Discovery › screen
Screen unswept fields
Date: 2026-07-29 · Command: /screen (coverage-gap sweep) · Framework: analysis_notes.md §0–§4
Method: mapped 41 prior screens + 80 stock files to find fields never foraged, then ran four parallel Scout sweeps across ~100 names.
1. Why this sweep exists — the coverage gap
The research library had run 41 screens and never once screened healthcare, energy (outside nuclear), consumer staples, materials, or telecom.
| Field | Stock files | Screens ever run |
|---|---|---|
| Technology | 39 | ~15 (software variants) |
| Semiconductors | 10 | 3 compares |
| Healthcare | 3 | 0 |
| Energy (oil/gas/midstream/utilities) | 0 (all energy work was nuclear/uranium) | 0 |
| Consumer staples | 0 | 0 |
| Materials | folder did not exist | 0 |
| Industrials (broad) | 1 | smallcap-defense only |
| Macro/ | — | never run |
The portfolios mirror it exactly — by position count, not dollar weight:
A portfolio-specific passage was removed from the public build.
The self-reinforcing problem: the agency only screened where it already held, so it only ever found more of what it already owned. The user's stated goal is a transition from pure high-growth toward balanced growth + income — and the unswept fields are precisely where income lives.
2. Field-level valuation context
Sector ETF multiples, pulled 2026-07-29. None of these fields is cheap at the sector level.
| Field | P/E (TTM) | Yield | Read |
|---|---|---|---|
| XLC Comm Services | 15.4x | 1.33% | ⚠️ Mirage — ~half of XLC is GOOGL + META, both already held in both accounts. Buying the "cheapest sector" would concentrate, not diversify. |
| XLU Utilities | 20.6x | 2.64% | Near 52-wk low. Genuinely unswept. |
| XLE Energy | 21.8x | 2.85% | Cheapest true diversifier, highest yield |
| XLP Staples | 25.9x | 2.64% | |
| XLB Materials | 26.6x | 1.67% | |
| XLV Healthcare | 29.4x | 1.60% | |
| XLI Industrials | 30.5x | 1.11% | Most expensive |
Implication set before the sweeps ran: any gem would have to be an individual mispricing inside an expensive field, not "the whole field is cheap." All four Scouts were explicitly instructed to report "already stripped" rather than manufacture a pick.
3. ⭐ The meta-finding
All four sweeps independently reached the same structural conclusion:
The rotation bought the label, not the cash flow.
Money fleeing growth went into things that looked defensive, and specifically avoided anything carrying a scary narrative — even where the narrative was demonstrably wrong or already inflecting.
The pattern repeats in every field:
| Field | The rotation BOUGHT (now expensive) | The rotation SKIPPED (where value survives) |
|---|---|---|
| Energy | C-corp midstream (+25.9% YTD, 10.8x fwd EV/EBITDA vs 9.9x 3-yr avg) | K-1 MLPs — AMZI at 8.4x vs a 10-yr avg of 10.1x |
| Healthcare | Pharma, managed care, distributors (all near 52-wk highs) | Animal health — left behind on one dated guidance cut |
| Industrials | Rails, waste, aerospace aftermarket, aggregates | Coatings, E&C, HVAC distribution |
| Staples/Telecom | Tobacco + AT&T — MO, PM, BTI, T all now yield below their own 5-yr averages | Packaged food, alcohol, Verizon |
The cleanest single expression of it: in energy, the C-corp/MLP gap is not a mispricing that mean-reverts — it is a fee the market pays you for filing a K-1. ~200–300bp of extra yield that the ETF-and-IRA crowd structurally cannot reach.
The trap embedded in the same finding: most names carrying a volume-decline narrative genuinely are in volume decline. Applying the organic-volume test killed the majority of the cheap screen output (GIS, CAG, CPB, FLO, TAP, UPS, OKE, EMN). The discipline is separating "wrongly feared" from "rightly feared," and only a handful cleared it.
