Cheniere Energy LNG

Reference table

1. What they actually own — two terminals, two pipelines, one product

Cheniere Energy, Inc. (NYSE: LNG, founded 1983, headquartered at 845 Texas Avenue, Houston; CEO Jack A. Fusco; 1,717 full-time employees; listed since 1994-04-04) is described in this dataset as "an energy infrastructure firm predominantly focused on liquefied natural gas (LNG) related activities within the United States."

The asset list is short, which is itself the point:

Revenue concentration (FY24, the most recent product disclosure in this dataset):

Roughly 95% of revenue is one product from two sites. That concentration is the whole risk-and-return character of the company: enormous operating leverage to LNG volumes and spreads, essentially no diversification, and a small number of very large physical assets whose operational availability is the business. Compare the ET dive of the same date, where the revenue mix spans six product lines across nine states.

Data caveats, stated:

1. The most recent product segmentation is FY24 (2024-12-31) — there is no FY25 or FY26 segment breakdown available, so the mix above is roughly nineteen months stale.

2. The FY22 segment row is corrupt: it reports Liquefied Natural Gas of $112.0B against total FY22 income-statement revenue of $33.756B. That figure is off by more than 3x and is excluded.

3. Geographic segmentation is effectively absent. The only entries are FY17 (Ireland $787M, Japan $1.2B, Korea $762M, non-US $4.0B, United States $1.6B) and FY16. Cheniere provides no current geographic segment reporting in this dataset — which is a real gap for an export business whose entire economics depend on destination markets. It is named rather than papered over, and no geographic analysis is offered.

2. The Q1'26 problem — the most important section in this dive

This is the section that determines the verdict, so it is placed before the financials rather than buried in them.

What Q1'26 (2026-03-31) reported:

Do the arithmetic that the income statement forces you to do. Operating income was +$1.973B. Subtract interest expense of $255M and you are at +$1.718B. Reported pretax income was -$3.753B. The gap is approximately $5.47B of non-operating losses, and this dataset does not identify a single dollar of it.

What this dive will and will not say about it.

Will say: the operating business was strong and accelerating. Revenue +24.8%, gross profit +29.6%, operating income +30.6% — every operating line improved materially year over year. Whatever happened, it did not happen in the terminals.

Will say: for a company holding large long-dated natural-gas supply agreements and LNG sale-and-purchase contracts, a multi-billion-dollar non-operating swing is structurally consistent with a derivative mark-to-market movement, and such marks are typically non-cash and reverse as contracts settle. Note that FY25 operating cash flow was $5.539B against net income of $5.330B — a 1.04x conversion — and the TTM incomeQuality metric is 2.71x, both indicating the reported earnings and the cash generation diverge substantially.

Will NOT say: that the loss was a derivative mark. That is inference from the shape of the business, not a fact in this dataset, and asserting it would be fabrication. The honest position is: a $5.5B below-the-line loss occurred, its nature is not disclosed here, and until the 2026-08-06 print clarifies it, GAAP earnings for this company are not a usable valuation input.

The consequences are concrete and appear throughout the rest of this dive:

• FY26E consensus EPS is NEGATIVE $6.62222 (seven analysts, range -$8.19990 to -$5.79821). A forward P/E cannot be computed for the current year.
• The TTM P/E of 42.0x is arithmetically real but analytically meaningless, since the trailing twelve months contain the charge.
• Every valuation figure in §4 is therefore built on EV/EBITDA, using consensus EBITDA — and the reason is stated there explicitly.
• Consensus EPS is non-monotonic afterwards: $16.49119 (FY27E) → $14.93334 (FY28E) → $13.22983 (FY29E) → $15.415 (FY30E), on ten, ten, four and four analysts. Declining EPS on rising EBITDA is consistent with new capacity bringing depreciation and interest ahead of full contribution, but the FY29-30 estimates rest on four analysts and should be treated as noise rather than forecast.

A further data discrepancy worth flagging: the earnings calendar records Q1'26 revenue as $5.868B (filed 2026-05-07) while the income statement reports $6.650B for the quarter ended 2026-03-31 — a $782M difference. The income statement is used throughout this dive.

3. The financials — a strong operating business under a heavy balance sheet

Annual series:

Read that table with appropriate suspicion. EBITDA of $6.226B in FY22 against operating income of $11.097B, and EBITDA of $17.538B in FY23 against operating income of $7.643B, are not reconcilable through depreciation alone. The reported EBITDA line is heavily influenced by the same non-operating items that produced the Q1'26 anomaly, and it swings violently. The stable, trustworthy series in this table is operating income and revenue. Interest expense is also clean and has been falling — from $1.406B to $948M, a 33% reduction over four years, which is genuine deleveraging showing up in the income statement.

