A-Share Pre-Market Brief | Tuesday, 2026-10-06

Note: The A-share market is closed today (10/6) for the National Day holiday (markets closed 10/1–10/7; normal trading resumes from Thursday 10/8; source: SSE holiday schedule¹). This report is a pre-market watchlist for the first trading day after the holiday (October 8); the news window runs from the close of the previous trading day, 2026-09-30 15:00, to the time of writing (10-06 07:00), covering all incremental information over the National Day holiday. Close of the last pre-holiday trading day (9/30): SSE Composite 3842.19 (+0.31%), SZSE Component 12887.62 (−0.11%), ChiNext 3135.28 (−0.23%), STAR 50 −2.51%; combined turnover approx. RMB 1.44 trillion, a slight volume increase versus the prior session, still overall in a pre-holiday shrinking-volume regime; 52–56 limit-ups and 12–13 limit-downs; the mainlines were pharma/biotech (+2.73%, top net inflow of RMB 5.7 billion) and the property chain, while electronics/PCB/semiconductors led the declines (net outflow RMB 13.3 billion). Sources: 21jingji.com 9/30 close review², Eastmoney 9/30 close summary³, Lianban.net 9/30 recap⁴.

0. Today's one-line summary

The strongest catalyst over the National Day holiday (9/30 15:00 → 10/6 07:00) is innovative drugs going overseas: AstraZeneca's USD 2 billion strategic investment in Summit (~12% stake, USD 18.36/share), taking the lead on the global combination development of Akeso's ivonescimab (9/29), plus Hengrui Medicine licensing its GLP-1/GIP dual agonist HRS-1596 to Novo Nordisk (USD 300 million upfront, up to USD 2.6 billion total potential, 9/29–9/30), layered with the "15th Five-Year" Pharmaceutical Industry Development Plan from the MIIT and nine other departments (9/18) and the approaching ESMO congress (10/23–27, Madrid). Innovative drugs/CRO are the strongest pre-holiday mainline, and overseas mapping continued through the holiday (US vaccines/pharmacy retail led gains). The second-strongest branch is the property chain (the resident home-purchase loan interest subsidy landed on 10/1, with the six major banks offering "automatic enjoyment, no application needed") and the computing-hardware repair trade (10/5 Hong Kong PCB names surged — Kingboard Laminates closed up over 9%, Kinwong Electronic H up over 9% — against the divergence of A-share electronics net outflow of RMB 13.3 billion and PCB limit-downs on 9/30). Driver types: overseas industry + policy primarily, macro secondary (US September nonfarm payrolls only +29k, rate-hike expectations cooling). Pre-market state assessment: the overseas backdrop is neutral-to-positive (US Nasdaq at a record closing high on 10/5), but the US 10-year Treasury yield rising to 5.34% (highest since 2002) plus the Hang Seng Index's cumulative ~−2.3% over the holiday form a constraint; the 10/8 open is likely a structural gap-up — the validation focus is the auction strength of innovative drugs and the volume absorption of the PCB repair, while staying alert to pre-holiday mainline (pharma) gap-up profit-taking.

Dedicated macro risk note (UST 5.34%): This 10-year UST rally coexists with the weak "September nonfarm payrolls only +29k" print; the divergence shows this yield high is driven mainly not by growth expectations but by term premium/fiscal supply and sticky inflation — a more long-tailed drag on high-valuation growth stocks (Hengrui PE ~40x, Shengyi PE ~61x). Trigger-based watch conditions: if the 10Y UST keeps rising over the two nights of 10/6–10/7 and breaks above 5.40%, or the US Nasdaq turns from gains to losses, then the high-valuation branches in this list (high-PE innovative-drug names, PCB repair) should be down-weighted as a whole, keeping only the "statement-delivered" names (Hengrui Medicine, WuXi AppTec, COSCO Shipping Energy) as the watch focus.

