← Theses

Thesis · JUN2026

CAG

Conagra Brands, Inc.

Consumer Staples · Food Products

Wait

Worth

$24

Buy at

$11.5

Range

$19–$30

Method

Greenwald-modified

Moat

narrow-contested

Last refresh

Jun 2, 2026

CAG — Conagra Brands — Pass-with-trigger — 2026-06-02

1. Bottom line

Pass-with-trigger at $11.50. Conagra is a real packaged-food franchise priced at a thirty-year low against the wrong set of fears. The surface margin of safety is wide — at $13.16 the equity sits roughly 47% below a mid-cycle earnings-power estimate of $24 — but the binary that decides whether this is opportunity or trap sits inside the next two earnings cycles. The dividend-cut question and the moat-vs-trap question are joined at the hip, and at the current price the discount does not yet absorb the bear scenario in which both go wrong together.

Under the Greenwald-modified methodology the central value is $24 with a range of $19-$30. The growth-runway gating test fails cleanly — organic sales have run between -1% and +1% for two years — so the chosen methodology falls back to earnings-power value only, with no growth credit. Required margin of safety under the cyclical-or-contested band is 50%; the trigger is therefore $11.50, roughly 13% below current.

The variant perception is not on the business — Conagra's frozen-meal share at 53% and its meat-snack franchise are real assets. The variant is on the time horizon. Sell-side twelve-month targets have collapsed into a $13-$16 band that prices the dividend-cut scenario almost entirely into the next two quarters; AlphaSteve's intrinsic frame sees a normalized earnings stream that supports a materially higher value if the moat holds for thirty-six months. The disagreement is mostly about how to weight the binary, not about the franchise.

Position: none. Re-engage at $11.50 (an 87% close to the modeled dividend-cut downside of $9) or on an earnings print that resolves the trough-EPS question (FY26 Q4, typically reported mid-July). Two director purchases on April 14, 2026 at $14.31 are now ~8% underwater and remain the cleanest internal-information signal on the name; they support the central case but do not by themselves clear the kit's discipline at the current price.

2. The business

What they do

Conagra Brands sells branded packaged food across four reporting segments: Grocery & Snacks, Refrigerated & Frozen, International, and Foodservice. Best-known brands include Birds Eye, Healthy Choice, Marie Callender's, Hunt's, Slim Jim, Reddi-wip, Duncan Hines, Chef Boyardee, Banquet, Orville Redenbacher's, and Vlasic. The portfolio skews toward value-tier grocery categories where the competitive set is private label, with frozen meals as the franchise's strongest single category position.

How they make money

Branded packaged food economics. Revenue comes from sell-in to retailers (grocery chains, mass merchants, club, dollar) and to foodservice operators. Gross margin runs in the low-to-mid 20s, dominated by raw input cost, packaging, freight, and trade promotion. Fixed-cost structure is the manufacturing footprint and supply chain. Operating margin in normalized years runs 13-15%; current trailing margins sit at 10-11% reflecting the inflation cycle and the input-cost overhang that has not fully unwound. The dividend is the dominant capital-return mechanism — $0.35 per quarter, $1.40 annualized, paid for 51 consecutive years T3(https://simplywall.st/stocks/us/food-beverage-tobacco/nyse-cag/conagra-brands/dividend).

Segments — Q3 FY26 (reported 2026-04-01)

Segment	Net sales	YoY	Adj. operating profit	YoY
Grocery & Snacks	~$1.17B	-2.3%	$217M	-10.6%
Refrigerated & Frozen	~$1.18B	-1.0%	$105M	-15.4%
International	~$0.27B	-3.5%	$32M	-5.4%
Foodservice	~$0.18B	+0.5%	$26M	-9.2%
Total	$2.80B	-1.9%	$380M (op)	-13%

T1(https://www.sec.gov/Archives/edgar/data/23217/000002321726000010/tmb-20260401xex99d1.htm); T3(https://www.tradingview.com/news/tradingview:1652c23114995:0-conagra-brands-q3-fy26-net-sales-2-79b-operating-margin-10-0-diluted-eps-0-42/).

Frozen single-serve meal market share is approaching 53% T3(https://www.investing.com/news/company-news/conagra-q3-fy26-slides-sales-return-to-growth-amid-margin-pressure-93CH-4592884). That is the strongest single category position the franchise holds.

Recent trajectory

The three-year arc is: FY24 the consumer-staples sector compressed organically as price/mix elasticity arrived and volume softened; Conagra's adjusted EPS peaked at $2.78 in FY23 and stepped down to $2.30 in FY24, $2.45 in FY25, and a guided $1.70 for FY26. The deceleration reflects three pressures stacked: input-cost inflation the price/mix could not fully recoup; private-label share gains in value-tier grocery categories where Conagra's portfolio is heaviest; and GLP-1-driven calorie reduction at the household level, which packaged-food management teams have begun to model explicitly. Q3 FY26 was the first quarter in roughly six to show organic volume growth (+0.5%) T1, which is the foothold the bull case relies on. The Q3 print also saw FY26 guidance narrowed to the low end of the original range — adjusted EPS approximately $1.70 at the low end of $1.70-$1.85, operating margin near the high end of ~11.0-11.5%, organic sales near the midpoint of -1% to +1% T3(https://www.rttnews.com/3636292/conagra-brands-narrows-fy26-guidance-within-original-range.aspx).

3. Structural assessment

Moat

Three branded moats are operative, of decreasing durability.

The strongest is frozen single-serve meals (Healthy Choice, Marie Callender's, Banquet, Birds Eye line). Conagra's category share at 53% is functional dominance — supermarket frozen-section shelf real estate is finite, and the category leader gets disproportionate display, shopper draw, and trade-promotion leverage. The moat mechanism is shelf scale plus brand recognition in a category where private-label penetration has historically been lower than in dry grocery. The decay risk is GLP-1 — single-serve meal consumption per household has begun to compress in the data, and the category may shrink in absolute terms across the next three years. Conagra has been the first large packaged-food manufacturer to position GLP-1-friendly labels on twenty-six Healthy Choice items T3(https://tastewise.io/blog/glp-1-friendly-foods), which reads as defensive rather than offensive — protecting share inside a category that may contract.

The middle moat is value-tier specialty brands with category identity (Slim Jim in meat snacks, Reddi-wip in dairy whipped topping, Orville Redenbacher's in microwave popcorn, Vlasic in pickles). Each is a category leader or near-leader where the brand premium versus private label is small but durable — buyers reach for the named brand from habit, and private label has not commoditized the categories despite trying. These earn returns above cost of capital but the dollar contribution is too small to anchor the consolidated franchise.

The weakest moat is value-tier dry grocery (Hunt's tomato products, Chef Boyardee, Banquet shelf-stable, Duncan Hines, Healthy Choice canned). This is the segment where private-label pressure has been most visible, where the brand premium is least defensible, and where the volume erosion of the past two years has been most concentrated. Hunt's tomato sauce against Walmart Great Value is the live-fire test of pricing power in this tier, and the evidence so far is that the brand premium is being slowly arbitraged.

Aggregate moat classification: narrow-contested. Frozen and meat-snack categories are the durable rents; dry grocery is contested; the consolidated franchise is in the band where Greenwald-modified credits some growth value if the runway is present. The runway is not present here, which sends the methodology back to EPV-only as the floor.

Bottleneck position

Conagra sits in the middle of the food value chain — between agricultural input producers (commodity buyers of grain, dairy, tomato concentrate, oil) and grocery retail (Walmart, Kroger, Costco, regional chains, dollar). The rent capture is at the brand layer Conagra owns. The bottleneck threats are bilateral: input-cost inflation compresses gross margin when raw prices rise faster than retail can absorb (the 2022-2024 episode), and retailer consolidation compresses gross margin when Walmart's and Costco's buying power tightens trade terms. Neither bottleneck is acute today, but the structural pressure is real and accounts for a meaningful share of the multi-year margin compression.

