VLCC 超级周期深度分析 | VLCC Super Cycle Deep Dive

Multi-Model AI Valuation Research — 2026

📊 研究报告 | Research Reports

🥇 Gold Miners — US-Listed Western Majors vs China Majors (NEW — Sep 3, 2026)

Compares the top gold miners — Newmont / Agnico / Kinross (US-listed Western) vs Zijin / Shandong Gold / Zhaojin (China) — and stress-tests a common thesis (“US = higher cost but pure-gold focus; China = low cost but non-lucrative diversification”). Verdict: half right, half inverted. On cost it’s true vs Newmont ($1,609) / Barrick ($1,637), but the lowest-cost major is Western — Agnico at $1,339 (below every Chinese name except Shandong). On focus it’s backwards for the flagship: Zijin is a copper-gold major (gold only ~33% of revenue); copper is its most lucrative, fastest-growing engine, while Newmont is the >85%-gold pure-play. The “non-lucrative” tag fits SOE smelting/refining, not Zijin’s copper. China screens cheaper (Zijin fwd P/E ~9.3, div ~3.0%, ROE ~36%) but carries a China/SOE-governance + geopolitical discount. At gold ~$4,474/oz, scale beats cost — Newmont’s gold gross profit (~$16.9B) is ~2× Shandong’s despite a $360 higher AISC. Right like-for-like gold pair = Newmont vs Shandong Gold. Reproducible cost/margin model + chart. Analysis, not investment advice.

📄 Gold Miners (EN) → 黄金矿业股（中文）→ data + code →

🚢🏭 VLCC Supply — Does the 2027/2028 Newbuild Wave Break the Cycle? (NEW — Aug 22, 2026)

Turns the scary GROSS delivery numbers (~68 VLCCs in 2027, ~125 in 2028 ≈ 21% of the ~900-ship fleet) into NET fleet growth after scrapping + shadow-fleet exit. Verdict: materially — but as a 2028 rate-normaliser, not a 2027 cycle-killer. 2028 is +13% gross but only +4.9% to +9.9% net (depending on scrapping), and 2027 stays tight (+2.5–5.8% net, absorbed by SPR restocking + the shadow-fleet exit). The offset is a record aging pool — ~130 VLCCs already >20yo (~20%), doubling to ~300 by 2029–30, forced out by IMO-2030/EEXI/CII. The whole answer reduces to one variable: does scrapping accelerate? High-scrap → healthy renewal (fleet shrinks 2029–30); Low-scrap → real oversupply. Confirms the cycle’s expiry window (late-2027/2028) and the repo’s exit discipline (CRule 8). Reproducible net-growth model + chart + 3 scrap scenarios. Analysis, not investment advice.

📄 VLCC Supply (EN) → VLCC 供给（中文）→ data + code →

🏡 Bellevue Buy vs Rent — Housing Investment or Lifestyle Consumption? (NEW — Aug 16, 2026)

A reproducible seven-year terminal-wealth model for a $1.8M Bellevue home with $900K down, compared with rents from $4,300–$8,000/month. Key matched-home result: if the same house rents for $5,200/month, buying needs about 4.31% nominal annual appreciation to match renting under the stated assumptions (4% after-tax alternative return, 1% buy cost, 7% sale cost). At 3% appreciation, ownership’s lifestyle premium is about $1,960/month—the price of ownership, customization, and tenure certainty after renting already supplies the same home. Also covers Bellevue’s high-end price-to-rent ratios, affordability, tech-employment concentration, and European evidence on nominal versus real house-price growth. Includes the Two-Step Research Protocol and a dependency-free reproducible model. Analysis, not tax or investment advice.