1. News Overview

2. Strongest positive branches, descending

Key notes per branch: Innovative drugs — why positive: global pharma is pricing Chinese innovative assets with real money; BD has gone from isolated cases to an industry trend (2026H1 China BD = 58% of global share); short-term already rallied, mid-term relayed by ESMO and Q3 reports; earnings delivery path = upfront/milestone recognition + commercialization ramp; main risks: pre-holiday already fully climactic + 10Y UST suppressing valuations. Property — policy escalating in sequence (subsidy landed), but "policy lands = divergence" was already rehearsed on 9/30; short-term policy trading, mid-term watch October sales data. PCB repair — the 10/5 HK violent rebound and the A-share 9/30 crash form a strong divergence; the highest-elasticity direction on 10/8, but repair vs dead-cat bounce depends on auction volume; by board, the main board (Shengyi Tech/Kinwong) and ChiNext (Victory Giant/Zhongji Innolight) both have core names — note ±20% band names swing harder. Oil & gas/shipping — the most "solid" logic (COSCO Shipping Energy H1 net profit +143% already in the statements), but "old trend continuation" rather than a new catalyst — limited point in chasing. Solid-state battery/robotics — sentiment themes with no new news, generally poor high-tier fundamentals; watch only, don't act.

3. Individual stock positive-strength leaderboard (descending by stock)

4. Stock scoring model (total 100)

Items: source authority 0–15 / benefit directness 0–20 / earnings elasticity 0–15 / industry position & fundamentals 0–15 (revenue, net profit, gross margin, cash flow, debt ratio, moat, niche leadership) / expectation gap 0–10 / theme sustainability 0–10 / A-share trading attributes 0–10 (recognizability, liquidity, chip structure) / risk deduction 0 to −15. PE/PB valuations herein are static calibers as of the 9/30 close.

Full itemized breakdown (format: authority/directness/elasticity/position/gap/sustainability/trading/risk deduction): Hengrui Medicine 15/18/13/15/7/9/8/−3=82; WuXi AppTec 14/16/13/15/7/9/9/−2=81; Shengyi Tech 14/15/13/14/6/8/8/−6=72; COSCO Shipping Energy 12/16/13/13/6/7/7/−4=70; CanSino 13/9/9/12/6/9/10/−5=63; Trinomab 12/11/9/11/7/7/8/−8=57; Kinwong Electronic 13/9/8/11/6/7/8/−5=57; JOINN 12/9/7/13/6/7/9/−8=55; Vanke A 11/7/4/12/5/6/8/−7=46; Menovo 10/8/6/8/5/6/7/−5=45; China Film 11/8/4/11/4/4/6/−5=43; Shenzhen Property A 11/6/4/7/5/6/10/−7=42.

Rating vs total score: the rating (S/A+/A/B+/B/C) describes "the level of the news itself" (S = changes an industry trend or direct order elasticity; A = clear positive for a branch with multiple beneficiaries; B = long chain, needs validation; C = weak sentiment stimulus); the total score sums the eight model items. The two measure different dimensions and are allowed to diverge — e.g., CanSino's news level is A+ (many pharma-mainline sentiment events) but benefit directness is low, total 63 below "direct earnings elasticity" COSCO Shipping Energy (A/70); Trinomab (A/57) and Kinwong Electronic (A−/57) share the same score with different ratings, because the rating reads the news while the score reads the delivery path.

5. Top 10 stocks in detail

51 恒瑞医药600276total 82, priority deep-dive · Hengrui Medicine · Shanghai main

• Related news: 9/29 announcement — HRS-1596 (once-weekly oral GLP-1/GIP dual receptor agonist, Phase I already approved in China) overseas rights licensed to Novo Nordisk; USD 300 million upfront + up to USD 2.3 billion milestones + sales royalties; Greater China rights retained; closing expected in Q4 (jjckb⁷, Yicai³⁶).
• Positive logic: Direct. The ~RMB 2.1 billion upfront is expected to be recognized in Q4, equal to ~27% of full-2025 net profit attributable to parent — the only increment in the holiday window "written in black and white into the quarterly report"; it also endorses the pricing of the entire GLP-1 pipeline's internationalization — it affects revenue, not just sentiment.
• Industry branch stage: innovative-drug BD going-overseas main uptrend leg (sector fully exploded on 9/30), with tier structure and central names (WuXi AppTec, CanSino 20CM).
• Fundamentals check: 2026H1 revenue RMB 15.456 billion (−1.9%, first negative in years, dragged by generic centralized procurement), net profit attributable to parent RMB 4.465 billion (+0.3%), gross margin 86.3%, debt ratio only 8.2% — the most stable financial structure among A-share pharma; PE TTM ~40x (9/30 close caliber).
• Industry position: domestic innovative-drug #1, strongest BD internationalization delivery capability.
• Technical/sentiment: 9/30 +3.46%, not climactic amid the pharma limit-up wave — room to spare; main board ±10%.
• Final call: the hardest catalyst of the holiday. 10/8 watch points: gap-up magnitude and institutional absorption; a gap-up ≤3% with volume is a normal range; if the gap-up is above 5%, watch the pullback absorption — don't chase blindly.