FY25 versus FY24 on the clean lines: revenue +24.4%, operating income +9.4%, net income +63.9%, EPS +70.0% (helped by an 3.8% reduction in the share count).

TTM profitability: gross margin 36.2%, EBITDA margin 31.1%, operating margin 30.5%, net margin 7.00% (depressed by the Q1 charge). Return on equity 23.5%, return on invested capital 11.2%, return on capital employed 16.2%. Interest coverage 6.60x. Effective tax rate 27.4% — Cheniere is a C-corporation and pays full corporate tax, unlike the two partnerships in this batch.

Cash flow — the series that actually works:

Two things stand out.

First, the capital return has overtaken free cash flow. FY25 returned $3.175B against $2.461B of free cash flow — a 1.29x ratio, funded by drawing cash down $1.606B. Cash fell from $4.066B (FY23) to $2.638B (FY24) to $1.584B (FY25). That is a deliberate choice to run a leaner balance sheet while buying back stock, and it is sustainable only while cash flow holds. It is the same structural pattern flagged in the BAM dive — returning more than you generate — and it deserves the same scepticism.

Second, the buyback is doing genuine work. Diluted share count: 253.4M (FY22) → 242.6M (FY23) → 229.1M (FY24) → 220.3M (FY25) → 210.5M (Q1'26). That is a 16.9% reduction in three and a half years, roughly 5% a year, which is a material and reliable contributor to per-share value regardless of what the derivative line does. At $258.08, FY25's $2.724B of repurchases equals about 5.0% of the current $54.08B market capitalisation.

Balance sheet (FY25, 2025-12-31): cash $1.584B, total current assets $3.692B, net PP&E $38.455B, goodwill $77M, total assets $49.127B; total current liabilities $3.916B, short-term debt $933M, long-term debt $24.670B, total debt $28.609B, net debt $27.025B, total liabilities $36.049B; total equity $13.078B including $5.163B of minority interest (the public units of Cheniere Energy Partners, CQP — which trades separately at $67.00 with a $32.43B market capitalisation per the peer list).

Note the leverage ratios honestly: net-debt/EBITDA 3.76x (TTM), debt-to-equity 7.03x, debt-to-capital 87.6%, current ratio 0.57x, working capital -$3.113B, net current asset value -$34.014B. This is a heavily levered balance sheet. What makes it serviceable rather than dangerous: interest coverage of 6.60x, interest expense that has fallen four years running, $5.539B of annual operating cash flow, and $38.455B of long-lived contracted physical assets behind it.

The EV bridge, stated because the numbers differ. The TTM key-metrics enterprise value is $78.720B against a market cap of $54.081B — implying net debt of $24.639B, versus $27.025B at FY25 year-end. The difference reflects a more recent balance-sheet date not included in the annual series. This dive uses the $24.639B TTM-bridge figure consistently in all fair-value arithmetic, and flags that using the FY25 year-end figure instead would reduce each fair-value anchor by approximately $11 per share.

Insider signal: none usable. All eight recent filings are director A-Awards at a $0 reported price (Vitalone 614 shares 2026-07-14; Shear, Gray, Edwards, Mitchelmore and Moreland between 809 and 1,411 shares on 2026-05-14) plus one F-InKind disposition of 394 shares at $239.38 by Mitchelmore on 2026-05-13, which is a tax-withholding transaction, not a sale decision. There are no open-market purchases and no discretionary sales in this dataset, and no insider signal is claimed.

4. Valuation — priced in or room?

Method note, stated up front: this dive values Cheniere on EV/EBITDA, not on earnings. The reason is §2 — FY26E consensus EPS is negative $6.62222 and the trailing series contains a $5.5B unexplained charge. An earnings multiple built on that base would be theatre. EBITDA estimates, by contrast, are internally coherent across all seven forecast years and are supported by seven to nine analysts each.

At $258.08 (market cap $54.08B, EV $78.72B, ~210.5M diluted shares, net debt ~$24.64B per the TTM bridge):

Three observations.

First, the EBITDA path is modest. $9.619B (FY26E) to $11.017B (FY30E) is a 3.5% compound annual rate over four years. For the company at the front of what the knowledge base calls a historic structural gas deficit, that is a strikingly unambitious consensus — and it is the same pattern seen in the ET dive (1.9%) and the MPLX dive (4.8%). The street models American gas infrastructure as a low-growth utility complex.

Second, the FY29-30 numbers should be discounted. Revenue falls from $26.047B (FY29E) to $25.377B (FY30E) and EBITDA falls from $11.307B to $11.017B, on four analysts. That is thin-coverage noise. All fair-value work below anchors on FY27E, which carries nine EPS and nine revenue estimates.

