Yield

Bottom line

Devon's FY2025 adjusted free cash flow computes to $1,661 million — reported FCF of $3,119 million, less $99 million of share-based compensation, less a $1,359 million five-year average acquisition charge [1]. On the pre-merger share base that is a 5.9% yield, and the three-year average is 6.2%. On the post-Coterra company, consensus FY2027 adjusted FCF lands near 11% of a $51.2 billion market cap — at the 10% reference line the balance sheet selects, not comfortably above it.

The adjustment, line by line

fit_features.adjusted_fcf returns not_computable for Devon — the structured cash-flow feed carries only two years with a free-cash-flow field and no share-based compensation before FY2018, so the derivation never finds a complete five-year acquisition window. The table below rebuilds the same formula directly from the filed Consolidated Statements of Cash Flows in the FY2021, FY2023 and FY2025 Form 10-Ks, which together cover 2019 through 2025 [2][3][4].

Figures in $ millions. Adjusted FCF = operating cash flow − capital expenditures − share-based compensation − trailing five-fiscal-year average of "Acquisitions of property and equipment"; derived from the filed cash-flow statements in the FY2021 [5], FY2023 [6] and FY2025 [7] Form 10-Ks. The five-year average column is blank for 2019–2022 because the corpus holds no 10-K covering 2017–2018 acquisitions.

Two things the adjustment removes. Share compensation is small and steady — $88 million to $115 million a year across all seven years, $99 million in FY2025, or 3.2% of that year's reported FCF and 1.5% of operating cash flow [8]. The acquisition charge is the whole of the adjustment. Devon spent $6,795 million of cash on acquisitions across 2021–2025 — $2,583 million in 2022 and $3,808 million for Grayson Mill in 2024 against $18 million, $64 million and $322 million in the other three years [9][10]. Averaged, that is a $1,359 million annual charge — 44% of FY2025 reported FCF. The reported yield overstates the owner's yield by that much because the cash Devon calls free has, at this company's demonstrated rate, gone into buying reserves rather than to shareholders.

Devon's own FY2025 free-cash-flow headline of $3.1 billion [11] ties to the $3,119 million computed here — an 11.1% yield on the same $28,108 million base — so the disagreement between the framework's number and the company's number is entirely the adjustment, not a definitional dispute about capex.

The cash-acquisition line also understates how much of Devon's growth has been bought rather than drilled, because the three largest transactions of the period were paid in stock: WPX in 2021, part of Grayson Mill in 2024 [12], and Coterra in 2026, where each Coterra share converted into 0.70 of a Devon share [13]. Share count went from 377 million at the end of 2020 to 633 million at the end of 2025 to 1,153,403,107 in May 2026 [14]. A rising share count is an outright exclusion under the framework, and it is treated at length in Self-Help; it is named here because per-share adjusted FCF, not aggregate adjusted FCF, is what a yield means.

The yield, three ways

Partial = (reported FCF − share-based comp) ÷ fiscal-year-end market cap. Full = partial less the trailing five-year average acquisition charge, computable only from FY2023. Market cap uses same-year period-end share count and the last close on or before December 31; derived from the filed cash-flow statements [15] and company filings, as reported.

Current, on the pre-merger base. FY2025 adjusted FCF of $1,661 million against the feature file's $28,108 million market cap (633 million shares at $44.405, July 29, 2026) computes to 5.9% — $2.62 of adjusted FCF per pre-merger share against a $44.405 price.

Three-year average. Adjusted FCF averaged $1,749 million across FY2023–FY2025 ($2,027M, $1,560M, $1,661M). On the same $28,108 million market cap that is 6.2%.

The baseline. fit_features.yield_baseline returns not_computable for the same reason the adjusted series does. Rebuilt, the three computable years sit at 7.0%, 7.5% and 7.2% — a median of 7.2% and a coefficient of variation of 11.5%, which is a genuinely stable baseline over the window it covers. The current 5.9% is below that baseline, not above it: the stock has recovered from $26.80 on April 8, 2025 to $44.405 while adjusted FCF has moved sideways. The wider seven-year partial-adjustment series (0.2%, 3.7%, 9.6%, 14.7%, 8.8%, 13.8%, 13.0%) has no stable level at all — it is the commodity cycle, not a baseline.

There is no jump signature here. The fortress pattern the framework hunts — a name that sat at 3.5–4% for years and suddenly prints 8–9% — requires both a stable prior level and a step up from it. Devon's stable level is roughly 7%, and today's reading on the pre-merger base is 1.3 percentage points beneath it.

Which bar applies

fit_features.balance_sheet_class returns unknown because EBITDA is missing for FY2025 in the structured feed. The computation from the filed statements:

• Net debt = short-term debt $998M + long-term debt $7,391M [16] − cash and cash equivalents $1,384M [17] = $7,005M
• EBITDA = earnings before income taxes $3,466M + financing costs, net $455M + depreciation, depletion and amortization $3,595M = $7,516M [18]
• Net debt / EBITDA = 7,005 ÷ 7,516 = 0.93x. Vendor consensus records FY2025 EBITDA of $7,413M, which gives 0.95x — the same answer.

