2026-07-282026-07-272026-07-242026-07-232026-07-222026-07-212026-07-202026-07-172026-07-162026-07-152026-07-142026-07-102026-06-292026-06-192026-06-182026-06-172026-06-162026-06-152026-06-122026-06-112026-06-102026-06-092026-06-082026-06-052026-06-042026-06-032026-06-022026-06-012026-05-302026-05-292026-05-282026-05-272026-05-252026-05-222026-05-212026-05-202026-05-192026-05-182026-05-152026-05-11
Stock-market dispatch — 2026-06-15
Auto-generated by the daily research routine (headless Pi run). Summary of what changed in the wiki today; the wiki is the truth. Research output, not advice — brain never trades.
Top of mind
Risk-on recovery continued; thin-cluster steer produced two net-new chains and one belief-flip. The semis/uranium/materials complex kept rallying (ARM +11.3%, KLAC +5.5%, INTC +6.5%, SCCO +4.2%, TLN +4.6%, PWR +3.6%, CEG +2.9%, CCJ +2.0%; LIN printed a fresh 52w high $523.57). With the book ~50% ai-infrastructure (⚠ over), the day's generative budget went to thin verticals: prospect-chains mined two hypotheses — water-infrastructure off the El Niño forcing function (XYL/AWK) and a second NdFeB-magnet demand vector from commercial humanoids (MP/USAR/TSLA) — and the bucket scans (transport, consumer) corroborated existing transport/consumer hypotheses. The sharpest update: a NOW reframe — today's research + the live-trade feedback both say ServiceNow is a relative winner of the agentic-seat-erosion chain (monetizing Now Assist >$600M→>$1B ACV), not its victim; the per-seat incumbents (CRM/WDAY/ADBE/TEAM/MNDY) take the derate. One calibration fired. GLP-1/WST chain hardened into its Jun 29 / Jul 1 catalysts. 0 live-feedback flags (<−10% sector-excess); ingest was corroboration-heavy.
Strongest-conviction buys
Ranked shortlist from step 4c — chains, not advice. Prices 2026-06-15 (twelvedata, 59/63 fetched; 4 expected non-fetchables ASE/BESI/GNKG/HXSCL carry prior).
| # | Ticker | Causal chain (mechanism) | Conviction | Valuation vs base | Fundamentals (one line) | Next catalyst |
|---|---|---|---|---|---|---|
| 1 | LIN | ras-laffan-halt-to-lin-helium-pricing-power | High (0.78) | At base — $523.57 new 52w high | High-margin industrial gas, net-cash, dividend grower; best live leg (+2.3% sector-excess) | Late-summer Ras Laffan restart; APD Q3 (late July) |
| 2 | MU | helium-cliff-to-hbm-supply-crunch + hbm-cowos-as-binding-bottleneck | Medium-high (0.68) | Narrowed — $981.61 −1.4%, −10% from hi | HBM sold out on 3–5yr fixed-price contracts; ~74% GM; DRAM inventory the tell | Q3 FY2026 earnings Jun 24 (the gate) |
| 3 | CCJ | kazatomprom-supply-cut-to-western-uranium-premium + ai-capex-to-power-and-materials-cascade | Medium-high (0.72) | Yes — $100.96 +2.0%, −25% from hi | Term $90 (14-yr high), ~1B lbs uncovered, Kazatomprom 2026 cut | Utility contracting; India 2027 |
| 4 | WST | glp1-injectable-supply-chain-bottleneck | Medium-high (0.66 ↑) | Marginal — $328.82, −2% from hi (held up) | GLP-1 ≈18% of sales; HVP devices +29% (additive vs oral); per-unit consumable | 503B comment-close Jun 29; Medicare GLP-1 $50 demo Jul 1 |
| 5 | TSM | hbm-cowos-as-binding-bottleneck | Medium-high (0.62) | Marginal — $423.93 +0.7%, −6% from hi | CoWoS/HBM binding bottleneck; 50%+ GM, backlog | CoWoS adds; HBM4 ramp |
Watchlist
| Ticker | Thesis | Conviction | Price | Undervalued vs base? | Next catalyst |
|---|---|---|---|---|---|
| LIN | ras-laffan-halt-to-lin-helium-pricing-power | High | $523.57 ↑ | At base (new hi) | Late-summer restart; APD Q3 |