Third, the FCF yield picture is genuinely attractive and worth naming. The TTM freeCashFlowYield is 8.38% and priceToFreeCashFlow is 11.93x. On the FY25 annual figures the picture is tighter — $2.461B of FCF on a $54.08B cap is 4.55% — because FY25 capex jumped to $3.078B. Both figures are real and they measure different windows; the gap is the FY25 capex step-up. Either way, this is a cash-generative business, which is what makes the buyback credible.

Synthos fair values — stated arithmetic on EV/EBITDA, no DCF theater. Method: target multiple × FY27E consensus EBITDA of $10.263B = enterprise value, minus net debt of $24.639B (TTM bridge) = equity value, divided by 210.5M diluted shares.

• Bear ~$224 = 7.0x FY27E EBITDA → EV $71.84B → equity $47.20B → $224.24/share. The de-rating case: the export-cargo-shutoff scenario bites, marketing margins compress, the derivative charges recur, and the market applies a discount for earnings opacity. Sits near the 200-DMA ($233.08) and above the 52-week low ($186.20). -13.1% from spot.
• Base ~$273 = 8.0x FY27E EBITDA → EV $82.10B → equity $57.47B → $273.03/share. An 8.0x multiple is set as base because it is roughly a third of a turn above where the shares trade today and consistent with a contracted, cash-generative export infrastructure asset with a ~5%/yr buyback — normalisation, not enthusiasm. It also lands just below the street's $278.50 consensus. +5.8% from spot.
• Bull ~$322 = 9.0x FY27E EBITDA → EV $92.37B → equity $67.73B → $321.75/share. Requires the full re-rating: the gas-deficit wedge lands in export volumes rather than throttling them, the derivative noise resolves, GAAP earnings become usable again, and the market pays an infrastructure multiple. Sits above the street's $300 high target and above the 52-week high of $296.91. +24.7% from spot.

Cross-check on earnings: the base $273 is 16.6x FY27E consensus EPS of $16.49119, against 15.7x today. For a company generating $5.5B of annual operating cash flow with a 5%-a-year buyback, 16.6x forward earnings is unremarkable. The methods agree.

Sensitivity note: using the FY25 year-end net debt of $27.025B instead of the TTM bridge's $24.639B lowers each anchor by approximately $11 — bear $213, base $262, bull $311. The verdict does not change.

6a. What today's price assumes (the inversion)

At $258.08 (7.67x FY27E consensus EV/EBITDA), today's price embeds roughly the following. Every figure is consensus-derived arithmetic, labelled as such.

• EBITDA compounds just 3.5% a year through FY30. Consensus: $9.619B (FY26E) → $11.017B (FY30E), with a decline from FY29E's $11.307B. Today's price assumes contracted train capacity ramps modestly and then plateaus. Falsified in the bull's favour by: any upward revision to FY28E EBITDA above ~$11.5B (from $10.595B), which would mean the LNG export wedge is landing in Cheniere's numbers. Falsified in the bear's favour by: FY27E EBITDA drifting below ~$9.5B.
• The Q1'26 charge does not repeat, and FY27 EPS recovers to $16.49. Consensus goes from negative $6.62222 (FY26E) to positive $16.49119 (FY27E) — a $23.11 per-share swing modelled as a one-year reversal. This is the most fragile assumption in the price. If the below-the-operating-line losses recur in Q2 or Q3, the FY27 recovery does not arrive on schedule, and the market has no earnings anchor to fall back on. The 2026-08-06 print is the direct test.
• The buyback continues at roughly $2.7B a year — about 5.0% of the current market cap annually — funded by free cash flow that was $2.461B in FY25 against $3.175B of total capital return. The price assumes free cash flow recovers to cover the return. Falsified by: capex staying at or above the FY25 level of $3.078B while operating cash flow stays near $5.5B, which forces either a smaller buyback or a further cash draw (cash already fell from $4.066B in FY23 to $1.584B in FY25).
• The market keeps paying ~7.7-8.0x forward EV/EBITDA despite an earnings statement that is currently unreadable, 3.76x net-debt/EBITDA, a 0.57x current ratio and negative $3.113B of working capital. That is a modest multiple that already embeds some discount for those facts — which is why the bear case is only -13% rather than -30%.

