs 269,836 lots, +1,602 lots d/d; shorts 261,966 lots, -6,787 lots d/d; net long 7,870 lots, +8,389 lots d/d. The net direction flipped from the prior value to net long, with a large daily change, driven mainly by the 6,787-lot reduction in shorts while longs rose only modestly by 1,602 lots. On a net-direction basis, the current net long of 7,870 lots reflects a recovery driven by active short exit rather than large-scale long positioning.

Main contract notional open interest: DCE.B Soybean No.2 main contract notional open interest is CNY 5.77bn, down CNY 360m on the day; corresponding main contract open interest is 142,005 lots, down 8,855 lots, with the settlement price up +0.17% over the same period. Prices edged higher while open interest fell, which by our definition is a decline-in-open-interest rally (short covering). This combination implies the rally lacks fresh long capital to sustain it; the durability of the rebound depends on whether open interest turns to increase.

Volatility dimension: DCE.B Soybean No.2 RV20 is 20.1%, ATR14 is 74.21 (1.83%). In the international comparison, implied volatility indicators are: ^OVX (WTI implied volatility) 48.4 (-0.21 points d/d, 1-year percentile 41%), ^GVZ (gold implied volatility) 22.68 (-0.71 points d/d, 1-year percentile 12%), ^VXSLV (silver implied volatility) 37.74 (+0.45 points d/d), and ^VIX 15.41 (+0.33 points d/d, 1-year percentile 18%). Cross-asset implied volatility is broadly in the low-to-mid percentiles, with VIX and GVZ percentiles low, indicating the market is not pricing significant macro tail risk. Soybean No.2's own RV20 of 20.1% is at a moderate level, and the options side shows no event premium.

Risk indicators (international comparison): DD52w is 11.58%, DD20d is 3.75%, Sharpe30 is 1.0354, VaR95 is -1.5%. The 20-day maximum drawdown of 3.75% is larger than the mild 20-day price decline of -1.48%, indicating a sharp single-day drop occurred during the period; position management should leave room for gap risk.

V. Cross-Asset Relative Value & Global vs China

Cross-asset relative value snapshot (as-of 2026-10-08): Gold/Silver ratio 69.95, 1-year percentile 80.16%, 3-year percentile 26.72%; Copper/Gold ratio 1.58, 1-year percentile 96.83%, 3-year percentile 56.61% (alternative measure 1.61, 1-year percentile 99.6%, 3-year percentile 61.11%); Oil/Gold ratio 0.022, 1-year percentile 92.06%, 3-year percentile 47.88%; Crack spread 3:2:1 at USD 69.72/bbl.

The Copper/Gold ratio sits at a high 1-year percentile of 96.83% and a mid-range 3-year percentile of 56.61%, implying that growth expectations are strong on a 1-year horizon but neutral on a 3-year horizon. The Oil/Gold ratio, with a 1-year percentile of 92.06% and a 3-year percentile of 47.88%, likewise exhibits a 1-year strong, 3-year neutral structure. The Gold/Silver ratio is elevated at the 80.16% 1-year percentile but low at the 26.72% 3-year percentile, indicating that silver has been relatively strong versus gold over a 3-year horizon but relatively weak over a 1-year horizon. These ratios reflect the current state of macro risk appetite and growth expectations; they do not directly forecast future returns and do not constitute a direct driver for Soybean Meal.

Global vs China comparison: DCE.B Soybean Meal [10-08 settlement] CNY 4,086/t; ZS=F Soybeans [10-08] settlement 1,287.5 cents/bushel. A comparable alternative observation is directional consistency: the China market is -1.02% over 5 days and -1.48% over 20 days, while the international market is +0.27% over 5 days and -3.36% over 20 days. Over the 20-day window, both markets weakened in tandem; over the 5-day window, a slight divergence emerged, with the China market marginally weaker than the international market, though the magnitude is insufficient to support an arbitrage conclusion.

Term structure (international comparison): CONTANGO, M1-M2 at -17.25 (-1.32%), RY at -7.91%. The near month trades at a discount to the far month, coexisting with mid-to-high crush margins (1-year percentile 60.71%), pointing to a combination of ample spot supply and profitable crushing, which does not provide upward impetus for the Soybean Meal futures price. Synthesizing cross-asset and global vs China factors, Soybean Meal lacks independent bullish or bearish signals from relative value; direction remains dominated by its own positioning and cost side.

VI. Supply Chain & Seasonal Inventory (Historical ...)

This section uses only the year-over-year win rate and median change from the seasonal statistics block, without introducing other statistical measures.

Judging by the historical win rate for the same period, the early-October window shows no significant directional bias for Soybean Meal; the win rate does not lean one-sided, and the historical odds for longs and shorts are close to even. This is consistent with the neutral, wait-and-see conclusion in Section I: seasonality itself provides no directional premium and does not constitute a reason to add positions.

In terms of median change, the absolute median gain/loss for the same period is small, falling within the single-day full range characterized by ATR14 (74.21, 1.83%). This indicates that the seasonal driving force in this window is insufficient to trigger a trending move on its own. In other words, even if trading in line with the seasonal direction, expected returns would struggle to cover the cost of a normal intraday fluctuation.

Structurally, early October is a transition period between the advancing harvest of new crops in the Northern Hemisphere and the start of planting in South America. The dispersion of seasonal statistics is typically higher than the information content of the mean; the combination of win rate and median change is better used as a constraint against “adding positions against the trend” rather than as an entry signal.

