Energy Vault filed its 10-K today. The question isn't whether grid-scale storage is a good business — it is. The question is whether NRGV's specific pivot from EPC contractor to infrastructure owner generates enough recurring EBITDA to bridge a coming revenue gap before the company burns through its freshly raised capital. That answer isn't in the filing. It's on May 11.

The Filing in Four Numbers

FY2025 revenue: $203.7M — a 340% increase from $46.2M in FY2024. Gross margin improved from 13.4% to 23.6%. Operating cash burn was ($5.6M), near breakeven for the first time. Backlog tripled to $1.3B.

Those numbers look strong. The context strips them back.

The revenue surge is almost entirely two Australian EPC customers. 61% of revenue is Australia. Two customers account for 88% of the total. The Australia revenue went from $1.0M in FY2024 to $124.3M in FY2025 — a 12,300% increase in one year — because a handful of large EPC projects ramped. When those projects complete, that revenue largely goes away.

The near-breakeven operating cash flow is partly illusory: $60M of it is accounts payable/accruals building up on large construction projects. That unwinds when projects settle. The backlog includes $490M in "contingent option backlog" on a single project (Ebor NSW) that hasn't been contracted. Strip it out and contracted backlog is ≈$815M — still tripled from last year, but not $1.3B.

What Changed in the Capital Structure

FY2024: zero long-term debt. FY2025 (as filed, December 31): $94.6M in debt that didn't exist twelve months prior. Then in February 2026, after the fiscal year ended, the company raised $150M in 5.25% Senior Convertible Notes due March 2031.

The path matters. May 2025: $10M bridge loan from Crescent Cove at 24% interest. That's a distress-tier cost of capital — they couldn't access normal credit. September through December 2025: three tranches of convertible debentures from Yorkville Advisors totaling $65M, with floor conversion prices at $0.60/share. That's the structure a company uses when institutional equity markets are closed to it.

Then February 2026: $150M in senior converts placed with institutional investors at 5.25%, $5.18/share conversion, capped calls at $8.12. That's a different market — and it used $49.7M of proceeds to pay off the Yorkville tranches.

The distress path is blocked for five years. Pro forma unrestricted cash is approximately $129M. That matters for how we size the bear case.

This isn't a clean capital structure, though. The common equity waterfall is: $150M senior converts, ≈$13.5M remaining Yorkville principal, $300M OIC preferred equity (return hurdle senior to common), $21M redeemable NCI, and then you. The $300M OIC commitment sounds bullish. It isn't bullish for common shareholders — it's preferred equity in a subsidiary, with OIC's returns senior to NRGV's common. The ITC monetization, the O&O asset cash flows, OIC's hurdle rate — that all gets paid before common equity sees a dollar.

The Own & Operate Thesis

The bull case is the pivot. NRGV is transitioning from "build and transfer" to owning the assets it builds, collecting recurring cash through long-term offtake contracts and monetizing Investment Tax Credits (ITC) via transfers to tax-equity buyers.

Two assets are operational: Cross Trails (57MW/114MWh, Snyder TX, commercial since May 2025) and Calistoga Resiliency Center (8.5MW hybrid hydrogen/lithium-ion, Calistoga CA, commercial since September 2025). Cross Trails has a 10-year physically-settled revenue floor contract with Gridmatic, an AI-enabled power marketer — described as the first such structure in ERCOT. These are real assets generating real ITC: $47.7M in FY2025 ITCs sold at 85 cents on the dollar, $11.8M collected on Cross Trails in February 2026.

Behind those: SOSA Energy Center (150MW/300MWh, Madison County TX, construction started January 2026, guided commercial operation Q2 2027), Stoney Creek (125MW/1GWh, NSW Australia, acquired August 2025, development stage, no project financing confirmed in the filing), and Ebor (100MW/870MWh, NSW, 14-year LTESA awarded February 2026, operations expected 2028). Management has guided $40M O&O EBITDA by year-end 2027, $100-150M by year-end 2029.

Orion Infrastructure Capital ($3B AUM generalist fund) committed $300M in preferred equity to the Asset Vault subsidiary, with $35M deployed at Series A. OIC's warrants are struck at $4.24/share. This is genuine institutional validation — OIC doesn't invest for charity, and $300M is approximately 10% of their AUM, which is a real conviction bet. The catch: they're a generalist infrastructure fund, not a BESS specialist. Asset Vault is their only grid-scale BESS platform investment.

The Cross-Ticker Reality Check

Australia's boom is real. It is also not NRGV's. The federal Capacity Investment Scheme is driving $14B+ of storage investment through 2030. Australia commissioned 4.9 GWh in 2025 alone — matching its entire 2017-2024 cumulative total. That demand exists and is growing.

