I. Price Action & Technicals
As of the data cutoff date 2026-09-21, the latest available complete trading day for the SHFE.RB main continuous contract is 2026-09-18, with a close of 3096.00, a change of -1.0230%, volume of 884,858 lots, and open interest of 1,642,367 lots. This was the largest single-day decline among the past 5 trading days and the only trading day with a single-day drop exceeding 1%, indicating a concentrated release of short-term selling pressure.
Looking at the 5-day sequence, 2026-09-14 closed at 3095.00 (-0.7058%), 09-15 closed at 3110.00 (+0.4198%), 09-16 closed at 3126.00 (+0.7412%), 09-17 closed at 3122.00 (-0.1280%), and 09-18 closed at 3096.00 (-1.0230%). After touching the stage high of 3126.00 on 09-16, the price fell for two consecutive days, giving back a cumulative 30.00 points and essentially erasing the gains from 09-15 and 09-16. On a weekly basis, the latest close of 3096.00 was only 1.00 point below the week's opening level of 3095.00 on 09-14, leaving the week nearly flat, though the pattern shows an “upside spike and pullback” with an upper shadow.
On the moving average system, the 09-18 MA5 was 3109.80 and MA20 was 3126.80. The close of 3096.00 was below both MA5 and MA20, and MA5 (3109.80) has crossed below MA20 (3126.80), with the short-term moving average group forming the early shape of a bearish alignment. Comparing with 09-14, when MA5 (3139.00) was above MA20 (3109.65) by about 29.35 points, by 09-18 MA5 had shifted to about 17.00 points below MA20. The moving average structure completed a switch from bullish to bearish within 5 trading days, which is the most important technical signal in this reporting period.
On pivots and key levels, the 09-18 pivot P was 3105.67, first resistance R1 was 3122.33, and first support S1 was 3079.33. The close of 3096.00 was about 9.67 points below P (3105.67) and about 16.67 points above S1 (3079.33), placing it in the neutral-to-weak quadrant of the pivot framework. The 20-day high H20 was 3203.00 (maintained at that value over the past 5 days, indicating the high was formed on an earlier trading day), and the 20-day low L20 was 3034.00 (as of 09-18, up 40.00 points from 2994.00 on 09-14). The 20-day range was 169.00 points. Based on the 09-18 close of 3096.00, the price was about 62.00 points above the lower bound and about 107.00 points below the upper bound of the 20-day range, at roughly the 36.7% position within the range, skewed toward the lower bound.
At the indicator level, the raw values for RSI, MACD, and ATR were not provided in this dataset, and the data is pending update; however, a qualitative inference can be made from the relationship between price and moving averages: with the close below MA5 for two consecutive days and MA5 crossing below MA20, momentum indicators are most likely in a neutral-to-weak zone, not yet sufficient to confirm oversold conditions. On the volatility dimension, 09-18 volume of 884,858 lots was about 39.5% larger than 09-17's 634,449 lots, the highest volume in the past 5 days, and combined with the price decline this constitutes “volume-expanded decline,” with short-term volatility rising.
Key level conclusion: the first upside resistance is the 3122-3127 dense band formed by R1 (3122.33) and MA20 (3126.80), above which lies H20 (3203.00); the first downside support is S1 (3079.33), below which lies L20 (3034.00). If the price remains below P (3105.67), the weak setup continues; if it recovers MA20 (3126.80), the bearish structure is broken.
II. Fundamental Drivers
On the macro rates dimension, the U.S. 10-year Treasury yield ^TNX last printed 4.9980%, up 1.03% on the day, approaching the 5.00% round-number threshold. For Rebar, this transmission chain is not direct, but it exerts pressure through two paths: first, rising global risk-free rates suppress the overall valuation center for commodities, making it difficult for the ferrous complex to strengthen independently; second, a high-rate environment reinforces the relative appeal of USD assets, indirectly weighing on globally USD-denominated commodity pricing and affecting domestic risk appetite through sentiment channels.
On the dollar dimension, the U.S. Dollar Index DX-Y.NYB last printed 100.2540, +0.03% on the day, essentially flat above 100. With the dollar index holding the 100 mark, it exerts mild external valuation pressure on RMB-denominated domestic steel products, but the magnitude is limited — a 0.03% single-day move is insufficient to constitute a directional driver and is more a case of “not providing upward support.”
On the inflation dimension, this dataset does not provide CPI, PPI, or steel spot price index readings, and the data is pending update. Therefore, no quantitative judgment can be made on the “real rates-commodity relative value” chain; only an inference of a tight macro environment can be drawn from the absolute levels of rates and the dollar.