| MU | helium-cliff-to-hbm-supply-crunch | Medium-high | $981.61 | Narrowed | Q3 earnings Jun 24 |
| CCJ | kazatomprom-supply-cut-to-western-uranium-premium | Medium-high | $100.96 ↑ | Yes | Contracting; India 2027 |
| CEG | pjm-capacity-prices-to-nuclear-premium | Medium | $253.76 ↑3% | Dislocated (−39% from hi) | FERC RM26-4 end-June (large-load rule, likely NOPR); Calpine lockup Jun 30 |
| WST | glp1-injectable-supply-chain-bottleneck | Medium-high ↑ | $328.82 | Marginal | 503B Jun 29; Medicare demo Jul 1 |
| CF | hormuz-nitrogen-supply-shock-to-cf-risk-premium | Medium | $109.48 ↑3% | Yes | H2 nitrogen season |
| PWR | pwr-transformer-moat-to-eps-doubling | Medium | $707.74 ↑4% | Marginal (near hi) | FERC end-June |
| MP | drone-warfare-demand-to-mp-hree-rerate | Medium | $57.55 | Yes (−43% from hi) | HREE circuit; + new humanoid-demand leg (below) |
| NOW | agentic-ai-seat-erosion-to-saas-rerate | Medium (role flipped → long/winner) | $102.15 | Yes (−52% from hi) | Q2 FY2026 earnings |
| UNP/NSC | up-nsc-transcontinental-merger-to-pricing-power | Medium | $272.70 / $313.91 | Merger-arb | STB supplemental Jul 27; close H1'27 |
| INTC | taiwan-chokepoint-to-allied-reshoring | Exit review | $124.57 ↑7% | No — above ~$89 | VLSI gate; equity-raise watch |
Undervalued candidates (today)
- CCJ $100.96 — term $90 (highest since 2008), contracting below replacement, −25% from hi; +2% today (recovery off the 06-11 dislocation).
- CEG $253.76 — still −39% from hi; +2.9% today; medium conviction into end-June RM26-4 (large-load rule, likely NOPR/defer) + Calpine lockup Jun 30.
- NOW $102.15 — −52% from hi but reframed as a relative winner (monetizing agentic ACV); the extreme derate is theme-wide (sector-excess only −2.9%), not NOW-specific — Q2 is the test.
New chains to investigate (hypothesis-stage)
- el-nino-drought-to-water-infrastructure-capex — XYL/AWK/WTS. Confirmed El Niño forcing function (NOAA 82–96%, record DSCI 202 drought) → structural water-infra capex step-up → water pure-plays the El Niño concept lists but never wired. Graduate on an XYL/AWK earnings/rate-case citing drought-resilience capex. Thin water/materials cluster. → /explore-chain.
- humanoid-actuator-magnet-demand-to-us-ree-stack — MP/USAR/TSLA. Commercial humanoids (Unitree G1 ~$30/hr parity) add a second NdFeB high-torque-magnet demand vector on the confirmed China-HREE chokepoint → demand support for the US producer stack. Graduate on a source tying humanoid volume to magnet tonnage. → /explore-chain.
- Carryover: talen-energy-replacement-cost-pjm-scarcity (TLN), mgm-diller-bid-floor-japan-dubai (MGM — premium-K tailwind added today), ai-capex-derate-to-private-credit-contagion (ARES/OWL/BX/APO).
New theses (now active)
- None promoted to active — corroboration-heavy ingest; today's two net-new chains landed hypothesis-stage (water-infra, humanoid-magnets).
Updated theses (existing active)
- glp1-injectable-supply-chain-bottleneck (WST) — both catalysts dated and the beneficiary leg hardened: Medicare $50 Bridge Jul 1, FDA 503B comment-close Jun 29 (both push volume to branded); WST HVP Devices +29%, raised guidance, new 165k-sq-ft injectable facility (oral-pill bear case additive-not-substitutive so far). Step 4 strengthened; signal re-dated.
- agentic-ai-seat-erosion-to-saas-rerate (NOW) — role flip + short-leg hardened: NOW reframed victim→relative-winner; short basket CRM/WDAY/ADBE/TEAM/MNDY (Feb-2026 SaaSpocalypse, Atlassian seat decline, Workday RIF). Calibration fired; signal re-dated.