📄 Buy vs Rent (EN) → Bellevue 买房还是租房（中文）→ data + code →

📈 Market Gauge — How High Is the S&P 500, and How Good Is the Quality? (NEW — Aug 4, 2026)

A four-axis “market-level” scorecard — breadth · valuation · positioning · quality. Verdict: expensive and stretched, but genuinely high-quality — “priced for perfection,” not a junk bubble. Valuation is ~99th-percentile extreme — Shiller CAPE 41.3 = the 98.9th percentile since 1881 (only the Dec-1999 peak of 44.2 was higher), forward P/E ~21, Buffett indicator ~225% of GDP. Breadth is two-faced: participation is healthy (~69% of the S&P above its 200-day MA) but leadership concentration is the narrowest in 20 years (equal-/cap-weight RSP/SPY at the 3rd percentile); the broadening is real on a 3-month view (+1.8%) but fragile (last week re-narrowed −3.5% on mega-cap earnings; only 3/11 sectors beat the index over 1mo). Positioning is stretched (CTAs net long with $100B+ mechanical downside; VIX 48th pct — no cushion). But quality is the anchor — record earnings + record margins make this the AI-bubble report’s “1998→late-1999, loaded but unlit” market. Thin margin of safety + high quality → stay invested but hedged/diversified; watch positioning/credit, not the P/E, for the turn. Reproducible (CAPE percentile from Shiller data, breadth/VIX computed). Analysis, not investment advice.

📄 Market Gauge (EN) → 市场标尺（中文）→ data + code →

💼 Portfolio Strategy — The 30/30/40 Gold + Index + Alpha Barbell (NEW — Aug 4, 2026)

Assesses a 30% gold / 30% index / 40% concentrated alpha (max 2 domains) barbell, plus dividend/blue-chip (SCHD) and “ballast” (XLP vs low-vol vs Treasuries) sleeves. Verdict: the skeleton is sound; the risk is entirely in the 40%. The 60% core is the free lunch — gold and the S&P both ~11% CAGR but corr 0.08, so 50/50 keeps the return and halves drawdown (−51% → −25%, Sharpe 0.75→~0.97) if you rebalance mechanically. 30% gold is a macro/regime bet (~0 long-run real drift — insurance, not compounding). On the dividend idea: SCHD ≈ a quality-value S&P (corr 0.85), a core tilt not a diversifier; and a corrected-data finding — XLP is NOT a near-zero ballast (corr 0.65, ~54% down-capture), it’s lower-beta equity — the only true diversifiers (corr ~0) are Treasuries + the gold you already own. The 40% alpha must clear a ~10% hurdle (or just index it), stay uncorrelated across both domains, and run with hard exit discipline (a 20% domain −50% = −10% to the whole book). Reproducible backtest + committed CSVs. Analysis, not investment advice.

📄 Portfolio Strategy (EN) → 组合策略（中文）→ data + code →

🚢📅 VLCC Seasonality — Is the Q4 Stock Bump the Calendar, or the Year’s Rate Strength? (NEW — Aug 2, 2026)

Tests whether the seasonal Q4 VLCC TCE peak reliably lifts DHT/FRO — or whether Q4 stock performance is really about that year’s rate level. Verdict: it’s the level, not the calendar. Over ~16 years Q4 is a coin-flip (DHT 50% / FRO 44% of Q4s positive; FRO Q4 median −4.3%); the genuinely strong seasonal quarter is Q1 (DHT +12.5% avg, 75% positive) and November is the worst month (DHT −5.6%). But cross-year Q4 return correlates R ≈ 0.60 (DHT) / 0.66 (FRO) with the Q4 TCE level — high-rate Q4s rip (2019/2014/2022), low-rate Q4s fall (2021/2017). The reason is CRule 1: the stock leads the rate 1–3 months, so the predictable winter bump is pre-priced (→ Q1 confirmation + Nov sell-the-news), and only a surprise in the level pays. 2026 already fired its move in Q1 (DHT +53%, FRO +65%) front-running the Mar/Jun >$400k spikes — so “buy for Q4 seasonality” is not an edge. Reproducible script + committed CSVs. Analysis, not investment advice.