The framework's rule places fortress at net cash or 0.5x and below, and levered at 3.0x and above. At 0.93x Devon is squarely moderate, so the 10% reference line applies, not the 8–9% fortress line and not the 25% levered line.

Devon labels the same balance sheet "Fortress" in its own decks [19] while disclosing 0.9x adjusted pro forma net-debt-to-EBITDAX on the same slide deck family [20]. Both statements describe the same number; the framework's arithmetic and the company's adjective simply set different thresholds. Forward, the gap narrows: consensus carries net debt to $9,005M against $12,546M of FY2026 EBITDA (0.72x) and to $5,593M against $14,828M in FY2027 (0.38x), against a stated company target of gross debt near $9 billion by year-end 2027 and leverage held under 1.0x through the cycle [21]. If the FY2027 consensus balance sheet arrives, the class flips to fortress and the applicable line drops to 8–9%. That is a forward observation, not the basis on which this tab measures.

Where the name sits. 5.9% on FY2025 adjusted FCF against the 10% bar — 410 basis points short. On the three-year average, 6.2% against 10% — 380 basis points short. Both computed on Devon standalone against the pre-merger equity base. The forward figures below are what change this picture.

Normalized mid-cycle yield

Devon is meaningfully cyclical — this is an oil and gas producer whose cash margin is a direct function of two index prices — so the current-year number needs a mid-cycle check rather than a sentence of dismissal.

Source: WTI and Henry Hub index prices as disclosed in Devon's own realized-price tables — FY2021 Form 10-K [22], FY2023 Form 10-K [23], FY2025 Form 10-K [24].

The mid-cycle assumption, stated so it can be recomputed. The window is FY2020–FY2025, the six fiscal years Devon's own filings disclose an annual index price for. It contains one collapse (2020, WTI $39.59) and one spike (2022, WTI $94.39), which is what makes it usable as a cycle rather than a trend. The arithmetic mean is WTI $70.02 and Henry Hub $3.50. A reader who prefers to drop the two extremes gets WTI $71.54 and Henry Hub $3.07 on the remaining four years — oil essentially unchanged, gas 12% lower.

Against that mid-cycle, FY2025 was modestly below, not depressed: WTI averaged $64.87, about 7% under the six-year mean, and Henry Hub $3.43, roughly at it [25]. The current environment is at or above mid-cycle on oil and well above it on gas: Q1 2026 printed WTI $72.10 and Henry Hub $5.05 [26].

Volume and cost assumption. Volumes and cost structure are held at the company's published full-year 2026 guidance for the combined company — oil 490–510 MBbls/d, NGLs 315–330 MBbls/d, gas 3,300–3,400 MMcf/d, total capital $4,800–5,000 million, with oil realized at 98–100% of WTI, NGLs at 24–26% of WTI and gas at 40–50% of Henry Hub [27]. Only price is flexed.

The two sensitivities. For oil, Devon published its own: standalone, Q2 2026 annualized free cash flow of roughly $4.8 billion at $90 WTI rising to $6.5 billion at $110 WTI [28] — $85 million of FCF per $1/Bbl on approximately 392 MBbls/d of oil. Scaled to the combined company's 500 MBbls/d guidance midpoint, that is $108 million of FCF per $1/Bbl of WTI. For gas, no company sensitivity is published; computed from the guidance inputs, 3,350 MMcf/d is 1,223 Bcf a year, at a 45% Henry Hub realization, net of 7% production and property taxes and a 20% current cash tax rate, giving $409 million of FCF per $1.00/Mcf of Henry Hub.

What the normalization shows. Consensus FY2027 EBITDA of $14,828 million against pro forma production of 1,604 MBoe/d — the 1Q26 combined figure Devon publishes [29] — is $25.33 per Boe of unit cash margin. Strip out the $1.0 billion synergy target [30] and it is $23.62 per Boe. Devon standalone earned $24.51 per Boe in FY2025: $7,516 million of earnings before income taxes, financing costs and DD&A [31] over 840 MBoe/d of production, at WTI $64.87 and Henry Hub $3.43 [32]. So the forward consensus embeds a per-barrel cash margin, before synergies, about 4% below what Devon actually earned in a year priced at $65 oil. That is a mid-cycle forecast, not a peak-price one — the forward yields below do not need to be discounted for an inflated commodity deck. The counter-fact: consensus is thin at the annual level (7 to 8 contributors on revenue, 17 on EBITDA) and 1Q26's actual $5.05 Henry Hub is 44% above the six-year mean, so a gas mean-reversion of $1.50/Mcf would remove roughly $614 million of combined FCF on the sensitivity above.