On the inventory and central bank gold buying dimension, this dataset does not provide Rebar social inventory, mill inventory, output, or apparent consumption data, nor does it include central bank gold buying readings, and the data is pending update. It should be noted that central bank gold buying belongs to precious metals pricing logic and has limited direct transmission to Rebar, so this report does not extrapolate across commodities.
On the ETF and fund flow dimension, the dataset does not provide domestic steel-related ETF shares or fund flow data, and the data is pending update.
On the geopolitical dimension, the dataset does not provide geopolitical event readings, and the data is pending update. Based on available information, VIX printed 14.81, -4.08% on the day, at a low level, implying that the market is not pricing much short-term geopolitical or macro tail risk. This echoes the absence of the ferrous complex's own supply-demand drivers — that is, the external environment neither provides a reason for panic selling nor offers a safe-haven premium.
In summary, during this reporting period, Rebar's fundamental drivers present a pattern of “externally tight, internally pending update”: rates and the dollar provide mild suppression, while core domestic variables such as supply-demand, inventory, and margins lack data support and cannot form a directional judgment. Therefore, price action is more dominated by technicals and flows, which corroborates the “volume-expanded decline, moving averages turning bearish” observed in Section I.
III. Positioning Structure & Fund Flows
On the open interest dimension, 09-18 open interest was 1,642,367 lots, up 57,705 lots from 1,584,662 lots on 09-17, an increase of about 3.64%; and up 35,234 lots from 1,607,133 lots on 09-14, a cumulative 5-day increase of about 2.19%. Over the same period, the price moved from the 09-14 close of 3095.00 to the 09-18 close of 3096.00, essentially flat. Price moving sideways while open interest keeps climbing is a typical “open interest build standoff” structure, with both longs and shorts adding positions and neither side yet capitulating.
On the volume-open interest coordination dimension, 09-18 volume of 884,858 lots was the highest in the past 5 days, and open interest of 1,642,367 lots was also the highest in the past 5 days, presenting a combination of “both volume and open interest rising, price falling.” In the futures analysis framework, falling price + rising volume + rising open interest is usually interpreted as new short money actively entering rather than long liquidation, with short-term shorts holding the marginal pricing power.
On the COT classified positioning dimension, this dataset does not provide classified long/short positioning for Rebar futures (such as futures company member long/short positions, or the split between hedging and speculative positions), and the data is pending update. Therefore, no quantitative characterization can be made of “speculative net short/net long” or “hedging share,” nor can crowding indicators be calculated.
On the crowding dimension, the available proxy observation is the absolute level and pace of change in open interest. The cumulative 5-day open interest build of 2.19% and single-day build of 3.64% indicate a relatively fast pace of position adding. If the price continues to decline while open interest does not fall, short crowding rises, creating the possibility of a rebound triggered by periodic covering; conversely, if the price rebounds while open interest falls, it is more likely a technical repair driven by short covering, with questionable sustainability.
On the options and volatility dimension, the dataset does not provide Rebar option implied volatility or open interest distribution, and the data is pending update. A cross-market volatility indicator for reference is ^VIX at 14.81, -4.08% on the day, in a state of low-level pullback, indicating that global risk asset volatility pricing is mild. This is generally unfavorable for the development of trending one-sided moves and more supportive of a range-bound judgment.
Fund flow summary: during this reporting period, Rebar showed “open interest build with downside,” with marginal flows skewed bearish; however, lacking classified positioning and options data, it is impossible to determine whether the shorts are hedgers or speculators, which directly affects the sustainability of the move — if hedgers dominate, downside space is constrained by the spot market; if speculators dominate, volatility may amplify. This judgment awaits supplementary data after the update.
IV. Cross-Asset Relative Value Analysis
This dataset does not provide price series for gold, silver, crude oil, copper and other instruments; therefore, the gold-silver ratio, oil-gold ratio, copper-gold ratio and their historical percentiles cannot be calculated, and the data is pending update. To avoid fabrication, this section only provides a qualitative discussion of the available macro variables and explicitly flags the unavailable items.
Available variable one: US Dollar Index at 100.2540 (+0.03%). The US Dollar Index is typically negatively correlated with USD-denominated industrial metals; trading above 100 exerts moderate pressure on copper, crude oil and others, which in turn indirectly affects domestic ferrous sentiment through the “commodity sector linkage” channel. However, it must be emphasized that Rebar pricing is dominated by domestic demand and the domestic property and infrastructure chains, so the direct explanatory power of the US dollar is limited.