📄 VLCC Seasonality (EN) → VLCC 季节性（中文）→ data + code →

🧭 Tail-Hedge Cheat-Sheet — Cross-Asset Decision Table (NEW — Jul 20, 2026)

The one-page hub for the whole tail_hedge/ study: a 7-asset decision table (S&P, XLF, JPM, AXP, XLK, DHT, FRO) with each asset’s hold verdict, break-even VRP, live paid VRP, and hedge verdict — plus the single decision rule (hedge only if paid VRP < break-even VRP and you have a regime reason) and the “deep AND frequent relative to drift” principle. Bottom line: 5 of 7 → hold, don’t hedge; only tactical XLK and top-of-cycle DHT clear the bar.

📄 Cheat-Sheet (EN) → 速查表（中文）→

🏦 Sector Convexity Hedging — Financials (XLF) & Technology (XLK) (Jul 20, 2026)

Extends the tail-hedge framework to long-term-holdable fat-tailed sectors to test whether they beat VLCC as “hold + tail-hedge.” Verdict — a useful split: better to HOLD (XLF/XLK CAGR +5.7%/+9.2% vs VLCC negative), but mostly NOT better to systematically HEDGE. The break-even VRP spectrum: S&P ≈0%, Financials ≈0%, Technology ≈27%, DHT-VLCC ≈67%. The positive drift that makes a sector holdable is what makes hedging it bleed — so only Technology (recurring dot-com/2008/2022 crashes) has a positive break-even, and live 1-yr 20%-OTM XLK puts price a paid VRP of ≈24% < 27% → a tactical XLK hedge is marginally defensible now, a direct expression of the AI-bubble §11 thesis (but XLK’s edge collapses to 0% post-2010 — crash-regime-dependent). Unifying rule: tail-hedging pays only where crashes are deep AND frequent relative to drift. Analysis, not investment advice.

📄 Sector Hedging (EN) → 板块凸性对冲（中文）→ data + code →

🚢 VLCC Convexity Hedging — DHT/FRO Backtest & Win-Rate-vs-VRP Framework (Jul 20, 2026)

Applies the tail-hedging logic to the assets a VLCC holder actually owns (DHT/FRO) and answers: how much can you overpay for tail insurance? Verdict: VLCC’s catastrophic tail (DHT vol 48%/maxDD −97%, FRO 61%/−98% vs S&P 17%/−57%) makes tail-hedging worth a MUCH higher vol-risk-premium — DHT’s CAGR break-even VRP ≈ 67% vs the S&P’s ≈ 0%. But it’s not reliable: FRO’s break-even VRP ≈ 0% (grind + high premium + 2011–12 restructuring defeat it), and DHT’s entire hedge value came from ~one year (2011). Introduces a win-rate-vs-VRP framework: raw win-rate is 4–18% (useless for a convex bet) — decide by the CAGR break-even VRP vs your paid VRP, conditioned on entry vol (buying puts mid-crash at peak IV is a near-guaranteed loss). Reproducible backtest + committed CSV data. Analysis, not investment advice.

📄 VLCC Convexity Hedging (EN) → VLCC 凸性对冲（中文）→ data + code →

🛡️ Tail-Hedging & Convexity — A 50-Year Backtest (Jul 20, 2026)

Empirical test of the Taleb/Spitznagel tail-hedging / convexity thesis on 50 years of real S&P 500 total return (1974–2024, Shiller data) — does buying puts actually raise the geometric return (几何收益率) and compensate for Kelly’s fat-tail fragility? Verdict: real, but NOT a free lunch — price is destiny. Convexity flipped Buy&Hold’s fat left tail (skew −0.90 → +0.02) and clipped the worst month −19.4% → −6.3%; cheap puts raised CAGR 7.80% → 8.57% with lower risk; a 1-year LEAPS put halved drawdown (−40% → −21%) for ~0.4%/yr — validating long-dated protection. But at realistic vol-risk-premium the hedge costs 0.4–1.4%/yr CAGR, and bought too dear it deepens drawdown (the AQR vs Universa debate, quantified). At equal drawdown, convexity beat a cash barbell only +0.3%/yr. Full reproducible backtest + committed CSV data tables. Analysis, not investment advice.