52 药明康德603259total 81, priority deep-dive · WuXi AppTec · Shanghai main

• Related news: October broker top-pick clustering (21jingji.com 10/3³⁴); 9/18 "15th Five-Year" pharma plan positive for CXO; pre-ESMO industry-prosperity narrative strengthening.
• Positive logic: Direct. No single explosive point, but fourfold validation of orders, revenue, profit, and cash flow: 2026H1 revenue RMB 28.897 billion (+38.9%), net profit attributable to parent RMB 11.08 billion (+29.4%, first H1 above RMB 10 billion), ex-non-recurring +89.4%, backlog RMB 66.43 billion (+25.2%) at a record high, operating cash flow net ~RMB 9.7 billion; full-year guidance raised to RMB 58.5–60.5 billion.
• Industry branch stage: CXO prosperity recovery confirmation phase; the "central army" of the innovative-drug mainline.
• Fundamentals check: global CRDMO leader, PE TTM ~23x (9/30 close caliber), at the absolute industry head; red flags: receivables RMB 10.47 billion (+44%), geopolitical (US legislation) risk being fought via litigation.
• Industry position: A-share CXO absolute central army; the "picks-and-shovels" certainty beats Biotech.
• Technical/sentiment: 9/30 main-force net inflow over RMB 800 million, no limit-up (institutional tape); main board ±10%.
• Final call: the best risk-reward large-cap in the innovative-drug mainline, suitable as the core anchor for observing the sector; validation points: 10/8 gap-up magnitude and northbound/institutional absorption.

53 生益科技600183total 72, priority deep-dive (watch pullback absorption) · Shengyi Tech · Shanghai main

• Related news: 10/5 HK PCB concept surged, Kingboard Laminates closed up over 9%, intraday nearly 12% (STCN 10/5²¹, NBD²²), driven by the AI server "component shortage" and CCL volume-price upswing; US NVIDIA +2%+ on 10/5 with an intraday record.
• Positive logic: Direct. The core logic (high-end CCL mix up → margin up) has delivered for consecutive quarters, M7/M8 in volume supply; M9 reportedly the only domestic CCL vendor certified through NVIDIA's Rubin/GB300 platforms and in ramp-up (Wallstreetcn and other industry-media caliber, unconfirmed by company announcement); the HK surge is only a mapping catalyst.
• Industry branch stage: the starting point of the repair trade after computing hardware ebbed on 9/30 (electronics net outflow RMB 13.3 billion); a sector-wide high open on 10/8 is likely.
• Fundamentals check: 2026H1 revenue RMB 19.026 billion (+50.1%), net profit attributable to parent RMB 3.287 billion (+130.4%), gross margin 30.7% climbing for consecutive quarters, OCF per share RMB 1.23, ROE (H1) 18.2%; debt ratio 50.3% (+7pct, balance-sheet expansion for capacity). Red flags: PE TTM ~61x, PB ~17x (both 9/30 close calibers) — extremely high valuation; upstream glass-fiber cloth/copper foil/resin price hikes squeezing mid-low-end margins.
• Industry position: world's #2 rigid CCL, China #1; domestic high-end AI substrate leader (~30% AI-server CCL share, industry-media caliber).
• Technical/sentiment: 9/30 −2.27%, fell significantly less than the PCB high-tiers that limit-downed the same day, no panic tape; main board ±10%.
• Final call: the name with the "most real delivery path" in the PCB repair branch. A gap-up within 3% plus auction volume is healthy; too large a gap — then watch the volume-price coordination on a pullback to the 5-day line; if the PCB sector shows no auction volume, treat it as a dead-cat bounce rather than a repair.