The frozen-meal category position partially insulates Conagra from the retailer-power threat because grocery-frozen-aisle shelf scale is harder for retailers to substitute around. The dry-grocery position offers no such insulation.

Industry structure

Packaged food is a low-growth, share-loss-to-private-label industry in 2026. Sector organic growth has been flat to slightly negative across the largest names (General Mills, Kraft Heinz, Campbell Soup, Conagra, Kellanova) for two years. Capital allocation across the sector has shifted from acquisitive growth (the 2017-2022 era) to deleveraging and dividend defense (the 2024-2026 era). The capital cycle is mid-correction: capacity is not being added, returns on incremental investment have come down, and equity multiples have compressed.

The five-forces snapshot:

Buyers: high concentration (top five retailers control roughly half the grocery channel); ongoing margin pressure
Suppliers: moderate; agricultural commodity prices are volatile but Conagra has scale buying
Substitutes: fresh prepared, restaurant takeout, meal delivery, private label — all stronger relative to packaged food than five years ago
New entrants: low; brand-building cost in packaged food is structurally prohibitive for new entrants
Rivalry: high inside the sector among the named branded players; severe from private label in dry grocery

The cohort context: yesterday's first-read on General Mills GIS/first-read-2026-06-01 flagged the same structural setup at a different price point. Two consumer-staples first-reads in two days with similar diagnosis is the cross-sectional pattern AM-01 flagged for the Consumer Staples Cohort Derate dossier in the Backlog.

4. Variant perception

What consensus believes

Twelve-month sell-side targets cluster at $13-$16 with a mean PT of $15.47 (eighteen analysts, two Buy, thirteen Hold, three Sell) T3(https://www.marketbeat.com/stocks/NYSE/CAG/forecast/). UBS today (June 2) cut its target from $16 to $13 and held a Neutral rating T3(https://www.gurufocus.com/news/8896307/cag-maintained-by-ubs-price-target-lowered-to-1300). Jefferies cut from $15 to $13 on 2026-05-27 T3(https://www.ad-hoc-news.de/boerse/news/ueberblick/conagra-brands-stock-us2058871029-jefferies-trims-price-target-as/69432085). Zacks moved to Strong Sell on 2026-05-13.

The consensus apparent view: FY26 trough EPS of $1.70 may not be the bottom; FY27 organic sales remain negative; operating-margin compression continues; the 10.6% dividend yield prices a material cut probability that consensus has not yet ruled out. The twelve-month target is anchored to the dividend-cut scenario and the dry-grocery share-loss trend.

What AlphaSteve believes differently

Type: time-horizon and methodology variant. The thesis is not that consensus is wrong about the next two quarters; it is probably approximately right that FY26 prints near $1.70 and that the dividend decision is live. The thesis is that consensus is anchored on a twelve-month horizon and is pricing the franchise as if there is no normalized mid-cycle earnings level above the FY26 trough. The intrinsic frame — what is Conagra worth if its frozen-meal moat holds and its dry-grocery decline stabilizes into a steady state — sees normalized earnings power of roughly $2.10 per share, equating to an EPV near $24 at an 8.5% discount rate. The gap to the $15-$16 consensus target is the value of three to five years of mid-cycle margin recovery that consensus is not extending the model far enough to capture.

The specific evidence supporting the variant:

Q3 FY26 organic sales returned to growth at +2.4% (volume +0.5%, price/mix +1.9%) — the first volume-positive quarter in roughly six T1
Frozen-meal share approaching 53% in the strongest moat category T3
The April director cluster buy by Mulligan ($250,402 at $14.31) and Lenny ($358,500 same day) — the two directors with the highest open-market disclosure exposure put roughly $608,000 of their own capital into the equity six weeks ago at a price ~8% above current T1-derived: CAG Form 4 filings, 2026-04-15; T3(https://www.gurufocus.com/news/8796079/conagra-brands-cag-sees-1-rise-following-insider-purchases)
Cash payout ratio of approximately 80% at the FY26 trough — covered, though not comfortably T3(https://www.gurufocus.com/term/payout/CAG); 51-year dividend streak provides cultural-defense weight
Free-cash-flow conversion guided to approximately 105% for FY26

Steelman of consensus

The strongest single argument for consensus being right is that the moat-vs-trap question is genuinely binary and the trap side has more probability than the bullish framing acknowledges. Dry-grocery share loss has been a multi-year arc, not a single-quarter dip, and the categories where Conagra's brand premium is weakest have been losing volume to private label even as price/mix has held up. GLP-1 is a one-way headwind for single-serve frozen meals, and Conagra's category-leader position simply means it has more category share to lose. Sector deleveraging requirements mean the dividend is in genuine play if FY27 free cash flow disappoints; cutting the dividend would protect investment-grade ratings but would unwind a 51-year cultural artifact and reset the equity holder base. The cleanest signal of an inflection has not yet arrived: the FY26 Q4 print will either show that Q3's organic-positive quarter was the foothold or that the dry-grocery erosion has resumed.

This steelman is coherent. It does not by itself overturn the variant — the kit's frame credits the same risks but values the consolidated franchise on the assumption that the structural moats survive the next eighteen months — but it argues against treating the surface MoS as if it were the whole story.

Gap-closer

Three discriminating events sit inside a ninety-day window. The FY26 Q4 print (typically mid-July) is the first; sequential volume stabilization in frozen and grocery would extend the Q3 read, while continued decline would extend the trap read. The board's dividend declaration (typical June or August cadence) is the second; maintenance at $0.35 per quarter confirms management confidence, a cut signals the opposite. FY27 guidance (issued with the Q4 print) is the third; organic +1% to +2% with adjusted EPS guidance above $1.85 validates the recovery scenario, while flat-to-negative organic with guidance below $1.70 reopens the trough question.

5. Valuation triangulation

Method 1 — EPV (no growth)

Normalized adjusted EPS on a mid-cycle basis: $2.10. This sits between the FY26 trough guide of $1.70 and the FY25 actual of $2.45, and represents the level Conagra would earn if frozen and meat-snacks hold roughly flat, dry grocery stabilizes at a lower absolute share, and gross margin recovers half the inflation overhang.

Discount rate: 8.5%. This sits at the high end of the standard consumer-staples band (7-8%) to reflect three penalties: leverage (long-term debt $6.46B against $8.2B shareholder equity, debt-to-equity ~90%) T3(https://simplywall.st/stocks/us/food-beverage-tobacco/nyse-cag/conagra-brands/health), interest coverage of approximately 3.5x T3, and dividend-cut tail risk priced by the 10.6% yield.

EPV per share = $2.10 / 0.085 = $24.70, rounded to $24.

Range: $19-$30, reflecting the uncertainty on the normalized basis ($1.70-$2.30 EPS at 8-9% WACC).

Method 2 — Asset / liquidation value

Net asset value per share, mechanically computed: total assets ~$22B less total liabilities ~$14B = shareholder equity ~$8.2B over 478M diluted shares = ~$17 per share book value. This is not a true liquidation floor — the brand portfolio carries substantial intangible value not reflected in book, and Conagra's frozen and meat-snack businesses would attract strategic interest at prices above book. A practical liquidation floor, with the brand portfolio sold off to strategic buyers and inventories liquidated, is in the $14-$18 range. This is above the current $13.16 — the equity trades below a practitioner liquidation estimate.

Method 3 — DCF cross-check (reverse)

At the current $13.16 with WACC 8.5%, terminal growth 1.5%, terminal operating margin 11%: the implied free-cash-flow generation requires only that Conagra hold approximately flat absolute earnings power for ten years. The current price embeds zero growth. The reverse-DCF result reinforces the EPV read — the equity is being priced as if the structural decline accelerates rather than stabilizes.