📄 Tail-Hedging Backtest (EN) → 尾部对冲回测（中文）→ data + code →

🛢️ Saudi Oil Price War → VLCC — The Two Prior Analogs & DHT/FRO Record (Jul 6, 2026)

Saudi announced an oil price war — the 3rd this century after 2014–16 (vs US shale) and 2020 (vs Russia). Verdict: bullish for VLCC RATES, 2 for 2 (a price war = more barrels + floating storage, not about the oil price level; rates hit >$100k/day in 2015 and ~$200k–$279k/day in 2020; DHT’s Q2-2020 was its best quarter ever). But the STOCK record is a cyclical trap: FRO +21% (2015) → −46% (2016); in 2020 DHT −20.5% / FRO −25.9% full-year despite record earnings — the spike is a sell-into-strength event (CRule 5/8). What’s different in 2026: the near-zero orderbook to late-2028 removes the newbuild wave that killed 2016 and the storage-unwind of 2020 → a boost now could be more durable — a rare bullish “this time is different,” unless the war is a demand-collapse (recession) event. Live: FRO ~$37.02 (Jul 6), DHT ~$17.18 (Jul 2). Analysis, not investment advice.

📄 Saudi Price War → VLCC (EN) → 沙特价格战 → VLCC（中文）→

💰 VLCC Cycle Position — Are DHT/FRO Cheap or Expensive? (Jun 26, 2026)

Applies the Average × Duration model to live prices (DHT $17.44, FRO $35.12, Jun 26 — both down sharply this week as the spike premium unwinds). Verdict: neither is expensive — both CHEAP-to-FAIR, and the market is pricing them on the sustained ~$100k TCE, not the $420k Hormuz spike (which already collapsed). Spot TD3C ~$100k now (−76% from the ~$420k March peak) vs 2025’s ~$50–70k; structural elevation sustained ~9–12 months, supply-backed through 2027. PE just 5–6× @ $100k sustained (8–9× @ $70k) = mid-cycle, not peak. High-conviction supply case (§8): at a sustained $150k the stocks roughly double (FRO ~$66 / DHT ~$30, +70–88%); at $200k, FRO ~$91 / DHT ~$41 (+130–160%) — but only if it’s a sustained average, and PE 2.5–3.5× there is the sell tell. 12M base targets: FRO $38 (+9%) / DHT $17.5; plus 12–15% dividend yield. Sell-algo: 🛡️ “do not sell” (spike-unwind ≠ cycle turn). Includes a Fact-Check section (2008 peak corrected to ~$230k; 2026 Hormuz ~$420k; BDTI-vs-TD3C proxy resolved). Analysis, not investment advice.

📄 VLCC Cycle Position (EN) → VLCC 周期定位（中文）→

📉 TCE vs VLCC Stock Price — Why “Watching the Rate Tape” Fails (Jun 26, 2026)

The “Average × Duration” thesis, proven with data (FRO + DHT). Settles the debate that “watching spot TCE to trade VLCC stock is bad.” Verdict: the stock prices the AVERAGE TCE sustained over a DURATION, not spot spikes. Six evidence points: (1) stock-vs-rate R² rises with the averaging window (FRO 0.12→0.37, DHT 0.20→0.50 as spot→52-wk avg, real BDTI 2020–24); (2) amplitude compression — TCE peaks 5–10.6× baseline but stock peaks only ~1–3× (the direct answer to “how big is TCE peak vs stock peak”); (3) real episode contrast — 2020 short spike → FRO +11% vs 2022–24 sustained → FRO +307%; (4) simulation: same $200k peak, 2-wk spike ×1.0 vs 2-yr sustained ×1.82; (5) control — tripling the peak ($120k→$350k) moves stock only +10%; (6) sustained-avg signal forward return median +64% (80% win) vs spot +10% (63%). Ties to the Modeling Stash momentum+rate-confirmation algo. Includes the repo’s Two-Step Research Protocol (draft + peer review). Analysis, not investment advice.