54 中远海能600026total 70, watch closely · COSCO Shipping Energy · Shanghai main

• Related news: Brent holding USD 101–103/barrel (highest region since May), long-tailed Middle East Hormuz + Bab el-Mandeb dual-strait disruption (early-September US strike on an Iranian tanker, Houthi attack on Saudi facilities); 10/5 US oil & gas sector led gains (Schlumberger +3%).
• Positive logic: Direct and already delivered. VLCC-TCE at a record high within 2026 (BOCI, Dongxing Securities and other broker research calibers; Middle East–China route rates at one point reached historical extremes of hundreds of thousands of dollars per day — large variance across institutions' estimates); high rates are already in the statements.
• Industry branch stage: high oil/freight persistence phase; Q4 traditional peak season + risk premium as the upside option.
• Fundamentals check: 2026H1 revenue RMB 15.146 billion (+30.0%), net profit attributable to parent RMB 4.545 billion (+143.2%), ex-non-recurring RMB 4.449 billion, gross margin 41.4% (~+18pct y/y), OCF per share RMB 1.27; PE TTM ~17x, PB ~2.3x (9/30 close caliber). Red flag: strong cyclical — if rates/risk premium fade, earnings retreat fast.
• Industry position: one of the world's largest tanker fleets; A-share oil-shipping absolute leader.
• Technical/sentiment: 9/30 +1.43%, steady climb, not overheated; main board ±10%.
• Final call: within the holiday window, the most "solid" fundamentals among branch leaders — but old-trend continuation rather than a new catalyst; better suited to watching pullback absorption and tracking the hold logic than emotional chasing on 10/8; validation points: whether Brent holds 100 and VLCC spot TCE.

55 康希诺688185total 63, watch closely (sector sentiment anchor) · CanSino Biologics · STAR board

• Related news: 9/29–9/30 innovative-drug/vaccine sector fully exploded; 9/30 CanSino 20CM limit-up, late-session net inflow RMB 243 million, #1 across both markets; US vaccine surge during the holiday (NVAX +20%, Moderna +7%) and the New York measles emergency provide sentiment sidelights (note: measles/mRNA are weakly related to CanSino's core business — sector sentiment only).
• Positive logic: Indirect sentiment mainly + weak earnings support. No new company announcement during the holiday; sector β; sentiment anchor of the dual mainline (innovative drugs + vaccines).
• Industry branch stage: the high-recognition 20CM flag of the pharma main uptrend.
• Fundamentals check: 2026H1 revenue RMB 550 million (+43.8%), net profit −RMB 720k (loss narrowed 94.7%, near breakeven) but ex-non-recurring −RMB 50.05 million still negative; gross margin 76.9%; debt ratio 27.1% declining year by year; cash RMB 1.34 billion; OCF per share −0.43. Products: MCV4 曼海欣 (domestic only quadrivalent meningococcal conjugate vaccine, expanded to age 6 in Feb 2026, Argentina registration in June), 13-valent pneumococcal 优佩欣 (approved 2025/6, admitted to 20+ provinces, 2026 is the first full commercial year).
• Industry position: meningococcal conjugate vaccine niche leader; one of the A-share innovative-vaccine central names.
• Technical/sentiment: STAR board ±20%; after the 20CM first board, a 10/8 high open is likely — chasing risks profit-taking.
• Final call: the "wind vane" of pharma sentiment; its auction strength decides the whole branch's opening direction; good as an observation anchor, not as a heavy-position basis.