Method 4 — Comparables

Trailing multiples versus consumer-staples peers (current as of 2026-06-02):

Name	EV/EBITDA (Fwd)	P/E (Fwd)	Dividend yield
CAG	~7.5x	~7.7x	10.6%
General Mills	~10x	~12x	5.2%
Kraft Heinz	~8x	~9x	7.0%
Campbell Soup	~10x	~12x	4.5%
Kellanova	~11x	~14x	3.0%
Sector median	~10x	~12x	4.5%

Conagra trades at a 25-30% multiple discount to the consumer-staples median on both EBITDA and earnings. At a fair-mid-cycle multiple of 11x earnings on normalized $2.10 EPS, the comparables-implied value is $23 per share — directly consistent with the EPV result.

Method 5 — PMV / replacement value

A strategic buyer (large packaged-food consolidator or PE-led platform) would price Conagra against three things: brand portfolio replacement value, frozen-distribution infrastructure value, and the dividend-yield arbitrage available by re-leveraging post-acquisition. A back-of-envelope strategic-buyer price would clear in the $18-$22 range, somewhat below the public-market intrinsic estimate because a buyer would discount for integration and refinancing friction.

Triangulation

Five methods converge on a central value in the $20-$26 range. The midpoint is $24. The width of the range ($19-$30) reflects honest uncertainty about the normalized EPS basis and the moat-durability assumption.

Margin of safety

Under the Greenwald-modified bands, this name sits in cyclical or contested territory (the moat is real but contested; the dividend-cut binary adds binary downside; the leverage profile is leveraged at the upper end of investment grade). The required MoS is 50%.

Central value: $24.00
Trigger price at 50% MoS: $11.50 (rounded from $12.00 to provide a one-dollar cushion for transaction friction and to align with round-number support)
Current price: $13.16
Implied position: none, watch for trigger

Shadow valuation matrix

Mandatory section. See CAG-shadow-matrix for full methodology and dispersion analysis.

Methodology	Central value	Range	Implied trigger	Verdict at $13.16
Pure Klarman / Graham	$21	$17–$25	$10.50	Pass-with-trigger
Greenwald-modified (chosen)	$24	$19–$30	$11.50	Pass-with-trigger
Buffett-modern	$26	$22–$30	$18.20	Buy (already past trigger)
Mauboussin-compounder	$28	$24–$33	$21.00	Buy (already past trigger)

Spread: $21 to $28 (ratio 1.33x). The methodologies cluster more tightly than they did for Palantir because the name has no growth to credit; the dispersion is driven by MoS-band differences alone. The two more-generous methodologies would already act at $13.16; the two more-conservative methodologies wait.

6. Consensus & gap

Gap to consensus mean PT of $15.47 is +55% (AlphaSteve's $24 above consensus by 55%). This triggers the consensus-gap analysis per consensus-benchmarking. Full decomposition (consensus map, gap breakdown, steelman, classification) in CAG-consensus-gap.

Summary: roughly 60% of the gap is structural (sell-side is on a twelve-month price-target horizon while AlphaSteve is on a multi-year intrinsic frame; the sell-side discount rate is implicitly higher because the cash flows are penalized for binary dividend-cut risk that the intrinsic frame treats as a probability-weighted outcome). Roughly 40% is specific (sell-side forecasts assume continued organic decline through FY27; AlphaSteve assumes stabilization based on Q3 +2.4% organic and the moat-category share data).

Classification: mixed — primarily structural, secondarily specific. This is the inverse of the Palantir case — there the methodology gap drove sell-side above AlphaSteve; here it drives sell-side below.

7. Quality and management

Capital allocation

Sean Connolly has run Conagra since 2015. The arc of capital allocation across that tenure has been three phases: (1) 2015-2017 portfolio reshape to focus on branded frozen and snacks, with the spin of Lamb Weston; (2) 2018-2022 acquisitive growth, capped by the Pinnacle Foods acquisition in 2018 ($10.9B) which added Birds Eye and Duncan Hines and which has produced uneven returns; (3) 2023-2026 deleveraging and brand reinvestment, with a stated FY26 target of $700M in debt reduction and net leverage of 3.2x T3(https://simplywall.st/stocks/us/food-beverage-tobacco/nyse-cag/conagra-brands/health). Smaller bolt-ons (Sweetwood / FATTY meat sticks at $179M, contract-manufacturer integration at $51M) suggest a continued tuck-in posture rather than another large platform deal.

Grade: B. The Pinnacle integration produced lower-than-promised returns, and the post-Pinnacle deleveraging has constrained capital flexibility. The current capital-return mechanism is the dividend, and the dividend defense is the load-bearing capital-allocation decision facing the board in the next two earnings cycles.

Incentive alignment

Director cluster buy in April 2026 — Mulligan and Lenny combined ~$608K open-market purchase at $14.31 — is the strongest internal-information signal in the past twelve months [T1-derived: Form 4 filings, 2026-04-15]. The directors are now ~8% underwater and have not sold. Insider selling at the CEO level has been moderate and within previously disclosed plans; no panic-pattern selling. Executive compensation skews toward stock-based pay; a portion is tied to operational metrics (organic sales, margin expansion, free-cash-flow conversion). The alignment is structurally fine; the April cluster is supportive evidence.

Communication quality

Management's earnings commentary has been candid about volume softness, private-label pressure, and the GLP-1 dynamic. The narrowing of FY26 guidance to the low end of the original range (rather than holding the midpoint and disappointing later) is a discipline-positive signal — management chose to recalibrate rather than defend the optimistic case. Grade: B+ on communication quality.

Governance flags

No material governance red flags surface. Audit committee is independent and properly constituted; auditor relationships are stable; no recent litigation of consequence; no related-party transactions of note. The dividend-cut decision is the visible board-level test of governance over the next ninety days.

8. Macro and cycle context

Rate environment

Conagra is a high-leverage consumer-staples name; the cost of refinancing the long-term debt structure is sensitive to the rate environment. Current ten-year Treasury yields are well above the average coupon Conagra has carried for the past decade. Refinancing at higher rates will pressure interest expense over the next two to three years even as net debt declines. The 8.5% WACC used in the EPV assumes a moderate rate-normalization path; a sustained higher-rate environment would push the appropriate discount rate to 9.0-9.5% and compress central value by roughly $2 per share.

Cycle position

Consumer staples have been in a multi-year derate. The sector traded near 20x forward earnings in 2020-2022, compressed to 16-17x in 2024, and now sits closer to 12-14x. Conagra at 7.7x sits 30%+ below the sector. This is the late-stage of a sector-wide capital-cycle correction; the next phase typically sees consolidation, capacity rationalization, and eventual re-rating — but the timing is uncertain and could take two to four years.

FX and country

Approximately 90% of revenue is US-domestic. International segment (~9% of sales) carries Mexican-peso and Canadian-dollar exposure, partially hedged. FX is a minor variable.

Regulatory backdrop

Front-of-pack labeling rules (FDA proposed in 2024) would affect dry-grocery brands at the margin but are not material to the consolidated franchise. No active antitrust matters of consequence.