6b. The return bridge (why the multiple moves)

> expected return ≈ EBITDA growth + multiple drift + shareholder yield

• EBITDA growth: +6.7% (consensus FY26E $9.619B → FY27E $10.263B). The four-year rate is much lower at 3.5%, so 6.7% is the best single year in the forecast window, not the run rate.
• Multiple drift: +0.33 turns, or roughly +4% on enterprise value — from the 7.67x the market pays on FY27E today to the 8.0x base. This dive assumes only modest multiple expansion, and the driver is risk normalisation rather than growth maturation. Specifically: the earnings opacity resolving. If Q2 and Q3 show clean, positive GAAP earnings and the FY27 consensus of $16.49 holds, the discount the market currently applies for unreadable accounts should partially close. If it does not, there is no expansion.
• Shareholder yield: +5.9% — a 0.84% dividend plus roughly 5.0% of buyback at the FY25 pace. This is the largest single component of the return and the most reliable one, because it does not depend on any multiple or any estimate; it depends only on cash flow continuing.

Say it plainly: roughly 60% of the base-case total return is shareholder yield, not price appreciation. That is unusual, and it is the strongest structural argument for the name — a business retiring 5% of its equity a year does not need a re-rating to compound. It is also why the Hold is a patient Hold rather than an avoid: the buyback works whether or not the market ever repairs the multiple. But note the leverage effect works against you here relative to ET: net debt is only ~31% of enterprise value, so a 4% EV move produces about a 6% equity move, not a 17% one.

6c. Variant perception (where we differ, what would surprise)

• Where we differ from a near-unanimous street: on whether the demand wedge is unambiguously good for Cheniere. Twenty-seven analysts, zero sells, $278.50 consensus. The implicit view is that more American gas demand equals more Cheniere. We think the relationship is not monotonic, and the knowledge base's own sharpest gas voice says so explicitly. invest_like_the_best (2026-07-21, conviction 70): "In the coming shortage, gas could spike like Russia/Ukraine did ($8-10/MCF); he hesitates to put a hard target but sees prices high enough to force shutting off US export cargos." A domestic gas price high enough to shut in export cargos is not a tailwind for the largest US exporter. This is the actual variant perception in the name, it is sourced rather than asserted, and it is why the verdict is Hold against a street that has no sells.
• Where we side with an existing tracked bear. compound_and_friends (2026-05-01, bearish, conviction 55): "Wouldn't invest in Cheniere despite its LNG-export moat: margins are a function of the natural-gas price and management lacks control over its inputs, disqualifying it as a quality compounder." Note that the same source was bullish six weeks earlier (2026-03-31, conviction 60 and 65). A tracked voice flipping on a name inside six weeks is itself a signal about how legible the business is — and it corroborates §2's finding that the accounts have become hard to read.
• Where we differ from the bear case: the tolling structure is real and the buyback is real. Cheniere's long-dated sale-and-purchase and tolling agreements exist precisely to decouple cash flow from the gas price, and operating income rose 30.6% in Q1'26 while the gas backdrop was volatile — evidence that the structure works at the operating line even when it fails at the net line. And a company retiring 5% of its shares a year is compounding per-share value independent of any of this. We do not think this is a bad business. We think it is an unreadable one this quarter.
• Positive surprise that would force a reprice up: the 2026-08-06 print showing (a) EPS at or above the $3.08 consensus, (b) no repeat of the below-the-operating-line loss, and (c) an explanation of the Q1 charge that confirms it as non-cash and reversing. That combination validates 8.5-9.0x and puts the $322 bull anchor in play. Watchable numbers: reported EPS against $3.08, and the gap between operating income and pretax income.
• Negative surprise that would force a reprice down: a second consecutive quarter with a large unexplained non-operating loss, or an operating-income deceleration from the Q1 pace of +30.6%. In that case 7.0x and the $224 bear anchor becomes operative, and the earnings-opacity discount becomes structural rather than temporary. Watchable number: pretax income versus operating income minus interest — the same $5.5B arithmetic run on Q2.

5. The knowledge base — four direct claims, and one source on both sides

Direct coverage on Cheniere: 4 claims across 2 tracked sources — and they do not agree.

Bullish:

• all_in, 2025-05-06 (bullish, conviction 80), entities ['Cheniere Energy','Sabine Pass']: "LNG/methane has a massive growing global market with lower carbon footprint; US export demand exists regardless and is now the #2 dollar export." The highest-conviction direct claim on the name, and it names the flagship asset.
• compound_and_friends, 2026-03-31 (bullish, conviction 65), entities ['LNG','Cheniere Energy']: "US is the Saudi Arabia of natural gas (100-200yr supply); Cheniere's Gulf export terminals are active because the world needs LNG."
• compound_and_friends, 2026-03-31 (bullish, conviction 60), entities ['Cheniere Energy','LNG']: "US is the Saudi Arabia of natural gas; Cheniere's Gulf export terminals are active and running full because the world needs LNG."