Available variable two: US 10-year yield at 4.9980% (+1.03%). A yield approaching 5% is unfavorable for the valuations of growth-oriented and highly leveraged sectors, and also weighs on the financial attributes of commodities. If the yield effectively breaks above 5.00% and holds, the global commodity valuation center may shift further lower; if it stalls and retreats near 5.00%, the pressure eases.
Available variable three: VIX at 14.81 (-4.08%). A low-volatility environment typically favors carry and range strategies, while the win rate of trend strategies declines. For Rebar, this means that in the absence of domestic supply-demand data drivers, the price is more likely to remain fluctuating within the 20-day range of 3034.00-3203.00 rather than staging a one-sided breakout.
The absence of a relative-value framework is the main data gap in this report. In a normal research workflow, the copper-gold ratio is used to gauge global manufacturing activity, the oil-gold ratio to gauge inflation and growth expectations, and the gold-silver ratio to gauge the intensity of safe-haven demand; all three have cross-validation value for directional judgments on ferrous metals. As they are currently unavailable, this report will not cite any relative-value conclusions in the long-short scenario analysis in Section VII, relying only on price, moving averages, positioning and macro rates/US dollar.
V. Sentiment Monitoring
Sentiment sub-dimension: this dataset does not provide quantitative sentiment indicators for Rebar or the ferrous sector, and the data is pending update.
Headline bias dimension: the dataset explicitly marks domestic steel-related headlines over the past 48h (market=CN, asset=STEEL) as N/A, i.e., no usable news sample. Therefore, it is impossible to calculate the proportion of positive/negative/neutral reports, nor to identify sentiment focal points (such as property policy, production restrictions, environmental protection, exports, etc.). This report does not fabricate any media quotes or headlines.
The available indirect sentiment proxy is ^VIX at 14.81, down -4.08% on the day, at a low level and retreating, reflecting that global risk appetite has not deteriorated. However, VIX measures implied volatility of US equity options, and its mapping to domestic steel sentiment is weak; it can only serve as background reference for “no external panic.”
Comprehensive judgment: during this reporting period, the sentiment and public opinion dimensions for Rebar are missing, and no quantitative conclusion can be formed. In a sentiment vacuum, prices are more easily dominated by technical levels and capital flows, which is consistent with the earlier observation of “open interest build on the decline, moving averages turning bearish.” It is recommended that subsequent reports prioritize filling in domestic steel news sentiment and spot transaction sentiment data.
VI. Historical Statistical Patterns
This dataset does not provide historical seasonal series for Rebar (such as the distribution of gains/losses from late September to October over the past 10 years, seasonal inventory inflection points, seasonal operating rates, etc.), and the data is pending update. Therefore, it is impossible to conduct quantitative statistics on the “Golden September and Silver October” peak-season effect, pre-National Day restocking patterns, heating-season production restriction expectations, etc.
In the absence of historical statistical support, this report makes no directional judgment on seasonality and only flags two methodological considerations: first, Rebar seasonality is highly dependent on the actual strength of property and infrastructure in the given year, and the explanatory power of historical averages declines significantly in years when the demand structure shifts; second, late September is usually a window for position adjustment ahead of the National Day holiday, and seasonal changes in positioning and turnover may amplify short-term volatility, but this is a trading-behavior pattern rather than a price-direction pattern.
After the data is updated, it is recommended to supplement: the return distribution from September 21 to October 31 over the past 10 years, the maximum drawdown distribution within that window, and the average change in positioning over the 5 trading days before the holiday, in order to calibrate the holding period and stop-loss width for the strategies in Section VIII of this report.
VII. Long-Short Scenario Analysis
Bullish arguments (no fewer than 4):
1. The 20-day low L20 has already risen from 2994.00 on 09-14 to 3034.00 on 09-18, an increase of 40.00 points, indicating that the lower edge of the range is slowly moving up and the medium-term bottom structure remains intact.
2. The 09-18 close of 3096.00 is still about 16.67 points above S1 (3079.33); the first support has not been breached, and a breakdown has not yet been confirmed in the short term.
3. 09-15 and 09-16 posted consecutive gains (+0.4198%, +0.7412%), showing that there is absorbing buying below 3100 and that bulls have not fully exited.
4. Open interest has increased by about 2.19% cumulatively over 5 days, with both longs and shorts adding positions; if prices subsequently stabilize, short covering could steepen the rebound.