4. The five survivors, ranked for this portfolio
Ranked by fit to the stated growth→income transition, given zero existing exposure to any of these sectors.
| # | Ticker | Field | Price | Fwd P/E | Yield | vs 5yr avg yield | Conv. | Status |
|---|---|---|---|---|---|---|---|---|
| 1 | ZTS | Animal health | $75.78 | 10.25x | 2.72% | +186% 🔥 | 7.5 | ✅ in zone |
| 2 | PPG | Coatings | $112.00 | 12.9x | 2.65% | +32% | 7.0 | ~at zone ($110) |
| 3 | VZ | Telecom | $46.01 | 8.72x | 6.15% | +0.5% | 7.0 | ✅ in zone |
| 4 | STZ | Beer | $129.05 | 10.41x | 3.19% | +82% | 7.0 | ✅ in zone |
| 5 | MPLX | Midstream MLP | $57.73 | 11.85x | 7.45% | −9% | 7.0 | ⚠️ taxable only |
🥇 1. ZTS — Zoetis · Animal health · $31.8B · $75.78
The best single name in the sweep. Quality is not in question: 72% gross margin, 37% operating margin, ~23% ROIC, 7.2% FCF yield, FCF compounding at +19.9% over three years. It trades at 10.25x forward, −53% off its high, because of one dated event — the May 7, 2026 guidance cut on US companion-animal share loss that caused a −21.5% single-day drop. The market moved from pricing ZTS as a 30–40x compounder to pricing it as terminally declining generic pharma. The financials do not yet support the second story.
A portfolio-specific passage was removed from the public build.
Worth naming explicitly: this is the exact inverse of the AMAT signal found the same day — insiders dumping $157M into a top while analysts raised targets. Same signal, opposite sign, same week of research.
Moat — intangibles + efficient scale. Animal health is structurally one of the best-designed corners of healthcare: no PBMs, no insurance reimbursement, no Medicare, no IRA price negotiation. Pet owners pay cash, and pet spend is among the most recession-resistant consumer categories. Vets prescribe on habit and detailing relationships, creating real switching friction; only two or three full-line players are viable globally and ZTS is #1. In an RFK-era HHS environment, that policy insulation is worth real money and nothing else in healthcare offers it.
Bear case. Elanco has landed three simultaneous hits: Zenrelia vs Apoquel/Cytopoint, Credelio Quattro underpricing Simparica Trio, and Librela — the osteoarthritis monoclonal meant to be the next billion-dollar franchise — carries an FDA safety signal and weakening vet confidence. Revenue growth has fallen to 3%. Management levered up $3.85B to repurchase $3.23B of stock above $75, pushing Debt/Assets to 59.7%. A securities class action covers 2025-01-14 to 2026-05-06. If Librela is structurally impaired and Elanco holds its price undercut, 10x forward is fair, not cheap.
FV $100–125 · entry <$80 (in zone) · conviction 7.5 Graham IV $32.80 — disregarded; BVPS is $7.84 only because of buybacks, so the formula punishes an asset-light IP business for returning capital.
🥈 2. PPG — PPG Industries · Coatings · $24.96B · $112.00
A 55-year Dividend Aristocrat with 41% gross margins trading at 12.9x forward — a multiple normally reserved for commodity chemicals. The market repriced PPG as if it were LYB or DOW because it sits in the Basic Materials bucket; its margin structure says otherwise. Yield 2.65% vs a 2.01% five-year average (+32%).
The moat evidence is in the margin line. PPG pushed through 2022 inflation pricing and kept ~500bps of it four years later (36.1% → 41.3% gross margin) while volumes went nowhere. A commodity chemical cannot do that. The moat is specification lock-in — in aerospace and automotive OEM, a coating is qualified into an airframe or body-shop process and re-qualification is a multi-year regulatory exercise costing far more than the paint. Aerospace aftermarket coatings and sealants are an annuity tied to fleet flight hours.
Cycle position: trough-to-early-recovery. FY26 consensus EPS $7.90 vs $6.94 delivered in 2025 — rising, not peaking. FCF payout 54%, plus $790M of buybacks shrinking shares 1.5%/yr.