Bearish — and from the same source, six weeks later:

• compound_and_friends, 2026-05-01 (bearish, conviction 55), entity ['LNG']: "Wouldn't invest in Cheniere despite its LNG-export moat: margins are a function of the natural-gas price and management lacks control over its inputs, disqualifying it as a quality compounder."

That flip is the single most interesting fact in the KB read. A tracked source moving from "terminals running full because the world needs LNG" (conviction 65) to "wouldn't invest, management lacks control over inputs" (conviction 55) inside six weeks is not noise — it is a voice concluding that the business quality is lower than the business position. And it landed in early May 2026, immediately before the Q1'26 result (reported 2026-05-07) that produced the -$16.65 EPS. Net KB conviction: mixed-positive. Breadth 2.

The thematic lane — enormous, and genuinely double-edged.

The corpus carries 110 claims matching LNG-related terms, of which only four name Cheniere. The rest describe the commodity and the geopolitics, and the 2026 cluster is dominated by a Middle East supply disruption. Both directions are represented:

Supporting the demand case:

• invest_like_the_best, 2026-07-21 (bullish, conviction 82), entity ['natural gas']: "After 18 months of well-level study, US faces historic structural gas deficit from LNG exports (15→35 BCF/d) plus AI compute; upside price risk is convex and unbounded from ~$3.50 today."
• odd_lots, 2026-03-18 (bullish, conviction 82): "US LNG exports rising from ~10% to ~20% of US gas demand, fastest-growing demand wedge, adding multiple BCF/day out to 2030."
• odd_lots, 2026-03-18 (bullish, conviction 80): "The 'LNG glut 2026-27' call is wrong: facilities take 4 years to build, run full once built, and have no spare capacity — the system can't flex to shocks." Directly relevant: it says built terminals run full.
• money_of_mine, 2026-05-05 (bullish, conviction 70), entity ['LNG']: "Structural shift coming: Asian buyers (Japan, Korea, etc.) will diversify LNG sourcing away from crisis-hit Middle East toward Australia long-term." Note: toward Australia, not necessarily the US.
• money_of_mine, 2026-02-24 (bullish, conviction 60), entities ['WDS','LNG']: "Long-term LNG price likely $10-13/MMBtu vs consensus $9."

Cutting against it:

• invest_like_the_best, 2026-07-21 (bullish on gas, conviction 70): "In the coming shortage, gas could spike like Russia/Ukraine did ($8-10/MCF)… sees prices high enough to force shutting off US export cargos." This is the specific mechanism that would hurt Cheniere.
• geopolitical_cousins, 2025-12-24 (bearish, conviction 80): "Massive LNG oversupply coming online (Qatar, Australia, Canada, Mozambique, US) with weak import-side demand pushes gas prices down for years."
• doomberg, 2026-05-28 (bearish, conviction 65), entities ['natural gas','LNG']: "LNG/gas is least concerning risk: Qatar loss is tiny share, US export capacity racing, coal-switching absorbs it; US gas still given away, Europe hit hardest."
• money_of_mine, 2026-05-05 (bullish, conviction 55), entity ['coal']: "When LNG prices spike, price-sensitive Southeast Asian buyers can't afford it and revert to coal — and they've already begun shifting to coal in reaction to this crisis." Demand destruction at the margin.
• jordi_visser, 2026-07-25 (neutral, conviction 55, skill 2.0), entities ['oil','natural gas']: "Oil stocks rarely get multiple expansion because revenue is tied to the underlying commodity."

Read: the knowledge base is loudly bullish on American gas volumes and genuinely split on Cheniere. The bull case for the commodity is not automatically the bull case for the exporter, and the corpus contains the exact claim explaining why. Moderate conviction, two-sided, breadth 2 direct. That supports a position, not a large one, and not before the print.

6. Technicals — corrected, not broken

• Price $258.08 on 2026-08-04, -2.08% on the day, on 1.08M shares against a 2.17M average — light volume on a down day.
• -13.1% from the 52-week high of $296.91 (quote-based high: $300.89). +36.7% above the 52-week low of $188.83. Maximum drawdown from peak: -13.1% — a real correction, unlike ET's -3.8% or MPLX's -0.49%.
• Above the 50-DMA ($246.40) by 4.7% and above the 200-DMA ($233.08) by 10.7%. The trend structure is intact.
• RSI 44.9 — below the midline, neither oversold nor extended. MACD +3.44 — positive.
• Beta -0.008. Effectively zero market correlation, and the lowest absolute beta in this batch by a wide margin. Worth naming: this dataset reports Cheniere as essentially uncorrelated to the broad market, which for a $54B energy infrastructure name is unusual and, if durable, is a genuine portfolio-construction attribute.
• Relative performance: 12-month +9.4% against SPY +19.9% and QQQ +23.9%; 6-month +22.0% against SPY +9.5% (a strong stretch); 3-month -4.4% against SPY +5.1% (the recent fade).