5. The external environment has not deteriorated: VIX has retreated to 14.81 (-4.08%), the US Dollar Index is only +0.03%, and there has been no systemic selloff in risk assets.
Bearish arguments (no fewer than 4):
1. The 09-18 close of 3096.00 fell below MA5 (3109.80) and MA20 (3126.80), and MA5 has already crossed below MA20, turning the moving-average structure bearish.
2. The 09-18 single-day decline of 1.0230% was the largest in the past 5 days, and volume of 884858 lots was the highest in the past 5 days; the high-volume decline confirms selling pressure.
3. The close of 3096.00 is below the pivot P (3105.67), leaving the pivot system weak.
4. The price is at about the 36.7% percentile of the 20-day range (3034.00-3203.00), biased toward the lower edge, and there is 107.00 points of space to H20 (3203.00), with heavy overhead trapped positions.
5. Macro pressure: the US 10-year yield at 4.9980% is approaching 5%, and the US Dollar Index at 100.2540 is holding above 100, leaving the external valuation environment tight.
Short- versus medium-term balance of forces: in the short term (1-5 trading days), bears have the advantage, based on the triple resonance of a bearish moving-average turn, a high-volume decline, and trading below the pivot; however, the bearish advantage is not overwhelming, because S1 (3079.33) has not been broken, L20 is moving up, and VIX is low. The medium-term (1-3 months) direction depends on domestic supply-demand data, which is currently pending update and cannot be judged. If the price finds support in the 3034.00-3079.33 area and reclaims MA20 (3126.80) on volume, the medium term is bullish; if it breaks below L20 (3034.00), the medium-term lower edge opens up.
VIII. Trading Strategies & Risk Management
Strategy one (range-biased short, main strategy): the mirror of a LONG direction is a short bias, but this report presents an executable two-way long-short framework. Short idea: if the price rebounds into the dense band from R1 (3122.33) to MA20 (3126.80) and stalls, enter short, with entry reference 3122, stop-loss reference 3136 (about 14 points above R1, corresponding to the 09-17 R1 near 3135.00), target reference 3079 (S1), horizon 1-5 trading days, position size recommended no more than 15% of total capital, conviction 6.
Strategy two (buy the support, counter-trend strategy): if the price pulls back to near S1 (3079.33) and shows a low-volume stabilization signal, enter long, with entry reference 3080, stop-loss reference 3060 (about 19 points below S1, and below the lower edge of the 09-18 pivot system), target reference 3122 (R1), horizon 1-5 trading days, position size recommended no more than 10% of total capital, conviction 5.
Strategy three (breakout follow, conditional strategy): if the price holds above MA20 (3126.80) on volume, switch to long, with entry reference 3127, stop-loss reference 3105 (below P), target reference 3203 (H20), horizon 5-10 trading days, position size recommended no more than 12% of total capital, conviction 6.
Risk management points: first, all strategies are confirmed on a closing-price basis to avoid intraday false breakouts; second, single-trade risk exposure should be controlled within 1.5% of total capital; third, if open interest continues to climb while prices fail to fall, be alert to short crowding and tighten stop-losses promptly; fourth, this report lacks domestic supply-demand and inventory data, and any strategy should be recalibrated after the data is updated.
IX. This Week's Data Calendar
The domestic financial calendar events available over the next 7 days are as follows (all are overseas PMI data; China steel-specific data are pending updates):
| Date & Time | Region | Importance | Event |
|---|
| 2026-09-23 03:15 | EUR | MEDIUM | France Manufacturing PMI Preliminary |
| 2026-09-23 03:15 | EUR | MEDIUM | France Services PMI Preliminary |
| 2026-09-23 03:30 | EUR | MEDIUM | Germany Manufacturing PMI Preliminary |
| 2026-09-23 03:30 | EUR | MEDIUM | Germany Services PMI Preliminary |
| 2026-09-23 04:30 | GBP | MEDIUM | UK Manufacturing PMI Preliminary |
| 2026-09-23 04:30 | GBP | MEDIUM | UK Services PMI Preliminary |
Note: The direct transmission of the above events to Rebar is limited; they influence commodity sentiment indirectly mainly through EUR and GBP exchange rates and global manufacturing activity expectations. Domestic steel inventory, output, apparent consumption, and high-frequency property and infrastructure data are not provided in this dataset and are pending updates; it is recommended that these be prioritized for inclusion in the next report.
This report is generated automatically from public quantitative and macro data for research and market tracking only. It does not constitute any investment advice.