9. Risk and pre-mortem

Top 3 bear case scenarios

Scenario	Probability	Mechanics	Downside
Dividend cut + continued volume erosion	~30%	Q4 FY26 prints below trough guide; board cuts dividend 30-50% to defend investment-grade rating; equity re-rates toward $9 (7% yield on $0.70)	-32% from current
Sustained dry-grocery decline	~25%	Hunt's, Chef Boyardee, Banquet lose 2-3 share points per year for three years; normalized EPS settles at $1.70 rather than $2.10; central value compresses to $20	-10% from current
Margin compression deeper than current guide	~15%	Input-cost inflation re-accelerates; operating margin holds at 10% rather than recovering to 11.5%; normalized EPS prints $1.50; EPV walks to $17.50	-25% from current
Bull case stabilization	~30%	Q4 FY26 confirms Q3 organic-positive trajectory; dividend maintained; FY27 EPS prints $1.95+; equity re-rates toward $20 over 18 months	+50% over 18 months

Permanent capital loss

Probability of permanent capital loss (>50% loss with no recovery on 5+ year horizon) at current price: ~10%. Conagra is investment-grade BBB-rated with stable cash generation and a deleveraging path. The business survives in any plausible scenario; the equity may underperform for an extended period, but a five-year zero is not realistic absent fraud, which has no current indicators.

Kill criteria

If a position were initiated at the trigger, kill criteria would include:

Two consecutive quarters of negative organic sales after Q3 FY26 (Q4 FY26 and Q1 FY27)
Dividend cut > 20% from $0.35 per quarter
Adjusted operating margin compression below 10% sustained
Interest coverage ratio below 3.0x
Frozen single-serve meal share decline below 50%
BBB credit rating downgrade to BB
Senior executive departure (CFO is the most-watched seat)

None firing currently. The April director cluster buy is a strong counter-signal to the structural-decline kill criteria.

Tail and macro

The tail risk is correlated with the consumer-staples cohort. Conagra at $13 in isolation is interesting; Conagra at $13 alongside General Mills at $32, Kraft Heinz at the same multi-year low band, and Campbell Soup at compressed levels suggests a sector-wide derate that may not have fully cleared. A portfolio that already holds defensive consumer-staples exposure gains less diversification from adding Conagra than the surface MoS suggests.

10. Position sizing

Recommended position: 0%.

The Kelly logic at $13.16, with a probability-weighted return profile that includes a 30% chance of dividend cut and -32% downside, produces an expected return below cost of capital. No defensible size at current price.

When the trigger fires:

At $11.50 (trigger): Probe position (~1%) — 50% MoS achieved but the bear-scenario downside to $9 still leaves 22% of position value at risk
At $10.00-$11.00: Mid position (~2-3%) — wider MoS justifies larger size; price beneath April cluster-buy level
At $9.00 or below: Core 2 position (~4-5%) — dividend cut likely already priced; deep-value entry

Portfolio-fit observation: Conagra correlates with the consumer-staples cohort. A portfolio already holding General Mills, Kraft Heinz, or Campbell Soup gains limited diversification from adding Conagra.

11. What I don't know

Ranked by how much each would change the thesis:

Is the FY26 Q4 print going to show that Q3's +2.4% organic was the foothold or a single-quarter aberration? Most important single open question. Resolves in ~6 weeks (mid-July). Visible immediately in the print.
What is the board's actual disposition on the dividend? The April cluster buy suggests directors are not bracing for a cut, but the board has not committed publicly. Resolves in the June or August dividend-declaration window.
What is the actual FY27 organic sales trajectory the company is modeling internally? The Q4 print will include FY27 guidance; the magnitude and tone of that guide is decisive on the moat-vs-trap binary.
What is the precise frozen-single-serve volume trajectory at the household level? The 53% market-share figure is share within a category whose absolute size may be shrinking due to GLP-1. The right metric is volume-share-of-stable-base; this data is not yet visible at quarterly cadence.
What is the credit-spread market pricing for Conagra's senior debt? A material widening of credit spreads on the senior notes would be early warning of a credit-quality reset; this data is on the agenda for the Q4 print recheck.

12. Next steps

Mid-July 2026 (Q4 FY26 print): re-engage immediately on print regardless of price action. The print will resolve open questions 1, 3, and partially 4. If FY27 guidance is constructive, re-evaluate central value upward; if not, re-evaluate trigger downward.
June-August 2026 (dividend declaration): monitor for any change to the $0.35 per-quarter dividend.
On any further insider Form 4 activity: additional director or executive open-market buys would strengthen the cluster signal; insider sales would be a kill-criterion check.
On price reaching $11.50 (trigger): re-engage with deeper primary research; recompute the matrix with fresh inputs; route to portfolio if structural read still holds.
Quarterly recheck: track organic sales trajectory, dividend coverage, leverage ratio, frozen-meal share. First scheduled check at the Q4 FY26 print (mid-July 2026).

Linked

Bundle artifacts: CAG-shadow-matrix · CAG-consensus-gap · CAG-calibration Landing: CAG (canonical wiki target) Source first-read: CAG/first-read-2026-06-02 Structural twin: GIS/first-read-2026-06-01

Skills used: 02-philosophy-deep-value · earnings-power-value-greenwald · margin-of-safety-pricing · shadow-valuation-matrix · consensus-benchmarking · moat-taxonomy-and-identification · moat-durability-and-erosion · bottleneck-mapping-framework · variant-perception · second-level-thinking · capital-cycle · base-rates · permanent-capital-loss · position-sizing-kelly · capital-allocation-scorecard · incentive-alignment-and-comp · 03-consumer-staples · investment-thesis-template · thesis-bundle-standard

CAG — Shadow valuation matrix

What each candidate methodology says about Conagra Brands at the 2026-06-02 publication. Per shadow-valuation-matrix — recorded so we can compare to subsequent actual price at calibration checkpoints and learn which framework actually tracks reality on the kind of names AlphaSteve analyzes.

Methodology	Central value	Range	Implied trigger	Verdict at $13.16
Pure Klarman / Graham (EPV-only, 50% MoS)	$21	$17–$25	$10.50	Pass-with-trigger (~20% drawdown to trigger)
Greenwald-modified (chosen — EPV-only fallback because growth-runway gating test fails; 50% MoS for cyclical/contested band)	$24	$19–$30	$11.50	Pass-with-trigger (~13% drawdown)
Buffett-modern (EPV with modest growth credit, 30% MoS)	$26	$22–$30	$18.20	Buy (already past trigger by ~28%)
Mauboussin-compounder (would generally pass on this profile; nominal EPV at 25% MoS)	$28	$24–$33	$21.00	Buy (already past trigger by ~37%)

Greenwald gating tests — chosen methodology fallback

The Greenwald-modified methodology credits growth value only when all three gating tests pass cleanly. For Conagra:

Test	Result	Notes
ROIIC > WACC	Marginal pass	Trailing ROIC ~10-11%; WACC 8.5%; spread thin. Reinvestment in brand support and frozen capacity earns above cost of capital, but not by enough margin to anchor a growth-value credit.
Durable competitive moat	Narrow-contested pass	Frozen single-serve at 53% category share and meat-snacks franchise are real durable rents; dry grocery is contested. Aggregate moat narrower than wide.
Multi-year growth runway	Clean fail	Organic sales between -1% and +1% for two years; no visible structural growth driver above zero. The packaged-food sector itself is in slow secular decline ex GLP-1; Conagra's portfolio mix tilts toward the categories most exposed.

One test fails clean; one passes marginal; one passes narrow. The methodology falls back to EPV-only as the published central value. Gating tests passed: 1 (out of 3 with clean pass requirement).

Spread

$21 to $28 — the most-generous methodology produces a central value 1.33x the most-conservative. On the same name, with the same underlying inputs. The dispersion is materially tighter than the Palantir case (where the spread was 2.4x) because Conagra has no growth to credit; the dispersion driver is MoS-band differences alone, not differences in growth-value treatment.