The tactical read. This is the only name in the batch offering a meaningful pullback — 13% off the high with RSI at 44.9 and both major moving averages intact. That is a reasonable technical setup, and if the fundamentals were legible this would likely be a Stage-In. The reason it is not is §2, and the reason to wait is 2026-08-06. The $233 area (the 200-DMA) is close to the $224 bear anchor and is the level at which a starter position would carry a genuine margin of safety.

7. Moat and competitive position

The moat is permitting, capital and time, and it is wide.

A US LNG liquefaction and export terminal requires FERC authorisation, Department of Energy export approval, multi-billion-dollar construction budgets and — per the corpus's own framing (odd_lots, 2026-03-18, conviction 80) — "facilities take 4 years to build, run full once built, and have no spare capacity." Cheniere already owns two of them. $38.455B of net property, plant and equipment on $49.127B of total assets is the physical expression of a barrier that cannot be crossed quickly by anyone, however much capital they have.

Layered on top: long-dated contracts. Cheniere's model sells capacity forward on twenty-year sale-and-purchase and tolling agreements, which converts a volatile commodity into a fee stream. The evidence that this works is in Q1'26 — operating income up 30.6% in a quarter of extreme gas-market dislocation.

Two genuine weaknesses in the moat, both named honestly.

First, concentration. Roughly 95% of revenue from a single product at two sites (FY24 disclosure). An operational outage, a hurricane, or a regulatory action at Sabine Pass or Corpus Christi is a company-level event, not a segment-level one. Neither ET (nine states, six product lines) nor MPLX (two segments, marine plus pipe) carries anything like this concentration.

Second, input exposure — the point the tracked bear makes. compound_and_friends (2026-05-01, conviction 55): "margins are a function of the natural-gas price and management lacks control over its inputs." Cheniere buys feedgas and sells LNG. The tolling contracts are designed to pass that through. The marketing book is not. That gap between the contracted and the marketed volumes is where the gas-price risk lives, and this dataset does not disclose the split — which means it cannot be sized here.

Peer set (from this dataset): CQP (Cheniere Energy Partners, $67.00, $32.43B — Cheniere's own subsidiary partnership, the $5.163B minority interest on the balance sheet), KMI ($69.9B), ET ($69.8B — also in this batch), OKE ($55.6B), MPLX ($59.8B — also in this batch), TRP ($68.6B), BKR (Baker Hughes, $60.4B), plus FANG, IMO and SU. This dataset does not carry forward estimates for those peers, so no external peer-multiple table is offered. The clean within-batch comparison from §4 methodology: LNG at 7.67x FY27E EV/EBITDA against MPLX at 10.58x and ET at 6.41x — Cheniere sits in the middle, cheaper than the conservative partnership and dearer than the levered one, with 3.76x net-debt/EBITDA between MPLX's 3.41x and ET's 5.07x. That positioning is fair, and the absence of a valuation anomaly is part of why this is a Hold.

8. Verdict, kill-criteria and flip conditions

Hold. Cheniere owns the largest LNG export position in the United States at the front of what the knowledge base calls a historic structural gas deficit, trades at 7.67x FY27E consensus EV/EBITDA, generates $5.5B of annual operating cash flow, and is retiring roughly 5% of its equity a year — a 5.9% total shareholder yield that compounds regardless of whether the multiple ever moves. Twenty-seven analysts cover it with zero sell ratings and a $278.50 consensus. Base fair value is $273, or +5.8%.

Two things stop this being a Buy, and one of them is disqualifying on its own:

1. You cannot currently value this company on its reported earnings. Q1'26: operating income +$1.973B and up 30.6%; GAAP net income -$3.502B; EPS -$16.65 against a +$4.25 consensus; approximately $5.5B of losses below the operating line that this dataset does not identify. FY26E consensus EPS is negative $6.62222. The most probable explanation is a non-cash derivative mark that reverses — but that is inference, not data, and until the 2026-08-06 print resolves it, any position is being taken on faith about a nine-figure-plus line item.

2. The demand thesis is double-edged, and the knowledge base says so. The same conviction-82 source describing the gas deficit also warns of prices "high enough to force shutting off US export cargos" (conviction 70) — the specific mechanism a tracked bear cites in refusing to own it (compound_and_friends, 2026-05-01, conviction 55). More American gas demand is not unambiguously more Cheniere.