Combined with the technical conclusion in Section II (settlement price 4,086 is less than half an ATR above the pivot P 4,075.33, with R1 4,103.67 and S1 4,032.67 failing to open up effective space), seasonal statistics and price structure corroborate each other: there is currently neither a seasonal tailwind nor a seasonal headwind, making it preferable to maintain a wait-and-see stance and wait for invalidation conditions to trigger.

Conclusion: Seasonal statistics do not support any directional bet; they only support the position discipline of “not chasing rallies and not selling dips.”

VII. Scenario Analysis

The three scenarios sum to a probability of 100%; trigger and target levels all follow the pivot and channel definitions established earlier, with no new key levels created.

Base case (probability 55%): Neutral, wait-and-see. Trigger condition: settlement price continues to trade in a narrow range between S1 4,032.67 and P 4,075.33, with main contract open interest showing no directional expansion. Target levels are the two ends of the range, i.e., P 4,075.33 above and S1 4,032.67 below. Corresponding action: do not establish directional positions; manage existing positions according to the discipline in Section VIII, with a focus on tracking whether main contract open interest can shift from a cut to a build. This scenario is consistent with the directional conclusion in Section I; the core rationale is that although the 10-08 settlement of CNY 4,086/t is +0.17% versus the previous settlement, the 5-day change is -1.02% and the 20-day change is -1.48%, with the position within the 20-day channel of 3,996–4,222 at only 28.8%. Moreover, main contract open interest of 142,005 lots fell by 8,855 lots on the day, and main contract notional open interest changed by -CNY 360m on the day, which is characterized as short covering rather than long building; the rebound lacks support from new capital.

Bullish case (probability 20%): Trigger condition: settlement price effectively stands above P 4,075.33 and main contract open interest turns to an increase, i.e., the invalidation condition described in Section I is met. Target levels: first R1 4,103.67, then the upper band of the 20-day channel at 4,222 for further reference. Corresponding action: shift from neutral to bullish; on a pullback that holds above P 4,075.33, build long positions in batches, with stop-loss and position sizing executed according to the risk-control framework. The prerequisite for this scenario is that open interest shifts from a cut to a build; if the price stands above P but open interest continues to decline, treat it as the base case and do not chase the long.

Bearish case (probability 25%): Trigger condition: settlement price loses S1 4,032.67 accompanied by an open interest build, i.e., the invalidation condition described in Section I is met. Target level: the 20-day low of 3,996. Corresponding action: shift from neutral to bearish; on a rebound that fails at S1 4,032.67, establish short positions. This scenario's probability is slightly higher than the bullish case because, as noted in Section II, the medium-term downward slope remains unchanged, and as noted in Section III, the international market ZS=F 20-day -3.

Common observation variables across the three scenarios: main contract open interest direction, and the validity of the settlement price relative to P 4,075.33 and S1 4,032.67. Switch scenarios as soon as any invalidation condition is triggered; do not front-run.

VIII. Trading Strategies & Risk Management

【Defined Strategy】None; there are currently no compliant strategies that have passed risk-control validation. Therefore, this section does not provide entry, stop-loss, target, or duration figures, nor does it involve lot sizes.

Wait-and-see condition: when the settlement price remains between S1 4,032.67 and P 4,075.33 and main contract open interest shows no directional expansion, maintain zero directional exposure and do not participate in range-bound scalping. The rationale, as noted in Section II, is that the settlement price of 4,086 is about 10.7 points above P, less than half an ATR (37.1 points), with R1 17.7 points and S1 53.3 points away from the settlement price; the point space is insufficient to cover transaction costs and slippage.

Re-entry trigger levels: first, if the settlement price effectively stands above P 4,075.33 and main contract open interest turns to an increase, this is viewed as the entry condition for a shift from neutral to bullish, with a pullback that holds above P as the intervention window; second, if the settlement price loses S1 4,032.67 accompanied by an open interest build, this is viewed as the entry condition for a shift from neutral to bearish, with a rebound that fails at S1 as the intervention window. The two conditions are mutually exclusive; the one triggered first takes precedence, and no direction is preset before either is triggered.

Stop-loss placement logic: stop-losses should be placed where the invalidation condition is proven wrong in the opposite direction, i.e., long stop-losses reference below S1 4,032.67 and short stop-losses reference above P 4,075.33, so that stop-losses share the same source as scenario switching and avoid being swept by random volatility within the range. Specific stop-loss distances, ATR multiples, and position ratios must follow the 【Risk Conversion】 validated by risk control; this report does not calculate or fabricate them independently.

Position sizing and add/reduce rules: initial position size is executed according to the proportion approved by risk control; positions may only be added if the price effectively breaks through the next target level in the direction of the open interest (R1 4,103.67 for longs, the 20-day low of 3,996 for shorts) and open interest increases in the same direction. If the price returns within the entry trigger level or open interest changes in the opposite direction, first reduce to a half position, then decide whether to liquidate based on whether the invalidation condition holds.

IX. This Week's Data Calendar

The two items above have no direct transmission to Soybean Meal, acting only indirectly through overall commodity risk appetite; if data releases overlap with validity tests of P 4,075.33 or S1 4,032.67, handle them according to the scenario rules in Section VII.

This report is generated automatically from public quantitative and macro data for research and market tracking only. It does not constitute any investment advice.

Single-name briefs Deep dive

SHFE GoldSHFE.AU 2026-10-09

Shanghai Gold (SHFE.AU) In-depth Analysis Report | Bullish Bias: Gold-to-Silver Ratio at Low Levels | 2026-10-09

DCE.B Soybean Meal In-depth Report | Position-cut rebound meets 20-day channel lower edge, neutral watch | 2026-10-09

Today's Risk Map

Single-name briefs Deep dive

Previous briefs

⟠ This month's strategy backtest