56 泰诺麦博-U688806total 57, watch only (high risk) · Trinomab-U · STAR board

• Related news: core Class 1 innovative drug RSV mAb (新多妥/芮特韦拜单抗) approved for marketing by NMPA on 9/3; 9/30 20CM limit-up; "H1 revenue up over 6x y/y" widely cited (CNR³⁷).
• Positive logic: Small-base direct elasticity + event sentiment mix. The RSV product was approved only in September and has not yet contributed revenue; H1 revenue came from the tetanus mAb 斯泰度塔; the 9/30 limit-up looks more driven by "approval event + small float (~RMB 1.6 billion)".
• Industry branch stage: the second-highest-recognition elasticity name within the innovative-drug branch.
• Fundamentals check: 2026H1 revenue RMB 76.24 million (+623.4%, "up over 6x" is true), net profit attributable to parent −RMB 199 million (loss narrowed 31.3%); gross margin 66.6%; debt ratio ~95% (derived from total assets RMB 1.042 billion and net assets attributable to parent ~RMB 51 million, Time Weekly caliber); note the company's July STAR IPO net proceeds ~RMB 890 million have been received — read "low net assets" together with the received proceeds; -U = not yet profitable.
• Industry position: domestic first-tier RSV long-acting mAb (one of two companies), scarce but small-cap.
• Technical/sentiment: STAR board ±20%, 9/30 20CM limit-up, extremely hot sentiment.
• Final call: theme purity and statement risk are equally extreme — the ~95% debt ratio + RMB 51 million net assets is a hard flaw (though cushioned by the received IPO proceeds); watch only; if tracked at all, limit to the sentiment-trading layer.

57 景旺电子603228total 57, watch only · Kinwong Electronic · Shanghai main

• Related news: 10/5 HK H-share intraday up over 10% again (A/H premium ~30% forms the mapping basis), driven by the AI server component shortage; 9/30 H-share +10.26% to HKD 77.05.
• Positive logic: Indirect sentiment mainly (A+H mapping), weak earnings directness — the HK spike is sector β + recent-H-listing elasticity; the company has not yet delivered AI PCB profit elasticity in H1.
• Industry branch stage: sentiment-mapping name of the PCB repair branch.
• Fundamentals check: 2026H1 revenue RMB 8.611 billion (+21.4%) but net profit attributable to parent RMB 602 million (−7.4%) — rising revenue, no profit growth; gross margin 20.2% (−1.2pct); debt ratio 51.2% (+7.4pct); fundraising projects delayed twice; controlling shareholder cashed out nearly RMB 1 billion pre-listing; PE TTM ~92x (9/30 close caliber).
• Industry position: domestic full-category PCB platform leader (rigid + HDI + flex + metal-base), global top-10 by PCB revenue.
• Technical/sentiment: A-share 9/30 +0.68% resilient; 10/8 likely gaps up on H-share mapping; main board ±10%.
• Final call: if it spikes toward the limit-up on 10/8 it is sentiment rather than earnings; Shengyi Tech is the harder name in the same logic; the mapping trade can watch the A/H premium convergence rhythm — not a fundamentals basis.

58 昭衍新药JOINN / 603127total 55, watch only · Shanghai main

• Related news: 9/30 extra-large single net inflow RMB 552 million, #1 across both markets, CRO-branch sentiment core; "monkey price up + order recovery" expectation brewing.
• Positive logic: Indirect sentiment. The "order recovery" being traded currently exists only as broker paraphrase, with no concrete numbers; and the 9/30 limit-up rests on profits "unrelated to the core business".
• Industry branch stage: CXO-branch elasticity name.
• Fundamentals check: 2026H1 revenue RMB 704 million (+5.3%), net profit attributable to parent RMB 748 million (+1126.8%) — of which biological-asset fair-value change gain RMB 735 million (98%+); core business still loss-making (admitted at the company's earnings briefing); debt ratio 17.1%; operating cash flow acceptable. Red flags: profit entirely from lab-monkey revaluation (two-way volatility, unsustainable); 2025 net profit RMB 298 million corresponds to PE ~124x (9/30 close caliber).
• Industry position: drug non-clinical safety evaluation (GLP toxicology) leader, #1 lab-monkey reserves industry-wide.
• Technical/sentiment: 9/30 10CM limit-up, high recognizability; main board ±10%.
• Final call: the archetype of "pretty statements, core business asleep" — fine as a CRO sentiment wind vane, not as a fundamentals name; for CXO earnings delivery, WuXi AppTec first.