Add a base case worth only +5.8% and earnings in two days, and the answer is straightforward.

How to act on a Hold:

• Already own it: hold. A 5.9% shareholder yield on an irreplaceable asset with intact moving averages does not get sold into a print. Do not add before 2026-08-06.
• New money: wait for the print. Then either (a) initiate on a clean quarter — EPS at or above the $3.08 consensus with no repeat of the below-the-line loss and an explanation of Q1 — sizing to 1-2%; or (b) wait for a move toward $224-233 (the bear anchor and the 200-DMA), which offers a genuine margin of safety and a starting EV/EBITDA of roughly 7.0x.
• Sizing discipline: 1-2% maximum. The binding constraint is not valuation — it is that the reported earnings of this company are currently unreadable, and unreadable accounts justify small positions regardless of how good the assets are.

Pre-registered KILL / avoid-adding criteria:

• The 2026-08-06 print shows a second consecutive large unexplained non-operating loss. Run the same arithmetic as §2: operating income minus interest expense versus pretax income. If the gap reappears at multi-billion scale, the opacity is structural, not episodic, and this becomes an Avoid.
• Operating income decelerates from the Q1'26 pace of +30.6% year over year — the one clean signal in the accounts breaking.
• FY27E consensus EPS drifts below ~$13 (from $16.49119), or FY27E consensus EBITDA below ~$9.5B (from $10.263B).
• The buyback is cut materially from the ~$2.7B annual pace — that is ~60% of the base-case return disappearing. Watch cash, which has already fallen from $4.066B (FY23) to $1.584B (FY25).
• The 200-DMA (~$233.08) is decisively lost on earnings-related news rather than sector beta.
• Any credible evidence of export-cargo curtailment on domestic price grounds — the invest_like_the_best conviction-70 scenario becoming operative.

Pre-registered FLIP TO BUY:

• The 2026-08-06 print delivers EPS at or above $3.08 with no repeat of the Q1 charge and a clear explanation of it — earnings become usable again, the opacity discount closes, and 8.5x becomes defensible.
• Or the shares trade below ~$233 (the 200-DMA, roughly 7.0x FY27E EBITDA) with FY27E consensus EBITDA still at or above $10.263B — a 17% gap to base and the margin of safety the earnings opacity requires.
• Or FY28E consensus EBITDA is revised above ~$11.5B (from $10.595B), i.e. the street starts modelling the export wedge that invest_like_the_best (2026-07-21, conviction 82) describes rather than a 3.5% plateau.

Where LNG fits in the Synthos Framework Portfolio. The energy-infrastructure sleeve, as the export-terminal / capital-return leg — 1-2%, and not before 2026-08-06. On overlap within this batch, the three energy names are genuinely different instruments and should not be treated as substitutes. MPLX is the income anchor: a partnership, K-1, 7.11% covered distribution, 3.41x leverage, priced at 10.58x. ET is the valuation torque: a partnership, K-1, 6.58% distribution covered only 0.81x, 5.07x leverage, priced at 6.41x with the most upside to fair value. LNG is the corporate compounder: a C-corporation (1099, not K-1 — the only one of the three you can hold in a retirement account without UBTI concerns), a 0.84% dividend plus a 5%-a-year buyback, 3.76x leverage, priced at 7.67x, with beta -0.008. Its role in the sleeve is per-share compounding and market-correlation diversification, not income. Suggested construction: MPLX as the income anchor, ET as the valuation torque, LNG as the tax-simple, low-correlation compounding leg — with LNG's size held down until the accounts are readable. Logged as a tracked Synthos call (Hold) as of 2026-08-04 at $258.08.

Single biggest risk: the reported earnings are currently unreadable, and the reason is undisclosed. A company that produces $1.973B of operating income and a $3.502B net loss in the same quarter, with a ~$5.5B gap the data does not explain, cannot be underwritten with confidence at any price. This is not a claim that something is wrong — the cash flows, the operating line and the buyback all look healthy. It is a claim that the single largest input into the FY26 and FY27 earnings estimates is invisible, and that invisible inputs deserve small positions.

Single most fragile assumption in the price: that consensus EPS swings from negative $6.62222 (FY26E) to positive $16.49119 (FY27E) — a $23.11 per-share reversal — on schedule. Every earnings-based cross-check in §4 rests on that FY27 number. If the below-the-operating-line losses recur even once more, the reversal slips, the FY27 anchor moves, and a name with no readable earnings and a 3.76x-levered balance sheet has nothing left to stand on but its EBITDA multiple and its buyback. Check it on Thursday.