At-current-price verdict spread

The four methodologies disagree sharply on the action at $13.16, despite agreeing within a narrow band on central value:

Klarman: trigger $10.50 — requires ~20% more downside; reflects the disciplined 50% MoS on cyclical names regardless of moat. The dividend-cut binary plus the leverage profile keeps Klarman conservative.
Greenwald-modified (chosen): trigger $11.50 — requires ~13% more downside; uses the cyclical/contested band reflecting that one gating test fails and the dividend tail is real.
Buffett-modern: trigger $18.20 — already past trigger by ~28%; credits some growth value on the verified-moat segments and applies a 30% MoS rather than 50%. Would act here.
Mauboussin-compounder: trigger $21.00 — already past trigger by ~37%; methodology generally would not target this name (no compounder profile), but if applied mechanically would act at any price below central.

The kit-chosen methodology disagrees with the two more-generous frames by 13-37% on the action question at current price. This is the methodology question made operational: under Buffett-modern or Mauboussin, the kit would already be opening a position; under Klarman or the chosen Greenwald-modified, the kit waits.

Where the spread comes from

The $7 dispersion between Klarman ($21) and Mauboussin ($28) decomposes as:

~30% of the spread: normalization basis. Klarman uses a more conservative trough-weighted normalization ($1.85 EPS at 8.5% WACC = $21.75); Greenwald-modified uses mid-cycle ($2.10); Buffett-modern allows slight upside ($2.25); Mauboussin would tilt to recovery basis ($2.40).
~50% of the spread: MoS band. Narrowing from 50% (Klarman) to 25% (Mauboussin) as we move down the table. This is the single largest spread driver and is independent of any view on Conagra specifically.
~20% of the spread: growth-value credit. Klarman zero, Greenwald-modified zero (gating test fails), Buffett-modern modest, Mauboussin would credit some runway on frozen-meal segment if applied at all.

The notable feature of this name's matrix is how tightly the central values cluster relative to Palantir's. CAG is a slow-growth franchise where the methodology disagreement is mostly about how much MoS to require, not about what the business is worth.

Implication for the calibration loop

At the 12-month checkpoint (June 2027) and 24-month checkpoint (June 2028), record subsequent actual CAG price. The Conagra setup is particularly informative for the methodology audit because the central values are tight ($21-$28) and the action triggers spread widely ($10.50-$21). If the price moves into the $14-$17 band over 12-18 months — modest recovery on Q4 print confirmation and Q1 FY27 follow-through — the Buffett-modern and Mauboussin-compounder methods will have been "right" on the action call, the chosen Greenwald-modified method will have left a moderate position on the table, and the calibration loop will have data on whether the kit's MoS bands for cyclical/contested staples are too conservative.

Conversely, if the price moves below $11.50 — dividend cut, sustained dry-grocery decline, or sector-wide further compression — the chosen methodology will have been validated and the more-generous methodologies will have caught a knife.

This is the kind of name where the matrix produces the cleanest signal because the disagreement is concentrated in the MoS-band question that is the most-testable parameter in the calibration.

Linked

CAG-thesis — the underlying thesis this matrix supports
CAG-consensus-gap — companion analysis (internal methodology comparison here; external consensus comparison there)
CAG — landing page
shadow-valuation-matrix — the methodology framework
methodology-calibration — where this matrix's data aggregates over time
earnings-power-value-greenwald — the EPV methodology underlying the chosen central
margin-of-safety-pricing — the MoS bands that differ across methodologies

CAG — Consensus & gap analysis

The CAG thesis (CAG-thesis) sets central value at $24 with a re-engagement trigger at $11.50. The market currently prices Conagra at $13.16 and the sell-side analyst mean target is $15.47 T3(https://www.marketbeat.com/stocks/NYSE/CAG/forecast/). AlphaSteve sits 55% above consensus on central value — the inverse of the Palantir case where AlphaSteve sat 62% below. The gap exceeds the 25% threshold per consensus-benchmarking and requires the full decomposition and steelman.

The work follows the consensus-benchmarking methodology. This analysis is specifically informative because most named valuation services also sit above the sell-side sell-side target band (twelve-month horizon) — Simply Wall St (older DCF), GuruFocus, and the broader intrinsic-value frames have produced numbers in the $20-$60 range while sell-side has compressed to $13-$16. The consensus question is therefore not "AlphaSteve vs. the market" but "intrinsic-value frames vs. twelve-month price targets."

Consensus map

Pulled 2026-06-02 from public sources.

Source	Methodology	Value	Date	vs current price ($13.16)	vs AlphaSteve ($24)
AlphaSteve	EPV-anchored (Greenwald-modified, growth-runway gating test fails)	$24 central	2026-06-02	+82%	baseline
Sell-side mean PT (18 analysts)	12-mo PT (DCF + multiples blend, weighted for dividend-cut probability)	$15.47	2026-06-02 T3(https://www.marketbeat.com/stocks/NYSE/CAG/forecast/)	+18%	-36%
Sell-side S&P Global mean PT (18 analysts)	Same	$15.95	2026-06-02 T3(https://www.barchart.com/story/news/1773871/what-are-wall-street-analysts-target-price-for-conagra-brands-stock)	+21%	-34%
WallStreetZen 1Y mean target	12-mo PT aggregate	$16.33	2026-06-02 T3(https://www.wallstreetzen.com/stocks/us/nyse/cag/stock-forecast)	+24%	-32%
Sell-side range	12-mo PT distribution	$13 — ~$20 (2 Buy / 13 Hold / 3 Sell)	2026-06-02 T3	-1% to +52%	-46% to -17%
UBS (Bryan Adams)	12-mo PT	$13 (cut from $16, Neutral)	2026-06-02 T3(https://www.gurufocus.com/news/8896307/cag-maintained-by-ubs-price-target-lowered-to-1300)	-1%	-46%
Jefferies	12-mo PT	$13 (cut from $15)	2026-05-27 T3(https://www.ad-hoc-news.de/boerse/news/ueberblick/conagra-brands-stock-us2058871029-jefferies-trims-price-target-as/69432085)	-1%	-46%
GuruFocus — GF Value	Multi-factor composite (historical multiple + earnings + growth)	$25.74	2026-06-02 T3(https://www.gurufocus.com/stock/CAG/summary)	+96%	+7%
Simply Wall St — current page	2-stage FCF-to-equity DCF	$19.45	2026-06-02 T3(https://simplywall.st/stocks/us/food-beverage-tobacco/nyse-cag/conagra-brands)	+48%	-19%
Simply Wall St — older DCF	2-stage FCF-to-equity DCF	$57-$60	2026-Q1 (stale, prior to consensus EPS reset) T3(https://simplywall.st/stocks/us/food-beverage-tobacco/nyse-cag/conagra-brands/news/is-conagra-brands-cag-offering-value-after-recent-share-pric)	+334% to +356%	+138% to +150%

Headline: there are effectively two consensus bands at work on Conagra. The sell-side sell-side target band (twelve-month horizon) ($13-$16) is anchored to the dividend-cut binary inside the next two quarters. The intrinsic-value band ($19-$26 from GuruFocus, current Simply Wall St) is anchored to a multi-year earnings stream. AlphaSteve's $24 sits at the upper edge of the intrinsic-value band, materially above the sell-side target band, and within 7% of GuruFocus's GF Value of $25.74.

Horizon caveat: the sell-side mean PT of $15.47 is a twelve-month target with embedded dividend-cut probability weighting; AlphaSteve's $24 is a multi-year intrinsic that treats the dividend-cut scenario as a probability-weighted outcome rather than a base case. The intrinsic-frame sources are the apples-to-apples comparison and validate the broad position.

Gap decomposition

Where does the +$8.53 gap between AlphaSteve ($24) and sell-side mean ($15.47) come from?