📄 TCE vs Stock Price (EN) → TCE 与 VLCC 股价（中文）→

🫧 Dot-Com Bubble (1995–2000) vs AI Bubble — Where Are We Now? (NEW — Jun 23, 2026)

Following the repo’s Two-Step Research Protocol + CRule 1 two-cycle backtrack (dot-com = reference cycle, AI = current cycle). Verdict: Late-Build / pre-Mania, ~1998–early-1999 analog — past the inflection, mid-capex-mania, valuations stretched but not yet at the 1999/2000 blow-off. Decisive difference vs 1999: revenue still accelerating into the capex (in 1999/2001 revenue rolled over first) → keeps us pre-peak. Decisive risk: ~$500B/yr capex-vs-revenue gap (~$700B capex vs ~$150–200B AI revenue), $230B+ new 2026 debt, FCF collapsing (Amazon −95%). Bear analog = telecom 2000–02; bull analog = Cisco/Intel 1998. Includes 5-phase mapping, side-by-side bubble metrics, disanalogies, and 4 signals that would flip us to “1999/2000”. Education/analysis, not investment advice.

📄 Dot-Com vs AI Bubble (EN) → 互联网泡沫 vs AI 泡沫（中文）→

🤖 What Did AI Actually Revolutionize? — Capex vs Revenue Reality Check (Jun 1, 2026)

Pivot from VLCC to AI industry analysis. Honest answer to “with $1.2T of cumulative Mag7 capex flowing into Nvidia + memory + components, what has AI actually revolutionized?” Validated: financial geometry user described is correct — picks-and-shovels (Nvidia $100B+ net, memory all $1T+ market cap with $300B+ revenue, components $150B+) captured most realized profit. But pushed back: “AI usage is limited” understates 900M weekly ChatGPT users + 20M Copilot seats + 70% F100 using Anthropic. Honest “revolutionized” list is short: software engineering (50%+ productivity gains, statistically validated); advertising (Meta $60B+ Advantage+) and customer service are improved not transformed; everything else is augmentation (only 6% of enterprises see 5%+ EBIT impact per McKinsey 2025). Math gap: $700B 2026 capex vs $150-200B AI revenue = $500B/yr gap, funded by debt. Anthropic 30x growth in 17 months is the single strongest counter to bubble thesis. Personal handicap: 55% bull / 35% middle / 10% bear on capex justification by 2028.

📄 AI Industry Analysis (EN) → AI 行业分析（中文）→

Key data points (mid 2026):

🛑 Global SPR Refill — Math Correction (May 29, 2026)

User pushed back: SPR math was US-narrow. 400M release breakdown: US 172M (43%), Japan 79.8M (20%), Korea 22.5M, Germany 19.5M, France 14.6M, UK 14.0M, other 77.6M. Japan refill is most VLCC-intensive (100% imported crude, 70% MEG). IEA 90-day rule is LEGAL OBLIGATION for non-US members (faster refill than US’s politically-paralyzed pace). China structural building runs in parallel at 500K-1M bpd through Q3 2026 (Goldman). Combined: IEA refill + China = ~270-770 kbpd structural demand = 2.7-7.7% of crude trade = 28-66 VLCC-equivalents. Plus shadow-fleet shift (15-30) + MSC-Sinokor floor = ~60-90 VLCCs of mainstream demand boost = 6-10% of fleet. OFFSETS 2026-27 delivery wave (~50 ships).