59 万科A000002total 46, watch only (policy β observation) · China Vanke A · Shenzhen main

• Related news: Resident home-purchase loan interest subsidy effective 10/1 (six major banks, no application needed); 9/30 intraday went straight from a limit-down open to green, late-session net inflow RMB 208 million.
• Positive logic: Indirect sentiment. The subsidy helps industry β and purchase expectations; improvement to Vanke's own statements is limited; company level remains a "payment guarantee" logic.
• Industry branch stage: property policy-trading phase; "policy lands = divergence" was already rehearsed on 9/30.
• Fundamentals check: 2026H1 revenue RMB 70.169 billion (−33.4%), net profit attributable to parent −RMB 14.951 billion (loss widened 25%), gross margin 7.9%; full-2025 loss RMB 88.56 billion; after the 2025/11 debt crisis, all public bonds completed one extension (per Caijing, Caixin et al.); Shenzhen Metro cumulative shareholder loans RMB 33.52 billion (as of end-2026/3), plus RMB 4.5 billion of new borrowing line added in July (a line, not a drawdown); ~RMB 10.5 billion of domestic bond principal and interest due within the year (media caliber); PB ~0.50x.
• Industry position: property development leader (Onewo), now a deep risk-resolution entity.
• Technical/sentiment: 9/30 deep V, extreme long-short divergence; main board ±10%.
• Final call: only suited as the observation weight for "whether property policy is endorsed by money"; no profit-logic support; a debt-extension entity whose fundamentals risk has not cleared.

510 深物业A000011total 42, watch only (sentiment benchmark) · Shenzhen Property A · Shenzhen main

• Related news: 9/30 staged an "earth-to-sky board" for 3 consecutive limit-ups, property short-term sentiment benchmark; Shenzhen SOE (Shenzhen Investment Holdings) background.
• Positive logic: Pure concept mapping/sentiment. Property policy β + Shenzhen SOE shell-resource imagination (Shenzhen Investment Holdings' same-industry-competition commitment plan for 深深房/深物业 (Shenfang/Shenzhen Property) expected to be implemented before 2026/11/9, asset-integration expectation), diverging from its own swung-to-loss fundamentals.
• Industry branch stage: property branch short-term high-tier.
• Fundamentals check: 2026H1 revenue RMB 1.205 billion (+10.8%), net profit attributable to parent −RMB 68.18 million (swung to loss); gross margin 13.1% (nearly halved); debt ratio 79.2%; ex-non-recurring −RMB 72.6 million; PB ~2.2x.
• Industry position: Shenzhen SOE property + property-management platform, regional small-cap (~RMB 7.3 billion).
• Technical/sentiment: earth-to-sky board extremely recognizable; 10/8 large gap-up likely, relay risky; main board ±10%.
• Final call: a trading-attribute name, watch only; its opening performance gauges property-branch sentiment — if it gaps up and fades on 10/8, the entire property chain should be avoided short-term.

6. Pass list

7. Intra-branch ranking

Innovative drugs / BD going overseas + CRO (S-tier branch)

Hardest name: Hengrui Medicine; highest trading recognition: CanSino Biologics; followers: 蔚蓝生物 (Vland Biotech, 603739), 康辰药业 (Kangchen Pharma, 603590), 奥赛康 (Aosaikang, 002755), 义翘神州 (Sino Biological, 301047), 中红医疗 (Zhonghong Medical, 300981); Pass: 南华生物 (Nanhua Biotech, 000504, high position), 近岸蛋白 (Novoprotein, 688137, abnormal move).

Property chain (A-tier branch)

The hardest logic is actually not in any stock but in the policy itself; the branch has no "hardest fundamentals" name — an inherent attribute of policy-trading branches. Post-gap-up consistency and absorption matter more than stock picking.

Computing hardware / PCB repair (A− tier branch)

Hardest name: Shengyi Tech; highest recognition: Victory Giant (±20% elasticity); followers: second-tier CCL names; avoid: 澳弘电子 (Aohong Electronics, 605058), 协和电子 (Xiehe Electronics, 605258) (ebb high-tiers).

