Pipeline Report — 2026-05-05 #

Cycle Summary #

• Cycle started: 2026-05-05 (15 days after prior cycle, triggered as new)
• Universe: 989 stocks (S&P 500 + STOXX 600)
• Phase 1: yfinance throttling stalled fresh screen twice; reused Apr 20 composite (322 candidates passed Graham, 15 days fresh)
• Phase 2: 7 sector-diversified non-financial candidates deep-analyzed in parallel
• Phase 3: Reddit overlay sparse for EU names (PRU 27/neutral, MOS 5/neutral, SHEL 3/neutral, others 0); top Reddit hype non-overlapping with Graham passers
• Phase 4 outcome: 0 GO, 5 WAIT, 2 PASS

Top Ranked Recommendations #

#1 — BKG.L (Berkeley Group Holdings) #

• Verdict: HOLD with optionality T1 EUR 150 at 3,242p, OR wait June 24 results
• Conviction: MEDIUM-HIGH
• Composite score: 56.4 (#4 by raw rank, #1 by deep-dive verdict strength)
• Sector: UK Homebuilder (NEW exposure)
• Why: Strongest insider signal of the batch — Executive Chairman Robert Perrins bought 28,000 shares (£1M+) over April 1/7/28/30 at 3,829-4,386p, ABOVE today's price. Multiple other directors buying. Per shr-002 this is independent officer open-market with own capital — not family. P/B 0.86x. Net cash £338M. Lynch cyclical with turnaround optionality (50,000-home pipeline ready to release on BoE cuts).
• Entry zone: 2,900-3,100p ideal; 3,242p current is acceptable on insider strength
• Target 90d: 3,829p (analyst mean, +18%)
• Downside: 2,000-2,250p (-30-38%) in deeper UK housing recession
• Kill triggers: FY27 PBT guidance below £1.2bn total, OR net cash <£150M, OR rights issue announced
• Catalyst: 24 June 2026 FY26 results

#2 — PAH3.DE (Porsche Automobil Holding) #

• Verdict: WAIT for May 12-13 H1 results
• Conviction: MEDIUM
• Composite score: 60.7 (#1 by raw rank)
• Sector: Auto/Asset Play (NEW)
• Why: Trading at EUR 30.84, essentially the stress-case floor. 30% discount to NAV (analysts model EUR 43-47). Forward P/E 2x reflects expected VW recovery. But three RED flags: China volumes still falling, EU auto tariffs (May 4 news), Porsche AG EV pivot uncertainty.
• GO trigger: May 12 H1 shows VW equity contribution >EUR 2B AND Porsche AG operating margin >8% in Q1
• Entry zone: EUR 28.50-31.00
• Target 90d: EUR 37.40 (+21%)
• Downside: EUR 25 (-19%)
• Catalyst: 12-13 May 2026

#3 — AKE.PA (Arkema) #

• Verdict: WAIT for May 6 Q1 (tomorrow)
• Conviction: LOW-MEDIUM
• Composite score: 55.0
• Sector: Specialty Chemicals (NEW)
• Why: French specialty chems, EV battery materials secular tailwind (PVDF). Forward P/E 10.7x, 5.8% yield. BUT: 4 consecutive quarterly misses, PFAS escalation (France case Jan 2026 new), recurring restructuring (4 yrs, EUR 250M+) is Schilit AMBER flag.
• GO trigger: Q1 confirms EBITDA trough + EUR 50-55 entry zone
• Entry zone: EUR 50-55
• Downside: EUR 45 (-28%) in cyclical bottom-of-trough
• Catalyst: 6 May 2026 (tomorrow)

#4 — SHEL.L (Shell) #

• Verdict: WAIT for May 7 Q1
• Conviction: LOW-MEDIUM
• Composite score: 55.0
• Sector: Energy (NEW)
• Why: Q1 update flagged $10-15B working capital outflow, Pearl GTL (Qatar) damage = 12-month repair / ~$1B+ CFFO hit. Sector-wide insider sell ratio 237:1 (shr-031) — disruption already priced in.
• GO trigger: Brent stabilization $55-60, new $3.5B+ buyback announced May 7, P/FCF <8x on current FCF
• Entry zone: 2,700-2,900p
• Downside: 1,542p (-53%) at $50 Brent sustained
• Catalyst: 7 May 2026