Bear case. Revenue CAGR is +0.6% over three years — zero organic growth, so the entire earnings story is margin and buyback. If 41.3% gross margin is the peak (achieved on price while raw-material costs fell) and it mean-reverts toward 37%, there is no engine left. Graham IV $75.51 vs $112 — driven by $6.15B goodwill against $7.94B equity, i.e. near-zero tangible book.
FV $125–145 · entry <$110 · conviction 7.0
🥉 3. VZ — Verizon · Telecom · $192.1B · $46.01
The purest income pick, and the only large defensive-income name that did not re-rate. Its 6.15% yield sits on top of its own 5-year average (6.12%) while AT&T's compressed from 6.26% → 4.64% and Altria's from 7.64% → 5.66%. The rotation bought the peers and skipped this one.
The cash flows are inflecting: 2026 guidance of ≥$21.5B FCF against a $192B market cap is an 11.2% forward FCF yield. New CEO Dan Schulman (Oct 2025) has attached $5B of cost cuts, the closed $20B Frontier fiber acquisition, and a $25B/3-year buyback — the first real share-count reduction in Verizon's modern history. FCF payout 53–58%. 20 consecutive years of dividend increases.
Moat — efficient scale. Not brand, not switching costs (number portability killed those). A three-player US wireless oligopoly protected by a network and spectrum position costing well north of $180B to replicate, with C-band holdings effectively unbuyable. Frontier converts Verizon into a converged fiber+wireless operator — structurally the advantaged position, and precisely what is killing Charter.
Bear case. Growth is structurally ~0% and the dividend CAGR is only 2.2% over ten years — this is income today, not income growth; a bond proxy with equity risk, and it must be sized as such. Rising rates attack it directly: $181.6B of debt at ~3.7% average cost refinancing into a 4.71% 10-year is a mechanical drain, already visible in net income falling 6.9%/yr on flat revenue. The classic telecom traps (capex treadmill, debt-funded dividend) genuinely do not apply here — which is why it qualifies.
FV $47–62 · entry ≤$47 (in zone), add aggressively <$43 · conviction 7.0 🚩 Open question that decides the thesis: postpaid phone net adds. Verizon has been the share donor in US postpaid wireless for years, and the Scout could not verify current trajectory. Must be resolved before committing capital.
4. STZ — Constellation Brands · Beer · $22.0B · $129.05
The cleanest failure of the market's broad-brush GLP-1 / Gen-Z-drinking narrative in the entire field. De-rated to 10.41x forward at $129 against a 52-week low of $126.45 — but the volume data says the decline already inflected: FY26 depletions −2.1%, with Q4 FY26 turning positive at +0.6% and shipments +1.1%. More decisively, STZ was the #1 dollar-share gainer in US Circana-tracked beer, +0.4 share points — taking share in a shrinking category, the opposite of what's priced.
Moat — brand intangibles + a legally unbreakable license. The 2013 DOJ consent decree breaking up the AB InBev/Modelo merger handed Constellation a perpetual, exclusive, irrevocable US license to Corona, Modelo and Pacifico. It cannot be revoked, renegotiated, or bought back. Modelo Especial overtook Bud Light as America's #1 beer. A well-funded rival cannot license these brands, cannot buy them, and would need a decade to build an alternative. One of the more durable moats in consumer staples, priced at 10x. Yield 3.19% vs 1.75% 5-yr average (+82%); FCF payout ~40%; shares shrinking 3.0%/yr.
Bear case. The volume inflection is one quarter — full-year depletions were still negative and Modelo Especial, the crown jewel, declined ~3%. Roughly half of beer volume skews to a Hispanic consumer demographic under sustained immigration-enforcement pressure — a policy variable, not a business one. Leverage ~3.2x. Dividend growth has stalled to ~1%/yr, which reads as a management signal that they see less cushion than the 40% payout implies. Graham IV $106.65 is below the current price — overridden on goodwill distortion, and that override is the key judgment call in this pick.