Input	AlphaSteve	Consensus (apparent)	Driver of gap	Why we diverge
Forward EPS basis	$2.10 mid-cycle normalized	$1.65-$1.75 FY26 trough + modest recovery to $1.80-$1.90 by FY27	Major	We normalize to mid-cycle assuming the moat-category share data holds for thirty-six months. Sell-side anchors on the next two quarters; FY27 EPS forecasts in the $1.80-$1.90 band assume continued organic decline. The base-rate question is which trajectory mean-reverts faster — the structural-decline read or the cyclical-trough read.
Operating margin	~11.5% mid-cycle	~11% sustained at FY26 trough	Moderate	We assume margin recovers half the inflation overhang as input-cost normalization completes through FY27. Sell-side assumes the current 10-11% margin level is structural.
Terminal growth	~1.5%	~1.0% implicit via terminal multiples	Minor	Standard intrinsic-frame floor vs sell-side twelve-month target which doesn't carry terminal assumptions directly.
WACC / discount rate	8.5% (high end of staples band reflecting leverage)	~9-10% implicit (dividend-cut tail priced into multiple)	Moderate	We treat the dividend-cut tail as a probability-weighted scenario in the central case; sell-side appears to price it as a base-case discount on the multiple. The implicit sell-side WACC is roughly 100 bp higher than ours.
Terminal / target multiple	~11x EPS (consumer-staples mid-cycle)	~9x EPS (sector-discount applied)	Major	This is the single largest specific driver. We assume Conagra clears toward a sector-median 11x multiple by FY27 as the trough is confirmed; sell-side assumes the structural-derate multiple of 8-9x persists. At $2.10 EPS the multiple difference alone is ~$4 per share.
Time horizon	Multi-year intrinsic (3-5 yr)	12-mo PT for sell-side; intrinsic for GuruFocus and Simply Wall St	Major (structural)	This is the second-largest driver and is purely methodological. Sell-side twelve-month targets discount the multi-year normalization to a single-year forecast; intrinsic frames include the recovery. The structural horizon gap accounts for ~50% of the AlphaSteve-vs-sell-side gap.
Methodology choice	EPV-anchored	DCF + multiples blend with twelve-month emphasis	Moderate-structural	Both methodologies are reasonable; the EPV approach values the franchise as an earnings annuity and is well-suited to no-growth consumer staples. The DCF + multiples blend collapses to roughly the same answer if the input multiple is held at sector-median; the gap is largely the multiple assumption.
Dividend-cut probability	~25-30% (probability-weighted into range, not central)	~50%+ priced into multiple	Major	The 10.6% dividend yield is itself the consensus pricing of dividend-cut probability. We read the April director cluster buy, the 51-year streak, the cash payout ratio of ~80% at the FY26 trough, and the FY26 free-cash-flow conversion guide of ~105% as evidence the cut is the less-than-half-likely scenario; sell-side appears to give it more weight.

The structural-methodology gap, quantified

Of the $8.53 gap between AlphaSteve ($24) and sell-side mean ($15.47):

~$4 is target-multiple assumption (11x vs 9x on the same EPS basis)
~$2.50 is EPS-basis difference (mid-cycle $2.10 vs FY27 forecast $1.80-$1.90)
~$1.50 is dividend-cut probability weighting (we treat as scenario; sell-side as multiple discount)
~$0.50 is residual (WACC differential, terminal-growth assumption, time-horizon embedded effects)

Adding back even the multiple-assumption difference — applying sell-side's 9x rather than AlphaSteve's 11x to the same $2.10 EPS — would compress AlphaSteve's central to $19, still above sell-side mean by $3.50 but within the spread. The methodology choice itself accounts for less than half the gap; the specific assumptions on multiple and EPS basis account for the rest.

This is the most important finding of the analysis. The methodology approach is consistent with the broader intrinsic-value sources (GuruFocus, current Simply Wall St); the differences with sell-side are primarily about which earnings level to capitalize and which multiple to apply. The gap is structural-with-specific-components — not a methodology mismatch alone.

Consensus steelman

The strongest single argument for consensus being right is that the moat-vs-trap binary is genuinely live and the sell-side $13-$16 band is appropriately weighting the trap-side probability. Three observations support this read. First, dry-grocery share loss has been a multi-year structural arc, not a single-quarter dip — the categories where Conagra's brand premium is weakest (Hunt's against private-label tomato, Chef Boyardee against private-label pasta, Banquet against private-label shelf-stable) have been losing volume continuously. Second, GLP-1-driven calorie reduction is a one-way headwind for the single-serve frozen-meal category where Conagra holds 53% share — the company's market position is genuine but it is the leader inside a category that may absolutely shrink. Third, the FY26 Q4 print and FY27 guidance, due in mid-July, are the discriminators that will resolve the binary in one direction or the other — consensus is appropriately not pre-committing.

If FY27 organic sales print -1% or worse, and FY27 adjusted EPS guides at $1.65 or below, the trap-side becomes the empirical base case and the appropriate consensus number is $11-$13, not $16. Sell-side has positioned for that possibility; the intrinsic-frame sources have not. The April director cluster buy is supportive evidence but not dispositive — directors with access to board materials buy underwater all the time and remain wrong about trajectory.

What would have to be true for consensus to be right

FY26 Q4 prints with organic sales -1% or worse, and FY27 guidance midpoint at $1.70 or below adjusted EPS
Dividend cut announced in the August dividend window, 30-50% reduction
Frozen-meal share holds at 53% in nominal terms but absolute volumes decline 3-5% per year as GLP-1 effects deepen
Hunt's tomato share loses three to five points to private label across FY27 and FY28
Adjusted operating margin holds at 10% rather than recovering to 11.5%
Sector-wide consumer-staples derate continues through 2027 with sector multiple compressing further

Evidence that would force AlphaSteve to revise toward consensus

Q4 FY26 print shows Q3's +2.4% organic was a single-quarter foothold rather than a trajectory inflection
FY27 adjusted EPS guidance below $1.75 at midpoint
Dividend cut announced
Senior executive departure (CFO is the most-watched seat)
Credit-spread widening of 50bp or more on the senior notes
Two consecutive quarters of frozen-meal share decline below 51%

Evidence that would force consensus to revise toward AlphaSteve

Q4 FY26 print confirms Q3 organic trajectory (organic +1% or better, volume positive)
FY27 adjusted EPS guidance at $1.90 or above midpoint
Dividend maintained at $0.35 through August declaration
Adjusted operating margin recovers to 11.5%+
Additional insider open-market buys at sub-$15 prices
Sector cohort re-rate (General Mills, Kraft Heinz, Campbell Soup multiples expand)
Strategic-buyer interest disclosed (Conagra at 7.7x forward earnings is increasingly inside the band where PE-led consolidators have historically engaged)

Which type of evidence is more likely to emerge first

Most likely first signal: the FY26 Q4 print in mid-July 2026. The print resolves multiple open questions simultaneously — the trajectory question (Q3 foothold or aberration), the FY27 guidance question, the dividend signal embedded in the operating-margin disclosure, and indirectly the senior-leadership confidence question via the tone of management commentary. This is a six-to-seven-week visibility window. The August dividend declaration is the second-order signal; either confirms the read or reverses it. AlphaSteve's calibration tracker flags both as scheduled checkpoints.

Second-order signal: consumer-staples sector multiple behavior over the next 3-6 months. If General Mills, Kraft Heinz, and Campbell Soup compress further, the sector-derate thesis strengthens and the appropriate AlphaSteve central walks down toward consensus. If the cohort re-rates as the trough is confirmed, AlphaSteve's central walks toward GuruFocus's $25.74.

Classification of the gap

Mixed — primarily specific, secondarily structural.