📄 Global SPR Refill Recalculated (EN) → 全球 SPR 补库重新测算（中文）→

Cumulative revised stance across all pages 23-34:

🟡 Cycle Position v3 — Bull-Case Stress Test (May 28, 2026)

User pushed back hard on prior bear lean with 3 strong bull arguments. Validated against primary data: (1) MSC-Sinokor alliance formalized Feb 2, 2026 (SAS Lux acquired 50% of Sinokor) — combined fleet 130-150 VLCCs = up to 40% of compliant spot market — major rate-floor mechanism analogous to 2M container alliance. (I should have flagged this in #27/#28.) (2) IEA released 400M bbl SPR (largest ever); US 172M; refill demand real but slow at historical 12M/yr pace (33 yr to refill). (3) Iran normalization → shadow-fleet shift of ~30 VLCCs from sanctioned trade to mainstream fleet (95% of 1,100-1,400-ship shadow fleet already >15 yrs old). Honest revised stance: drawdown estimate -30 to -60% → -15 to -45%; bull/base/bear weights 25/50/25 → 40/45/15; cycle stage “Stage 5 confirmed” → “Stage 4 late / Stage 5 early — bimodal”.

📄 Cycle Position v3 (EN) → 周期定位 v3（中文）→

Key revised findings:

🎯 Mid-Rally Pullback or Post-Peak? — Cycle Position v2 (May 28, 2026)

Answers 4 sharp questions: (NEW Q) Did container 2020-2022 have similar 25-30% mid-rally pullbacks? YES — ZIM dropped −25 to −33% Sep-Oct 2021 then rallied to $91.23 ATH. But absorbed because freight + earnings were still rising. VLCC now has the inverse setup. (Q4) Q1 2026 VLCC orderer profile: Greek-dominated (Capital Ship Mgmt 11 ships at Hengli, Cape Shipping 2, Navios + Monte Nero 4) = textbook cycle-top signal. Chinese SOEs measured: 招商轮船 10 VLCCs at Dalian Mar 30 (¥8.57B), 中远海能 6 (Nov 2024, first since 2017). (Q2) Prior tanker cycles (2004/2008/2015/2020): stocks lead/coincident with rates by 0-1 quarter — NOT 9-18 months like container (tanker is spot-based; container had long contracts). So VLCC earnings-momentum cushion is much shorter. (Q3) DHT Q1 2026 dividend = $0.64 (annualized 17% yield); modeled bearish scenario still returns 40-50% of current price as dividends over 2026-27.

📄 Cycle Position v2 (EN) → 周期定位 v2（中文）→

Verdict — revised stance vs #27/#28:

🗺️ Where Are We in the VLCC Cycle? — Container 2020–2024 Analog Re-applied (May 28, 2026)

Direct timeline-and-orderbook comparison between the container 2020–2024 cycle and current VLCC cycle. COSCO Holdings (601919) peaked July 2021 → FY2022 was peak earnings → stocks led earnings by ~12–18 months. ZIM peaked March 2022 → bottomed Aug 2024 at $6.50 (−93%). Container orderbook hit 28.9% of fleet (early 2023); Q1 2026 VLCC orderbook is 17–26% — lower but lead time 30–48 months vs container’s 24. Best estimate: VLCC is in the container-equivalent of April–May 2022 — post-stock-peak, post-freight-peak, pre-confirmed-earnings-peak, pre-delivery-wave (2028–2030 for VLCC vs 2023–2024 for container).

📄 Container Analog Re-applied (EN) → 集运类比的重新校准（中文）→

Key findings (Step 1 — draft, subject to Step 2 audit):

🔁 VLCC Selloff Fact-Check & Revised Analysis (May 28, 2026, supersedes #23/#24)

Substantial corrections to the previous-day’s selloff analysis after reader review identified 4 errors. Real trading routes (TD22 USG→China $59.26/mt vs 2025 avg $34.01/mt) still print 2–3× last year’s averages; China is BUILDING crude inventory to a record ~1.17 bn bbl (tank util only 62%), not drawing; orderbook is a 2028–2030 problem (only ~29 VLCC deliveries in 2026 / ~24 in 2027); and stocks peaked at war-start (DHT $20.55 March 2, war Feb 28), not after — so only ~5–7 ppt of the drawdown is war-premium reversion, the rest is super-cycle thesis de-rating.