Provenance and disclosures

• Traceability: kb_claim_count 4 — all four name Cheniere directly, across 2 tracked sources, and they disagree. Bullish: all_in (2025-05-06, conviction 80, entities Cheniere Energy, Sabine Pass); compound_and_friends (2026-03-31, conviction 65) and compound_and_friends (2026-03-31, conviction 60). Bearish: compound_and_friends (2026-05-01, conviction 55). All quotes verbatim. Thematic lane cited verbatim from invest_like_the_best (2026-07-21, conviction 82 and conviction 70), odd_lots (2026-03-18, conviction 82 and 80), money_of_mine (2026-05-05, conviction 70 and 55; 2026-02-24, conviction 60), geopolitical_cousins (2025-12-24, conviction 80, bearish), doomberg (2026-05-28, conviction 65, bearish) and jordi_visser (2026-07-25, neutral, conviction 55, skill 2.0). Breadth 2 direct; net conviction mixed-positive. Note: a naive symbol sweep for "LNG" returns 110 claims; 106 of them describe the commodity or the geopolitics, not the company, and are treated as thematic context clearly labelled as such — never as coverage of the ticker.
• Data as-of: fundamentals 2026-03-31 (Q1'26; annual figures 2025-12-31) · estimates 2026-08-04 · prices 2026-08-04 (quote $258.08, 50-DMA $246.40, 200-DMA $233.08, 52-wk range $186.20-$300.89 per quote / $188.83-$296.91 per technicals) · KB claims swept 2026-08-04. All market data from the single pre-pulled FMP dataset for this ticker. Nothing was re-fetched, recalled from memory, or estimated.
• Data gaps and inconsistencies, stated explicitly: (1) The Q1'26 result contains approximately $5.5B of below-the-operating-line losses that this dataset does not identify — operating income +$1.973B less interest $255M equals +$1.718B, against reported pretax income of -$3.753B. This is the central analytical problem of the dive and is treated as such in §2. No cause is asserted. (2) FY26E consensus EPS is NEGATIVE $6.62222, so no FY26 P/E is computed; all valuation work uses EV/EBITDA and says so. (3) The reported annual EBITDA line is unreliable — FY22 EBITDA of $6.226B against operating income of $11.097B, and FY23 EBITDA of $17.538B against operating income of $7.643B, are not reconcilable through depreciation; operating income and revenue are used as the trustworthy annual series. (4) The FY22 product-segment row is corrupt — Liquefied Natural Gas of $112.0B against total FY22 revenue of $33.756B — and is excluded. (5) The most recent product segmentation is FY24, roughly nineteen months stale. (6) Geographic segmentation is effectively absent — only FY17 and FY16 entries exist — so Cheniere provides no usable current geographic segment reporting in this dataset, a real gap for an export business, and none is analysed. (7) The Q1'26 earnings-calendar revenue ($5.868B) differs from the income-statement figure ($6.650B) by $782M; the income statement is used. (8) Two different net-debt figures exist — $27.025B at FY25 year-end versus $24.639B implied by the TTM enterprise-value bridge; the TTM figure is used consistently, with a stated $11-per-share sensitivity. (9) FY29E and FY30E consensus rest on four analysts and show revenue and EBITDA declining; they are excluded from fair-value work. (10) Insider filings are director awards at $0 and one tax-withholding F-InKind disposition only — no open-market transactions — so no insider signal is claimed. (11) The reported beta of -0.008 is unusual for a $54B energy name; it is reported as found and flagged rather than adjusted.
• Corporate-structure note: unlike ET and MPLX in this same batch, Cheniere Energy, Inc. is a C-corporation — holders receive a 1099, not a Schedule K-1, and there is no UBTI issue in tax-advantaged accounts. The TTM effective tax rate is 27.4%. Note that Cheniere Energy Partners (CQP) is a separately traded partnership subsidiary — the $5.163B of minority interest on the FY25 balance sheet — and is a different security with different tax treatment.
• Fair-value caveat: the $224 / $273 / $322 anchors are scenario multiples on FY27E consensus EBITDA of $10.263B (7.0x / 8.0x / 9.0x), converted to equity by subtracting net debt of $24.639B (TTM bridge) and dividing by 210.5M diluted shares, then cross-checked against FY27E consensus EPS of $16.49119. They are scenario arithmetic, not a discounted cash flow. Using FY25 year-end net debt of $27.025B instead lowers each anchor by approximately $11.
• Timing caveat: this dive is written two days before the Q2'26 print (2026-08-06, consensus EPS $3.08 on revenue $4.8399B) — deliberately, because that print is the direct test of the dive's central open question. Both entry triggers are structured around it.
• Not investment advice. Independent research, educational and informational only, never personalized.
• Version: 2026-08-04-full.