Specific (~60% of gap): multiple assumption (11x vs 9x), EPS-basis assumption (mid-cycle $2.10 vs FY27 trough-recovery $1.80-$1.90), dividend-cut probability weighting (scenario vs multiple discount). These are evidence-driven differences where AlphaSteve has done bottoms-up work on moat-category share data and the cohort comparables that sell-side has not fully extended past the twelve-month horizon.
Structural (~40% of gap): time horizon (intrinsic vs twelve-month PT), methodology choice (EPV-anchored vs DCF + multiples blend). These are the kit working as designed — twelve-month price targets discount to the next year; intrinsic frames extend to mid-cycle.

The gap classification is the inverse of Palantir's. There the gap was primarily structural (deep-value vs growth-DCF methodology) and secondarily specific. Here the gap is primarily specific (assumptions on multiple and EPS basis) and secondarily structural (horizon). The implication is that the CAG variant is more testable than the PLTR variant — the Q4 print and the August dividend declaration will produce specific evidence that closes (or widens) the gap mechanically.

What this means for the verdict

The CAG thesis verdict (pass-with-trigger at re-engagement $11.50) is not overturned by this analysis. The reasoning:

AlphaSteve's central value of $24 is materially above current price but the trigger of $11.50 sits below sell-side mean PT — the kit is more conservative on action than the consensus that is itself bearish on the name. This is the appropriate deep-value discipline given the dividend-cut binary.
The specific component of the gap (multiple, EPS basis, dividend probability weighting) is the part that the FY26 Q4 print will resolve in one direction or the other. The thesis is positioned to update mechanically on the print.
The structural component of the gap (horizon, methodology) is irreducible — the kit will always be in disagreement with twelve-month-PT sell-side on names where the trough is in question. That is the kit working as designed per 02-philosophy-deep-value.
The convergence between AlphaSteve ($24) and GuruFocus GF Value ($25.74) is reassuring on the central value but does not change the trigger discipline. GuruFocus and the intrinsic-frame sources do not impose an MoS band; AlphaSteve does, and the kit's MoS band is the action-discipline mechanism that separates intrinsic-frame analysis from intrinsic-frame action.

Sources

Conagra Brands (CAG) Stock Forecast and Price Target 2026 — MarketBeat, retrieved 2026-06-02
Conagra Brands (CAG) Stock Forecast — Public.com, retrieved 2026-06-02
What Are Wall Street Analysts' Target Price for Conagra Brands Stock — Barchart, retrieved 2026-06-02
Conagra Brands Stock Forecast & Predictions — WallStreetZen, retrieved 2026-06-02
CAG Maintained by UBS — Price Target Lowered to $13 — GuruFocus, 2026-06-02
Jefferies trims price target as investors weigh outlook — Ad-hoc-news, 2026-05-27
Conagra Brands Inc — GuruFocus Stock Summary, retrieved 2026-06-02
Conagra Brands (NYSE:CAG) Stock Analysis — Simply Wall St, retrieved 2026-06-02
Klarman, S. (1991). Margin of Safety. HarperBusiness — variant-perception discipline; MoS bands.
Marks, H. (2011). The Most Important Thing. Columbia — second-level thinking; steelman discipline.
Greenwald, B. (2001). Value Investing. Wiley — EPV vs EPV-plus-growth methodology, ch. 6-8.
Mauboussin, M. & Rappaport, A. (2001). Expectations Investing. HBSP — market-implied-expectations framing for the reverse-DCF cross-check.

Linked

CAG-thesis — the underlying thesis
CAG-calibration — calibration tracker (Q4 FY26 print and August dividend declaration both flagged as checkpoints)
CAG-shadow-matrix — companion analysis (internal methodology comparison)
CAG — landing page
consensus-benchmarking — the methodology this file applies
methodology-calibration — the cross-thesis audit this feeds
earnings-power-value-greenwald — the valuation skill underlying the central
02-philosophy-deep-value — the doctrine guiding the disciplined action stance

CAG — Calibration tracker

Companion to CAG-thesis. This file is the calibration discipline the soul mandates (04-intellectual-virtues §10). Every quarterly check-in appends to this log. Supersedes the pre-thesis stub at 09-Theses/CAG/CAG-calibration.md (instance-level calibration is the canonical from publication forward per thesis-bundle-standard).

Thesis snapshot (initial — 2026-06-02 publication)


Date written	2026-06-02
Verdict	pass-with-trigger
Price at writing	$13.16
Central value	$24.00
Range	$19 – $30
Buy trigger	$11.50
Sell trigger (post-engagement)	$26 (trim to half), $29 (exit)
Position size	0% of portfolio
Conviction tier (at trigger)	Probe to Mid
Methodology	Greenwald-modified (EPV-only fallback; growth-runway gating test fails clean)
Gating tests passed	1 (of 3 — moat marginal, ROIIC marginal, runway clean fail)
Moat quality	Narrow-contested

Key predictions (testable)

Q3 FY26 organic +2.4% was the trajectory inflection, not an aberration. Q4 FY26 prints (mid-July 2026) with organic sales between -0.5% and +1.5%; volume positive or roughly flat.
Dividend maintained at $0.35 per quarter through the August 2026 declaration. The April director cluster buy and the 51-year streak together produce > 50% probability of maintenance through the next two earnings cycles.
FY27 adjusted EPS guidance issued with the Q4 print prints between $1.85 and $2.05 at midpoint. This implies modest organic recovery from the FY26 trough; sell-side consensus today is below this band.
Adjusted operating margin recovers to ~11.5% by FY27 from FY26 trough ~11.0%. Input-cost normalization and productivity offset half the inflation overhang.
Frozen single-serve meal share holds at 52%+ through FY27. GLP-1 effects are real but Conagra's category leadership absorbs the absolute volume compression without losing relative share.
No senior executive departures (CFO most-watched seat) through Q1 FY27. Stable leadership signals confidence in the trajectory.

Scenario probabilities (initial)

Scenario	Probability	Implied price (12-mo)	Implied price (24-mo)
Bull — Q3 trajectory holds; dividend maintained; FY27 EPS $1.95+; multiple re-rates	~30%	$18-$22	$22-$28
Base — Q3 trajectory partial confirmation; dividend maintained; FY27 EPS $1.80-$1.90; multiple held	~35%	$14-$17	$17-$22
Bear — Sustained decline; dividend cut; FY27 EPS $1.60-$1.70; multiple compressed	~25%	$9-$11	$10-$14
Tail — Deeper structural break; large dividend cut; FY27 EPS below $1.50; credit downgrade	~10%	$7-$9	$8-$12

Kill criteria (initial)

Specific, observable, falsifying conditions. When any fires, the thesis is re-evaluated immediately, not on the regular cadence.

Q4 FY26 print shows organic sales -1% or worse with volume negative
FY27 adjusted EPS guidance midpoint below $1.75
Dividend cut > 20% from $0.35 per quarter
Adjusted operating margin compression below 10% sustained two quarters
Interest coverage ratio falls below 3.0x
Frozen single-serve meal share declines below 50%
BBB credit rating downgrade to BB
Senior executive departure (CFO is the most-watched seat)
Insider open-market selling above $1M by directors who bought in April
Two consecutive quarters of negative organic sales after Q3 FY26

Re-read schedule

Date	Type	Owner	Status
2026-07-15 (estimated; actual print date TBD)	Q4 FY26 print and FY27 guide	alphasteve-thesis-builder	scheduled
2026-08-15 (estimated; actual declaration date TBD)	Q1 FY27 board dividend declaration	thesis-monitoring	scheduled
2026-10-01 (estimated)	Q1 FY27 print	alphasteve-thesis-builder	scheduled
2026-12-15 (estimated)	Q2 FY27 print	alphasteve-thesis-builder	scheduled
2027-06-02	12-month full re-read and calibration audit	alphasteve-thesis-builder	scheduled
2028-06-02	24-month full re-read and shadow-matrix audit	alphasteve-thesis-builder	scheduled

Checkpoint log

Each check-in appends a section. The question is not "was I right about price?" but "was the reasoning sound regardless of outcome?"