FV $135–170 · entry ≤$135 (in zone), strong add <$125 · conviction 7.0
5. MPLX — Midstream MLP · $58.6B · $57.73 ⚠️ account-constrained
The highest yield in the sweep: 7.45% (7.7% forward) with 1.3x coverage and 3.7x leverage — a combination that does not exist in C-corp midstream at any price today. Management guides 12.5% annual distribution growth for 2026 and 2027. Beta 0.45 — near-zero correlation to the tech book. Marathon-anchored (~64% MPC-owned), Marcellus/Utica gathering plus Permian NGL logistics.
Moat — efficient scale + captive sponsor. Gathering systems are natural local monopolies: once pipe is in the ground across a producer's acreage, a rival must duplicate right-of-way to win volume already under minimum-volume commitments. MPC is both 64% owner and anchor customer, so a large slice of revenue is intra-family and effectively non-competable.
🚩 The unresolved flag: all-in FCF payout went 69% → 71% → 73% → 98% over four years. Management's comfortable "1.3x coverage" is a DCF figure computed before growth capex. On an all-in basis MPLX is distributing nearly everything it earns and funding the $2.4B growth program with debt — FY25 shows $4.32B of acquisitions against $6.59B of debt issued, pushing debt $21.4B → $26.2B in one year.
A portfolio-specific passage was removed from the public build.
FV $54–62 · entry <$54 (≈8.0% fwd yield) · conviction 7.0
5. Named traps — do not buy the screen output
The single most valuable output of this sweep. Each of these screens as cheap and fails on inspection.
| Ticker | Headline | Why it's a trap |
|---|---|---|
| UPS | 12.8x fwd, 6.27% yield | Payout 1.22x earnings, EPS −53% yoy. Amazon glide-down is structural, not cyclical. The trap of the industrials screen. |
| OKE | 14.3x fwd, 4.76% yield, 8% off high | Diluted shares compounded +11.8%/yr (448M→626M) funding M&A while EPS went nowhere ($5.48→$5.42). FY25 dividends $2.58B exceeded FCF $2.45B. Growth bought with paper, income funded with debt. |
| GIS | 11.3x fwd, 6.43% yield vs 3.68% avg | Fails the volume test outright: organic sales −2% on −1% price/mix and −1% volume. FCF −8%/yr, 81% FCF payout and rising, ~4.1x levered. Dividend is a freeze candidate. |
| PEP | Dividend King, 4.13% vs 3.03% avg, near 52-wk low | 99.6% FCF payout ($7.67B FCF vs $7.64B dividends) while debt climbed $39.5B→$49.9B. The streak is part-funded by the balance sheet. Elliott's $4B stake is a real catalyst — watchlist, not buy. |
| CHTR | 3.2x fwd, 0.98x book, 23.5% FCF yield | Broadband losses accelerating (−120k Q1, −172k Q2 vs −72k/−80k prior yr). $11.7B/yr capex against $4.4B FCF on declining revenue, 4.4x levered, 45% short float. Bought back $5.13B of stock against $4.42B FCF — funding buybacks with debt while the subscriber base shrank. |
| SCCO | copper-deficit story | 25.4x fwd near 52-wk high after +73% EPS and +41% revenue growth. Textbook low-multiple-on-peak-earnings. |
| DMLP | 11.85% yield | Payout ratio 2.01, fully variable distribution tracking the oil strip. Commodity beta in an income costume. Also a K-1. |
| PFE | 8.76x fwd, 6.84% yield | Payout 1.31 — dividend exceeds earnings. Classic yield trap. |
| CAG · CPB · FLO · EMN · TAP | 6.7–7.9% yields | All growing revenue on price while volumes fall, levered from old M&A, payouts climbing. FLO at a 283% payout is uncovered. |
| BDX · ZBH | 12.3x / 10.5x fwd | Both earn ROIC ~5% — value destruction. On the healthcare list because cheap, not because good. |
| BTI | 11.8x vs PM's 21.0x | Smokeless transition genuinely working, but DYT disqualifies on price: 5.29% yield vs a 7.47% 5-yr average = already re-rated. The CAD$32.5B Canadian CCAA settlement is not a one-off — multi-year payments push true FCF payout to ~65–70%. Watchlist under $56. |
6. Verdict
Three of four fields came back "largely stripped," and that is the honest headline. Healthcare was the S&P's leading sector in 2026 — not a forgotten corner. Industrials are +12% YTD against a flat S&P and among the most overbought groups in the market. Energy's broad midstream returned +25.9% through June and now trades above its 3-year average multiple. The rotation the user has been watching has already washed through these meadows. No Scout was permitted to manufacture a pick, and each named its field's condition plainly.