📄 Fact-Check & Revised Analysis (EN) → 事实核查与修订（中文）→

Key revised findings (Step 1 — draft, subject to Step 2 audit; 38 primary sources cited):

🔻 VLCC Post-Peak Selloff: Initial Draft (May 28, 2026 — SUPERSEDED by #25/#26)

Original Step 1 draft contained four substantive errors (TD3C as paper price; China inventory direction; orderbook delivery timing; war timing). Read the corrected analysis at #25/#26 above. Original kept for transparency.

📄 Original draft (EN) → 原草稿（中文）→

📈 DHT Holdings Q1 2026 Earnings Deep Dive (May 5, 2026)

Blowout quarter: Revenue $186.5M (+134% YoY), EPS $1.02 (beat consensus by 67%), spot TCE $106K/day, Q2 bookings at $189,500/day. TD3C at all-time high $420K+/day. Full Day1Global framework analysis with 6 rate scenarios.

📄 DHT Q1 2026 Earnings (EN/CN Summary) → Full EN Report → 完整中文报告 →

Key findings:

📊 Sinokor Dominance & Container Shipping Analog (NEW — April 23, 2026)

How Maersk’s 2020–2022 container shipping playbook maps onto the 2026 VLCC super-cycle — with Sinokor controlling 40% of the global spot VLCC market (vs Maersk’s ~17%).

📄 Sinokor & Container Analog Analysis (EN) → 中文版 →

Key findings:

🚢 VLCC Market: Supply Shortage & SPR Restocking (NEW — April 2026)

Standalone market analysis: why the VLCC market faces a multi-year structural shortage driven by shadow fleet exit, regulatory attrition, and the largest SPR restocking cycle in history.

📄 VLCC Supply Shortage Analysis (EN) → 中文版 →

Key findings:

🇺🇸 US-Listed: DHT Holdings vs Frontline (FRO)

Deep-dive comparative analysis of the two largest US-listed pure-play VLCC operators, using the Day1Global tech-earnings-deepdive framework across 5 AI models.

📄 DHT vs FRO 深度分析（中文）→

Key findings:

📎 English version: 05_Deep_Dive_Day1Global_Framework.md

🇨🇳 A-Share: 招商轮船 (CMES) vs 中远海能 (COSCO Energy)

Comprehensive analysis of China’s two largest VLCC operators on the A-share market, with 中远海控 container cycle (2020-2022) PE/PB compression as valuation reference.

📄 招商轮船 vs 中远海能 深度分析（中文）→

Key findings:

📎 English version: 07_CN_AShare_VLCC_Report_EN.md

🔬 US-Listed: Keysight Technologies (NYSE: KEYS) — NEW

Multi-model deep dive on the #1 global test & measurement company — applying Day1Global framework, operating leverage analysis, and 6-perspective valuation to the AI/5G super-cycle beneficiary.

📄 Keysight Deep Dive (EN) → 是德科技研究（中文）→

Key findings:

🔩 A-Share: 中钨高新 (China Tungsten High-Tech, 000657.SZ)

Cyclical metals deep dive on China’s #2 tungsten producer — applying CRule 1–10 (Two-Cycle Backtrack, PE Compression, Operating Leverage) to the 2025–2026 tungsten super-cycle.