#5 — UHR.SW (Swatch Group) #

• Verdict: WAIT (entry zone far below current)
• Conviction: LOW
• Composite score: 56.4
• Why: Trading at CHF 181.8, 12% ABOVE analyst mean (CHF 159) — market priced hope, not trough. Gross margin compressing 80%→77%, inventory at 424 days. Hayek family resists capital returns. Activist defeated at AGM.
• Entry zone: CHF 135-150 (need 17-25% further drop)
• Downside: CHF 95-111 (-38-48%)
• Kill trigger: Gross margin <74% OR inventory/revenue >1.3x
• Catalyst: 9 July 2026 H1

PASS List #

MOS (Mosaic) #

• Why PASS: Negative FCF in FY25 (-$535M), confirmed sulfur-cost headwind for Q1 ($250M EBITDA hit), Belarus sanctions easing structurally bearish for potash, zero meaningful insider buying. shr-017 cyclical-trap risk fully active. Re-examine if MOS hits $16-18 trough zone.

BOL.PA (Bolloré) #

• Why PASS: Trades at PREMIUM (~5%) to current-value NAV, not the discount the 0.62x P/B suggests (book illusion from historical-cost equity accounting). Post-dividend (Jun 25) target ~EUR 3.59 NAV vs ex-div price ~EUR 3.89 — flat to slightly negative. Ackman UMG deal probability uncertain (Bolloré family likely to block structure that dilutes European listing). Family-controlled per shr-035.

Diversification Check #

Current Graham portfolio: AGN.AS (financial services) + AIG (financials) + RI.PA (consumer staples). Limited room for more financials.

• BKG.L (homebuilder): CLEAN — new sector
• PAH3.DE (auto/asset play): CLEAN — new sector
• AKE.PA (chemicals): CLEAN — new sector
• SHEL.L (energy): CLEAN — new sector
• UHR.SW (luxury): CLEAN — new sector
• All five candidates pass diversification rules.

Capital Allocation Suggestion #

Recommendation: Hold deployable capital this week. Four binary earnings events May 6-12 will dramatically refresh the picture for AKE/SHEL/MOS/PAH3. Premature deployment violates shr-034.

Earliest defensible deployment:

• After May 12 PAH3 H1: if VW recovery confirms, deploy EUR 200-250 T1 in PAH3
• If RI.PA delivers H2 FY26 GO trigger (China stabilizing, gross margin >57%): deploy pending T2 EUR 220
• BKG.L T1 EUR 150 at 3,242p is defensible NOW on insider strength alone, but suboptimal vs waiting June 24

Composite Weights Used #

Default weights (no IC adjustment yet — first IC computation needs ≥10 resolved predictions):

• 0.30 graham_pct
• 0.20 magic_formula_pct
• 0.30 llm_pct
• 0.10 schilit_clean_pct
• 0.10 reddit_divergence

Reddit Overlay Notes #

• 19,125 posts + 245,459 comments parsed across 15 days
• 2,482 unique tickers, NVDA/GME/MSFT/GOOGL dominate (none Graham-eligible)
• Phase 2 candidates: PRU 27/neutral, MOS 5/neutral, SHEL 3/neutral, all others zero
• No actionable contrarian divergence this cycle — EU names off-radar (consistent with prior cycles)
• No Reddit-hyped + fundamentally-weak names to flag as exit-liquidity traps among Graham candidates

Next Steps #

1. Phase 5 (today): Resolve 14 pending predictions from Apr 3 + Apr 20 cycles (resolution date 2026-05-03 has passed)
2. May 6: Read AKE.PA Q1, update verdict
3. May 7: Read SHEL.L Q1, update verdict
4. May 11: Read MOS Q1 (likely confirms PASS)
5. May 12-13: Read PAH3.DE H1 — primary GO trigger for top composite candidate
6. June 24: BKG.L FY26 results — secondary GO trigger