Oil & gas / oil tankers (B+ tier branch)

8. Opening validation signals for tomorrow (10/8, first day after the holiday)

• Auction signals: innovative drugs (Hengrui Medicine/CanSino) gap-up magnitude and sealed orders — if CanSino gaps up over 8% with thick sealed orders, the pharma mainline continues; if Hengrui opens flat or down, institutions doubt BD sustainability, and the sector's gap-up profit-taking risk is high. Shengyi Tech gap-up within 3% + broad PCB auction green = valid repair; gap-up above 5% = overstretched.
• Sector signals: ① Pharma: CanSino/Trinomab quick limit-ups + Hengrui/WuXi up on volume = mainline confirmed; only back-row (Menovo et al.) spiking while the central names sit still = ebb warning. ② PCB: 3+ names (Shengyi/WUS/Victory Giant/Kinwong) attacking fast with Zhongji Innolight stabilizing = repair spreading; only Kinwong spiking alone = one-day HK mapping. ③ Property: Shenzhen Property A's absorption after the gap-up decides the policy-branch sentiment.
• Stock signals: top-5 absorption (auction volume/sealed-order ratio), quick limit-ups that hold, large single net inflows, and whether a stronger name steals the slot (e.g., Trinomab stealing CanSino's slot means money abandoning high for low — divergence widening).
• Risk signals: ① Index level: a quick dump after the high open to fill the gap; if HK weakens further on 10/6–10/7 (the 10/6 early-hours HSI futures night session +0.9% is only a temporary repair), the A-share open is pressured. ② Standalone risk signals: pharma "single limit-up + back row not following"; PCB "gap-up fade + shrinking volume"; COSCO Shipping Energy opening down with oil (10/5 WTI −2%); Shenzhen Property A nuking after a one-way sealed limit-up open, crashing the property chain. ③ Yesterday's strong themes ebbing: robotics (Xiangyang Bearing) and solid-state battery (Chuanyi/Times Wanheng) high-tiers opening down together = pre-holiday huddled money leaving, overall short-term sentiment cooling. ④ Macro: US market direction over the two nights of 10/6–10/7 (under the 5.34% UST yield, a Nasdaq turn lower weakens the growth-repair logic, triggering the §0 down-weight condition) + the 10/7 Nobel Chemistry Prize (may stimulate concepts again).

9. Final recommendations

The following is only a pre-market observation priority ranking, not investment advice; all "validation points" are observation calibers, not operation instructions.

① Today's 5 most noteworthy stocks

② Today's 3 strongest branches

③ Directions not to chase today: ① property-chain high-tiers (Shenzhen Property A, Times Wanheng — both sentiment trading) — "policy lands = divergence" was already rehearsed on 9/30, and leader fundamentals (Vanke extension, Shenzhen Property swung to loss) have not reversed; ② solid-state battery concept (Chuanyi Tech 002866, Shanghai Xiba 603200, Times Wanheng 600241) — no new catalyst, poor high-tier fundamentals, pure concept mapping; ③ robotics high-tier (Xiangyang Bearing 000678) — zero holiday catalyst, no order announcement, Xuelong Group already demoed the ebb; ④ cinema chains (China Film 600977 et al.) — season ends 10/7, profit-taking near the end, China Film only co-invests with small elasticity; ⑤ Nobel concepts (optogenetics/brain science) — pure event mapping, no direct A-share beneficiary, likely a one-day pop.

④ Final one-line judgment: the holiday news window — "industry catalysts in pharma, repair trades in PCB, the suppressor in USTs" — 10/8 is a structural gap-up, not a broad one; on direction, take Hengrui/WuXi-style "statement-delivered" names as the mainline and Shengyi-style repair elasticity as swing-trade observations; property/solid-state/robotics high-tiers watch only; and treat the US 10-year at 5.34% plus the two US sessions on 10/6–10/7 as the global risk switch — if the 10Y keeps rising through 5.40% or the Nasdaq turns down (§0 trigger conditions), down-weight the high-valuation branches as a whole.

⚠️ Risk warning: this list is only a pre-market information review and observation, not investment advice. A-share volatility risk is extremely high; automatically generated content may carry information timeliness gaps or industry-chain mapping errors, and must not be used directly as a trading basis. Financial data herein are mainly from the 2026 interim reports (reporting period 2026-06-30) and public reporting; PE/PB valuations are static calibers as of the 9/30 close; certain names (WUS, Victory Giant, PetroChina, COSL) had their latest financial-statement details not itemized, as noted in the text; certain industry-position statements (M9 NVIDIA certification, AI CCL market share, VLCC rate extremes) are industry-media/broker-research calibers unconfirmed by company announcements, marked "to be verified"; price-limit bands differ for the STAR board (±20%) and ChiNext (±20%) and BSE (±30%) versus the main board (±10%) — mind the risk-scale differences.