The consensus check

fit_features.consensus_forward_yield reads FY2026 22.4%, FY2027 26.4%, FY2028 26.6% and FY2029 28.0%. Those are consensus free cash flow for the combined company divided by the pre-merger $28.1 billion market cap, and they overstate by a factor of 1.82. The consensus series is combined-company on its face: quarterly consensus revenue steps from $3,946 million for Q1 2026 (Devon standalone, actual $3,807 million) to $6,189 million for Q2 2026 and $6,874 million for Q3 2026, which is when Coterra enters. The correct denominator is 1,153,403,107 shares at $44.405 = $51,217 million [33]. Devon corroborates the scale independently: its $8 billion repurchase authorization is described as "representing 15% of our market value," which implies roughly $53 billion [6].

Consensus free cash flow means from the vendor estimates file, vintage July 30, 2026 (data/sp/estimates.json). Feature-file yield divides by $28,108M; corrected yield divides by $51,217M — 1,153,403,107 shares [34] at the July 29, 2026 close of $44.405.

The framework's yield basis is adjusted FCF, and the vendor metric is plain free cash flow — cash from operations less capital expenditures, the closest available proxy, with no direct adjusted-FCF consensus published. Applying the two adjustments to FY2027, the first clean full year with Coterra:

All four rows start from consensus FY2027 free cash flow of $7,417M and deduct $200M of assumed combined share-based compensation (Devon standalone ran $99M [35]; Coterra's figure is not in this corpus). The revenue-scaled charge takes Devon's five-year average acquisition spend as 8.1% of FY2025 revenue and applies that rate to consensus FY2027 revenue of $26,622M.

The bar requires $5,122 million — 10% of $51,217 million. On Devon's own demonstrated acquisition rate the forward figure clears it by $736 million; on a charge scaled to the larger asset base it misses by $61 million. The honest statement is that consensus forward adjusted FCF lands at the 10% line, within the width of the assumption about future acquisition spend, in a range of 8.8% to 14.1% and a central estimate near 11%.

Does the sell side already agree? On unadjusted free cash flow, yes and by a wide margin — 14.5% for FY2027 rising to 15.4% by FY2029, well clear of any of the framework's three lines. That is the configuration the framework treats as the strong version of the setup: consensus itself underwrites the cash, so the disagreement would have to be about fear rather than fundamentals. The problem for the setup is on the other side of the equation. Devon is not trading at a fear price. The stock is $44.405 against a $47.03 peak on July 31, 2024 — down 5.6%, having recovered from the $26.80 trough of April 8, 2025. The 43% drawdown the framework looks for happened and has already been repriced; the anatomy of it is in Dislocation.

The forward path. No mean-reversion underwrite is required here, because consensus is not below the bar. What has to hold instead is durability. Using the sensitivities above and the Devon-trailing acquisition charge, the $736 million cushion above the bar is equivalent to $6.79/Bbl of WTI or $1.80/Mcf of Henry Hub. Consensus is built on a per-Boe cash margin already 4% below Devon's FY2025 realization, so the cushion is not a peak-price artifact — but under the revenue-scaled acquisition charge there is no cushion at all, and any further stock-funded acquisition resets the per-share arithmetic the way Coterra just did. Weighing those: roughly a 55–65% probability that adjusted FCF yield sits at or above 10% on a one-to-three-year view, with the outcome turning far more on how much cash and stock the combined company spends on acquisitions than on the commodity deck. What would move that estimate: a board-set capital-allocation framework that caps acquisition spend, or a first full combined year in which acquisition spend runs at the low end.

FCF to revenue

Revenue is revenue from contracts with customers as disclosed in the FY2025 Form 10-K [36]; cash flows from the filed statements [37][38][39].

Reported conversion is stable, and stable in the direction that matters. FCF/revenue held at 17.6%, 18.6% and 18.6% across FY2023–FY2025 while the WTI index Devon realizes against fell from $77.62 to $75.79 to $64.87 [40][41]. Flat conversion into a 16% fall in the oil price is improving conversion — the arithmetic of the $1.0 billion business-optimization programme, 85% captured on the Q4 2025 exit rate [42], showing up in cash rather than in a slide.

Adjusted conversion tells a different story: 13.4%, 9.8%, 9.9%. The step down between FY2023 and FY2024 is the Grayson Mill purchase entering the five-year average, not an operating deterioration. It is a fair reading that the adjusted series is the honest one for an acquirer and the reported series is the honest one for the underlying assets, and the two only converge if acquisition spend falls.

On consistency of cash generation, the record is better than the yield level. Reported FCF has been positive every year of the last decade the corpus covers, including $311 million in 2020 at WTI $39.59 with a $2.7 billion net loss [43]. The rolling five-year average has risen through the last three windows — $2,401 million, $2,965 million, $3,527 million — with a coefficient of variation of 15.5%; the three computable adjusted-FCF years vary by 11.5%. Volatile year to year, but not unpredictable, which is the distinction the framework draws.