- up-nsc-transcontinental-merger-to-pricing-power (UNP/NSC) — STB accepted the revised app May 28 but held in abeyance; supplemental due Jul 27; close H1'27; $320/NSC. Step 1 updated; signal re-dated.
- iran-fuel-shock-consumer-bifurcation (MGM/TJX) — new premium-K arm: Iran gas +50% since Feb 27 widens the K-shape; hotel demand trifurcates (luxury wins, low-end RevPAR under pressure). Signal re-dated.
- ai-capex-to-power-and-materials-cascade (CCJ) — Chamath: AI capex ≈ $100B/GW (up 20×); AI marginal cost ≫ 0 reframing. Step 3 evidence added; signal re-dated.
Contradicted / weakened
- 1 calibration (
relative-position-mismeasured): agentic-ai-seat-erosion-to-saas-rerate — NOW was on the wrong side of its own chain (carried at the theme average); flipped to long/relative-winner with the per-seat basket as the paired short. Conviction stays medium pending Q2. - New tension (bounds, doesn't falsify): compute-utilization-overhang-as-latent-supply — Anjney Midha (Odd Lots): software can reclaim ~30–40% latent GPU supply, which bounds (not overturns) the HBM/CoWoS scarcity theses — demand is "perpendicular"/growing and the packaging chokepoint is orthogonal to utilization. No mechanism (AMP private, no tradeable).
Open questions worth a human's eye
- NOW Q2 FY2026: does consumption-ACV durability beat core-seat softness (confirms the winner-flip)?
- Does an XYL/AWK earnings/rate-case actually cite drought-resilience capex (converts the water chain)?
- What ties humanoid unit volume to NdFeB magnet tonnage on the US stack (converts the humanoid-demand leg)?
- FERC RM26-4 end-June: NOPR (defers CEG/TLN co-location) or substantive rule?
What I looked at
- Run status: full headless Pi daily run (steps 0.5–7) completed synchronously; paced 63-ticker price fetch ran to completion in-turn.
- Macro buckets researched: Transport/industrials ex-AI (#11), Consumer shift (#10) — both thin/absent verticals per 2a (ai-infrastructure ~50%, ⚠ over).
- Ingestion veins (0.5–0.7): podcast-ingest 5 transcripts (Odd Lots/Midha, Forward Guidance, Biotech Hangout, All-In weekly, Capital Allocators/EOS; Bg2 failed — AssemblyAI 400 again, logged); earnings-ingest 0 (no-op — between Q1/Q2 clusters, MU reports Jun 24; June reporters ORCL/ADBE/LEN not on watchlist); feed-ingest 1 (SemiAnalysis SMIC N+3, partial/paywalled).
- Step-1 refreshes: 2 autoresearch (GLP-1/WST near-term catalysts; NOW/seat-erosion). Step-2 buckets: 2 (transport, consumer). prospect-chains: 2 net-new hypotheses (water-infra, humanoid-magnets).
- Sources promoted: 9; ingested: 9 → new pages: 1 concept (compute-utilization-overhang-as-latent-supply) + 1 person (anjney-midha); 5 mechanisms materially changed (glp1/WST, agentic-seat/NOW, up-nsc, iran-fuel/MGM, ai-capex-cascade); 2 zero-novelty (Biotech Hangout, Capital Allocators/EOS).
- Valuation snapshots: 59/63 tickers fetched (twelvedata, 0 rate errors); anchor concept snapshots refreshed (helium/LIN, hbm/MU, nuclear/CCJ-CEG-BWXT-UEC); others carry prior dates (prices folded into watchlist + signals).
- Live-trading feedback (4e): digest pulled (
performance-2026-06-15, 21 positions) — 0 flagged <−10% sector-excess (worst MP drone −8.5%; NOW −16.3% return but only −2.9% sector → theme-wide, the winner-flip tell). - Paper ledger (4d): weekly mark not due (last 06-11); no falsification triggers fired.
- Calibration events: 1 (NOW relative-position). Conviction buys ranked: 5. New hypothesis chains opened: 2.
- Signal feed: full active/armed set, gated ≥0.4 — see step 7 (Supabase sync performed by the wrapper, not this run).