HScreen · WMB

WMB · The Williams Companies, Inc. 9d old

conviction2

coverage tierT3-PRIMARY-COMPLETE

horizon12 months

Return distribution probability-weighted scenarios, never collapsed to a point

-17%20%

+15%55%

+30%25%

EV +12.35%

why these probabilities: ALL LEGS BUILT FORWARD FROM 2026E ADJUSTED EBITDA OF ~USD 8.5bn (management's raised guide, WMB/WMB_2026Q2.json:67), 1.223bn shares (92,349,144,136 market cap / 75.5), and net debt of ~USD 27bn implied by EV/EBITDA 13.86x on FY2025 metrics. Bridge is EV/EBITDA → equity value → per share, plus the 2.10 annualised DPS (2.8% yield) in every leg. BASE — THE MARKET'S EXPECTATION, NOT OURS. Consensus (fmp:/stable/analyst-estimates? symbol=WMB&period=annual, as_of 2026-08-19, 34 analysts) is 2026 revenue growth +7.4% and EPS growth +14.1%, and management guides EBITDA growth of 11%+. We adopt that, we do not compete with it: no model, no management access, no channel checks, so an independent base case here would be a noisier copy of consensus. EBITDA grows 11% to ~USD 9.4bn on 2027E, multiple HELD FLAT at 14.0x, net debt drifts to ~USD 29bn as capex is part debt-funded → equity ~USD 103bn → ~84.5/share → +12% price, +15% total. Sense-check: street consensus target 85.82 (median 85) implies +13.7% price, +16.5% total — our base is 1.5pp BELOW it, by construction of the flat-multiple assumption rather than by disagreement. BULL — GROWTH MATERIALLY ABOVE THE MARKET. 2027 EBITDA growth of ~14-15% versus consensus revenue +9.9% / EPS +4.5%, on named evidence the street has not modelled: Shelby Connector and Delta Access, which management states are EXCLUDED from the run-rate multiple (WMB

not built — options-implied distribution on the same axis, with the KL number and where the disagreement sits
needs: desk-options implied bands (v1.4 Part 4a) — activates at 60% coverage; currently 33%

Visibility how far ahead this name can be seen

not built — visibility score
needs: visibility_score.py

Trust how much weight this name's own record can carry (credibility Z)

not built — credibility
needs: credibility.py

Sizing survival-first size (half-Kelly) vs the enforced ladder

kelly not computable — no data

Payout durability does the distribution survive a downturn?

PASS

DPS trajectoryrecovered_after_cut

interest cover now3.05x

interest cover at FORWARD rates3.23x

Falsifiers what would prove this wrong

2026 adjusted EBITDA lands at or above the raised ~USD 8.5bn guide
The EBITDA growth trajectory guided at 11%+ for 2026 carries into 2027 rather than decelerating to the consensus +4.5% EPS path
Shelby Connector and Delta Access reach in-service on the disclosed schedule, delivering the growth explicitly excluded from the current run-rate multiple
Dividend cover stays comfortably above 1.8x AFFO while DPS grows ~5% a year
Leverage stays inside the committed 3.5-4.0x debt/adjusted-EBITDA band
The ~USD 4bn/yr growth capex programme continues to be funded and deployed, i.e. the project pipeline is converting

Conflicts surfaced, never netted out (T4)

Our probability-weighted expected total return of +12.35% is 4.1pp BELOW the street's implied +16.5% (consensus target 85.82, 34 analysts, 27 buy / 7 hold / 0 sell). We are not disagreeing about the base case — we adopt	medium
The revision tape is UNAVAILABLE, not neutral: eps_drift and rev_drift both return n_up 0 / n_down 0 / breadth 0.0 with zero outliers over 365 days, meaning no estimate-change rows exist for WMB in the estimates DB. We t	medium-high
Management guides 11%+ EBITDA growth for 2026 (walked up from 8% then 9%) while consensus models 2027 EPS growth decelerating to +4.5% from +14.1% in 2026, with 2026 revenue dispersion at 17.1% and a target range of 75-1	high
p_bear 20% against an interpolated node-pooled base rate of ~2.1% at -17% — a 17.9pp deviation, defended on entry multiple, the FERC permitting falsifier, and consensus's own 2027 deceleration. On the other side p_bull 2	high
No cross-check available from any signal desk: options_view is null (no chain entry for WMB in desk_options.json, so no options-implied probability anchor on either leg), and crowding, technical and factor_profile are de	medium-high
Routing mismatch flagged rather than resolved: this dossier was produced by analyst-aisemi, whose coverage config is the AI/semiconductor/data-centre value chain, on a name whose state-file theme_tag is energy-dividend a	medium

The case against independent adversary, different model family (P6) weakened

Version history from the research record

archived versions3

archived	hash	bytes
2026-08-17T10:34:17	df95153d0d24	36,919
2026-08-19T05:54:53	47203ba67b08	62
2026-08-19T08:32:01	5ad51b7c94b2	36,720

What changed in the latest version

-	API call failed after 3 retries: HTTP 502 — 502 Bad Gateway
+	ticker: WMB
+	as_of: 2026-08-19
+	coverage_tier: T3-PRIMARY-COMPLETE
+	analyst: analyst-aisemi
+	model_producer: claude-opus-5
+	rubric_v: v1
+	provenance:
+	produced_by: analyst-aisemi
+	produced_at: 2026-08-19
+	inputs:
+	- /Volumes/SSD/Hermes/system/evidence-packs/WMB.md
+	- /Volumes/SSD/Hermes/system/state-files/WMB.json
+	- /Volumes/SSD/Hermes/system/policies/rubric_v1.yaml
…	diff truncated

Rendered from canonical artifacts only (T1) — no terminal database exists. Every figure is read from the system's own files at render time; where an artifact is missing the panel names the dependency rather than showing an empty box. No execution path exists anywhere in this interface (T5), not even a link.