What survived is narrow, coherent, and shares one shape: businesses whose earnings or sentiment are depressed by identifiable, dated, non-secular causes — a single guidance cut (ZTS), industrial-volume softness (PPG), a capex hangover the market hasn't re-underwritten (VZ), a demographic scare with the volume already inflecting (STZ), and a tax-friction discount that does not close (MPLX). In every case the underlying franchise is intact and the evidence is in the margin or share line, not the narrative.
Two structural cautions worth carrying forward. First, all three industrials picks fail Graham badly (PPG −33%, ACM −39%, WSO −54%) on goodwill-heavy, near-zero-tangible-book balance sheets — expected for asset-light businesses, but three-for-three is a pattern, and this cohort is more fragile in a hard landing than the multiples suggest. Second, rising rates are a live headwind to the income half of this list — VZ as a bond proxy, MPLX and PPG through refinancing cost — with the 10yr at 4.71% and ~72% odds of a 2026 hike.
Recommended sequence
/analyze ZTS— the highest-quality business found, in zone now, with the strongest insider signal in the sweep. Must settle: is Librela's safety signal a durable impairment, and can Zoetis hold parasiticide pricing against Credelio Quattro?/analyze VZ— the purest fit for the growth→income transition and the only large defensive-income name that hasn't re-rated. Must settle: postpaid phone subscriber trajectory./analyze PPG— cleanest §4 dividend grower of the five. Must settle: is 41.3% gross margin durable or the peak? That single variable is a $145 stock versus a $95 one.STZ and MPLX are strong but each carries one binding constraint (Hispanic-consumer/tariff concentration; K-1 taxable-account-only placement).
Sizing note. These are five first positions in sectors where both portfolios currently hold nothing. They are genuinely uncorrelated with the existing tech book — which is the point — but that also means no existing position funds them. Any purchase is new capital or a trim elsewhere.
Data gaps flagged
fin.pyreturns broken EV/EBITDA for GBp-denominated ADRs — BTI shows EV $716.8B / 61.0x, DEO $917.0B / 148.3x. Currency-unit errors; do not quote those fields.- VZ postpaid phone net adds not verified — the single most important operating metric for that thesis.
- BDX trailing financials are pre-spin (Waters transaction closed 2026-02-09); TTM P/E, FCF and Graham IV are all stale.
- MPLX fee-based EBITDA % is company language only, no disclosed figure — the >85% test is presumed passed, not verified. Yahoo's EV/EBITDA (13.9x) disagrees with the Scout's calculation (~11.6x).
- EXC 8–10yr dividend CAGR uncomputable across the 2022 Constellation spin.
- All ROIC figures are Scout-derived (NOPAT ÷ invested capital), not pre-computed. roic.ai is on the free plan (2 years only) and could not cross-validate.
fin.pycarries only ~4 annual periods, so no §1-compliant 5–8yr FCF CAGR was computable for any candidate. This is the main reason no conviction here exceeds 7.5.- PPG: Yahoo returned null ROE/ROA/current-ratio; FCF series distorted by working-capital swings, so the +34.6% 3yr FCF CAGR is not a usable trend.
- KR (Kroger) failed to fetch (DNS error), not screened. IART returned incomplete data.
- claude.ai connectors Alpha Vantage, FactSet, S&P Global are unauthorized this session and could not be used for cross-validation.
Scouts: Healthcare · Energy · Industrials & Materials · Consumer Staples & Telecom. Quantitative data from .mcp/fin.py (Yahoo Finance) and Public.com live quotes, retrieved 2026-07-29.