📄 中钨高新 周期性金属深度研究（中文）→

Key findings:

📎 English version: tungsten/report_en.md

📁 All Reports

# File Language Content
01 Full Report EN English Initial multi-model DHT vs FRO
02 Session Summary EN English Methodology summary
03 Full Report CN 中文 初始多模型DHT vs FRO
04 Session Summary CN 中文 方法论摘要
05 Deep Dive EN English DHT vs FRO Day1Global framework
06 Deep Dive CN 中文 DHT vs FRO Day1Global框架
07 A-Share Report EN English CMES vs COSCO Energy
08 A-Share Report CN 中文 招商轮船 vs 中远海能
09 Tanker Peer Universe EN English 7-Company Crude Tanker Comparison (DHT/FRO/INSW/ECO/TNK/NAT/CMBT)
10 Tanker Peer Universe CN 中文 7家原油油轮公司同业对比分析
11 DHT vs FRO Apr Update EN English UPDATED Hormuz Crisis + Structural Supply Thesis
12 DHT vs FRO Apr Update CN 中文 更新 霍尔木兹危机 + 结构性供给论证
13 VLCC Supply Shortage EN English NEW Supply Crunch, Shadow Fleet Exit & SPR Restocking
14 VLCC供给短缺分析 CN 中文 新 供给紧缩、影子船队退出与SPR补库
15 Tungsten Report EN English 中钨高新 Cyclical Metals Deep Dive
16 Tungsten Report CN 中文 中钨高新 周期性金属深度研究
17 Keysight Report EN English NEW Keysight Technologies (KEYS) T&M Deep Dive
18 是德科技研究 CN 中文 新 是德科技 测试测量行业深度研究
19 Sinokor & Container Analog EN English NEW Sinokor 40% Dominance, Container Boom Parallel & VLCC Bull Case
20 兴高海运与集运类比 CN 中文 新 兴高海运40%定价权、集运牛市类比与VLCC牛市论证
21 DHT Q1 2026 Earnings EN English Q1 2026 Earnings Deep Dive & Call Analysis
22 DHT Q1 2026 财报分析 CN 中文 Q1 2026财报深度分析与电话会议解读
23 VLCC Post-Peak Selloff EN English SUPERSEDED — Initial draft (corrected by #25)
24 VLCC 顶部回调分析 CN 中文 已修订 — 初版（被 #26 修订）
25 VLCC Selloff Fact-Check & Revision EN English Fact-check with 38 primary sources; revised core conclusion (war premium only ~5–7ppt of drawdown)
26 VLCC 回调事实核查与修订 CN 中文 38 个一手来源；修订核心结论（战争溢价仅占回撤约 5–7 个百分点）
27 VLCC Container Analog Updated EN English Where in the cycle: VLCC = container April-May 2022 equivalent
28 VLCC 集运类比重新校准 CN 中文 周期位置：VLCC = 集运 2022 年 4-5 月等价位置
29 VLCC Cycle Position v2 EN English Mid-rally vs post-peak; Q1 2026 orderer profile (Greek-dominated); prior tanker lead-lag; DHT dividend model
30 VLCC 周期定位 v2 CN 中文 中段 vs 顶后；Q1 2026 下单方画像（希腊主导）；过往油运时滞；DHT 分红模型
31 VLCC Cycle Position v3 EN English Bull-case stress test: MSC-Sinokor alliance, SPR refill, shadow-fleet shift
32 VLCC 周期定位 v3 CN 中文 多头论据压力测试
33 Global SPR Refill Recalculated EN English Math correction — global (not US-only) 400M SPR refill + China structural + 90-day IEA obligation; demand 4-6x prior estimate
34 全球 SPR 补库重新测算 CN 中文 算术修正 — 全球（非美国单一）4 亿桶补库 + 中国结构性 + IEA 90 天义务；需求是先前估计的 4-6 倍
35 AI Industry Analysis EN English NEW What did AI actually revolutionize? $1.2T capex vs $150-200B AI revenue; coding is the only fully revolutionized category
36 AI 行业分析 CN 中文 新 AI 究竟革命了什么？$1.2T 资本开支 vs $150-200B AI 收入；编程是唯一完全被革命的类别
— Modeling Stash English NEW Cyclical valuation framework, sell signal backtest, momentum algo
— 🔴 Hormuz Tracker English LIVE Daily VLCC transit monitor — manual input dashboard

📝 Project Tracking

🔬 Methodology

*Last updated: May 28, 2026 Created by liqiqiii*