Checkpoint — 2026-06-02 (genesis)

What happened: Initial thesis publication. CAG closed Friday (May 29) at $13.13-$13.39; June 1 close band similar; June 2 (publication day) intraday near $13.16 with UBS cutting target from $16 to $13. The thirty-year low context is intact. Two directors (Mulligan, Lenny) remain ~8% underwater on the April 14 cluster buys with no further open-market activity.

Predictions vs actuals: N/A — initial publication.

Kill criteria status: None fired.

Reasoning assessment: Methodology fallback to EPV-only is the correct discipline given the growth-runway gating test failure. Central value of $24 reconciles with the GuruFocus GF Value of $25.74 ($1.74 below) and the Simply Wall St page DCF of $19.45 ($4.55 above). The trigger of $11.50 at 50% MoS preserves the asymmetric position above the modeled dividend-cut downside of $9.

Thesis update: Initial publication; no change.

New kill criteria added or removed: Initial list set.

Next checkpoint: 2026-07-15 (Q4 FY26 print and FY27 guide).

Checkpoint — 2026-06-03 (active-idea refresh — Bernstein sector downgrade)

Why unscheduled (per Rules §"Active-idea refresh"): material new evidence within 24 hours of publication — a Bernstein sector-wide packaged-foods downgrade with a CAG price target below the watchlist trigger. The portfolio-task's daily note recorded the first inception-period 10% proximity-band touch on the watchlist for CAG.

What happened: Bernstein analyst Alexia Howard cut CAG, CPB, GIS, KHC, and SMPL to Underperform from Market Perform on 2026-06-03; CAG price target lowered $16 → $12 on "dividend cut imminent" framing T3(https://www.gurufocus.com/news/8899230/cag-downgraded-by-bernstein-price-target-lowered-to-1200); T3(https://www.investing.com/news/analyst-ratings/bernstein-downgrades-conagra-stock-rating-on-dividend-cut-concerns-93CH-4723577); T3(https://finance.yahoo.com/markets/stocks/articles/cpb-cag-gis-khc-smpl-114137407.html). BofA cut CAG to $13 Underperform; UBS at $13; Goldman successive downgrade T3(https://finance.yahoo.com/markets/stocks/articles/bofa-trims-conagra-brands-inc-141447441.html). CAG closed $12.82 (range $12.61-$13.00), down −2.6% from $13.16 T3(/brain/daily-notes/2026-06-03)]. Watchlist gap to trigger $11.50 narrowed to −10.3% — first inception-period 10% proximity-band touch.

Predictions vs actuals: too early to score the six testable predictions; the Bernstein downgrade cluster is consistent with the thesis's bear/tail scenario (probability ~25% / ~10%) and does not refute it. Sell-side mean PT $15.47 at publication is now cohort-pressured lower as multiple sell-side names converge below the kit's central value $24.

Kill criteria status: None fired. Kill criterion "Dividend cut > 20% from $0.35 per quarter" remains the load-bearing test — Bernstein's framing predicts the cut but no board declaration yet. Kill criterion "Insider open-market selling above $1M by directors who bought in April" has not been observed; Mulligan and Lenny carry the $608K April cluster underwater. The next dividend declaration window aligns with the Q4 FY26 print 2026-07-15.

Reasoning assessment: The Bernstein PT $12 sits between the current $12.82 and trigger $11.50 — it is the sell-side's pricing of the dividend-cut scenario, not a falsification of the EPV-only central $24. The thesis explicitly priced the dividend-cut path as the trigger-and-buy mechanism (modeled dividend-cut downside ~$9). The sell-side cluster moving in the thesis direction within 24 hours of publication ratifies the cohort read; the variant-perception edge (deep-value before consensus capitulation) compresses but does not invert. The portfolio task's thesis-cold-re-read on the proximity flag was performed per Watchlist discipline; conclusion was no-buy-ahead-of-trigger.

Thesis update: No update. Methodology, central value, trigger, and verdict carry. Add intermediate-check trigger to the re-read schedule: if CAG prints below $12.00 (the Bernstein PT, halfway between current and trigger), perform an unscheduled re-read regardless of the 2026-07-15 calendar date.

New kill criteria added or removed: None added or removed; the existing dividend-cut and insider-selling criteria already cover the discriminator path.

Next checkpoint: 2026-07-15 (Q4 FY26 print and FY27 guide) — carries. Intermediate-trigger checkpoint at $12.00 print added.

Checkpoint — 2026-06-09 (active-idea refresh — Q3 FY26 print landed early) — ⚠️ VOID, BUILT ON MISATTRIBUTION

[corrected 2026-06-10 — see audit-log #012] This checkpoint is VOID. It was built on a false premise: Conagra did not report on 2026-06-08. The June 8 print described here was Campbell Soup (CPB), a packaged-foods cohort peer, whose Q3 FY26 (quarter ended 2026-05-03) showed net sales −4% to $2.4B and adjusted EPS $0.50 (−32%). The cited "Conagra 8-K" (CIK 0000016732) is Campbell's filing. Conagra's actual Q3 FY26 was reported 2026-04-01 ($0.39 adjusted EPS, −23.5% — already known at the 2026-06-02 thesis), and its next print is Q4 FY26 in mid-July. No genuine new CAG evidence landed on June 8. The original (misattributed) checkpoint text is preserved below in strikethrough for the audit trail; it should not be read as a CAG data point.

New evidence: Conagra reported fiscal Q3 on 2026-06-08, ahead of the July 15 date this tracker had penciled in. Net sales fell 4% to $2.4B, adjusted earnings of $0.50 fell 32% year-over-year and beat the $0.48 consensus by a cent, and full-year guidance was reaffirmed; the stock traded roughly flat on the print T1.
Impact on thesis: neutral. The print continues the known top-line and margin softness without a fresh discriminator. The reaction stayed under 10%, so it does not clear the active-coverage action bar. The dividend declaration — the load-bearing kill criterion — was not in this release.
Action: no change. Methodology, central value $24, trigger $11.50, and the $12.00 intermediate-trigger all carry. The next discriminator is the dividend declaration, not this quarter's operating line. The $11.50 trigger was not approached — CAG sat near $13.02 on the rebound, gap about −11.7%, out of the proximity band.

Corrected reading: No CAG-specific evidence emerged June 8. Methodology, central value $24, trigger $11.50, and the $12.00 intermediate-trigger carry unchanged from the 2026-06-03 checkpoint — but on the basis of no new information, not a confirming print. The next genuine discriminator is the dividend declaration / Q4 FY26 print (mid-July).

Next checkpoint: 2026-07-15 (Q4 FY26 print and FY27 guide) or a $12.00 print, whichever comes first.

Calibration scorecard (rolled up after multiple checkpoints)

To be populated at first quarterly check-in. Tracking will include:

Hit rate on the six initial testable predictions
Calibration of the four-scenario probability assignment against actual price/business outcomes
Methodology shadow-matrix accuracy (whether the chosen Greenwald-modified central value was closer to subsequent actual price than the Klarman, Buffett-modern, or Mauboussin-compounder alternatives — see CAG-shadow-matrix)
Patterns of error (overconfidence in moat durability, underweighting of dividend-cut probability, multiple-assumption miscalibration)

Linked

CAG-thesis — the thesis
CAG-shadow-matrix — the four-methodology dispersion
CAG-consensus-gap — the external consensus comparison
CAG — landing page
CAG/first-read-2026-06-02 — the source first-read
GIS/first-read-2026-06-01 — the structural twin
methodology-calibration — where shadow-matrix data aggregates
05-decision-framework — calibration loop discipline
04-intellectual-virtues — what calibration is for
base-rates — base-rate context for consumer-staples mean reversion