Fluence Energy signed AGL Tomago (500MW/2GWh, NSW) and AMPYR Wellington (300MW/600MWh, NSW) in mid-2025. Fluence CEO called out Australia's 7-to-51-GWh trajectory explicitly. FLNC has institutional scale, a global footprint, and is now competing directly in the same market where NRGV generated 61% of its FY2025 revenue.

NRGV has 142 employees. The Australia revenue surge is not a moat — it's a project timing advantage in a market that is now contested by a better-resourced global operator. The next round of NSW tenders will not go uncontested.

The ITC discount is another signal. NRGV is selling credits at $0.85 when non-investment-grade sponsors are getting $0.91 in the market. That 6-cent gap on $47.7M of credits costs roughly $2.9-4.8M. The most likely explanation: purchasers are pricing in FEOC compliance risk. OBBBA, enacted July 4, 2025, added 10-year clawback provisions to energy storage ITCs for projects with Chinese-origin supply chains. NRGV's battery supply chain is predominantly Chinese-origin. Management says this is "not expected to be material." ITC buyers are already disagreeing.

Factor Decomposition

Two factors dominate this thesis and they are anti-correlated.

F1 — Own & Operate EBITDA ramp (≈40% of thesis weight): Everything depends on this. Cross Trails and Calistoga are operational. SOSA is under construction. The question is whether the 142-person company can deliver SOSA on time (COD Q2 2027), close Stoney Creek project financing, and exercise the Ebor option — all while managing the transition from EPC mode.

F2 — Australia EPC revenue continuity (≈35% of thesis weight): The current projects provide the revenue bridge while F1 ramps. When they complete, F2 falls sharply unless new contracts are won against Fluence.

The anti-correlation: when EPC work is heavy, the same team and management bandwidth are consumed delivering it. Building an O&O portfolio requires a different skill set than EPC delivery. The company can't fully do both simultaneously at 142 employees. F2 high means F1 builds slowly. F2 gaps means F1 must carry the company before it's ready.

The bull scenario requires F2 to bridge for 12-18 months while F1 ramps. The base scenario is the gap appearing before F1 has enough scale. The bear scenario is F1 delayed while F2 gaps, and the $150M in converts starts looking like a fixed liability on a shrinking revenue base.

F3 — ITC integrity: Correlated with F1 (more O&O assets = more ITC). The FEOC/OBBBA risk is real but not existential unless compliance fails on multiple projects. The 85-cent pricing tells you purchasers are watching.

F4 — Capital structure survival: With $129M pro forma cash and a $13M quarterly G&A burn (non-GAAP), that's approximately 10 quarters of runway plus $25-30M in remaining FY2025-vintage ITC collections pending. The existential scenario is off the table through late 2028. Beyond that, $150M due in 2031 against a $67M market cap is a real refinancing question if the equity doesn't recover.

What the Market Is Missing

At $3.59 the market backward-solves to approximately: Bull 15% / Base 50% / Bear 35%. That 35% bear probability is stale — it was reasonable before the $150M convertible notes closed in February 2026. The existential path required either toxic convert rollover or a distressed equity raise. Both are blocked for five years.

We assign: Bull 25% / Base 50% / Bear 25%. The 10-point reallocation from bear to bull is driven by: (1) the capital structure improvement, (2) four insiders purchasing open-market in a 10-day cluster (CEO, COO, two directors, February 26 to March 4 — people with access to forward project status and the Q1 revenue trajectory), and (3) the 10-K content published today that confirms Q4 2025 was the first profitable quarter ($0.02 actual vs. -$0.02 consensus).

Probability-weighted EV at our distribution: $4.44 vs. $3.59 entry. That's +23.6% expected return, with +17 percentage points of alpha over the market's implied EV. After removing market and sector factors, the annualized idio alpha is approximately +2.7% — modest. The case isn't the carry. It's the probability misprice.

Today's market also showed a 3.3x volume spike (11.6M shares vs. 3-month average) on the 10-K filing date. The P/C ratio is 0.05 — 21.9x calls to puts, with heavy open interest at the $4 strike expiring tomorrow. Someone is positioned for a near-term move.

Entry and Exits

Entry range: $3.40-3.70. Below OIC's $4.24 warrant anchor. Below the $5.18 convertible conversion price. Both institutional reference points say the current price is below where sophisticated capital was placed.

Hard stop: $2.50. Below that, the base case intrinsic value is approached and the bear path is probably already in motion.

Soft stop — May 11 Q1 print: If revenue is below $30M with no new EPC contracts announced and no O&O EBITDA contribution visible, exit on the print. That combination means the bridge was cut before the model was ready. The $150M converts are then a fixed liability on a shrinking revenue base and the bear probability reasserts.

Key catalysts in sequence:

• March 20 (tomorrow): Options expiry, heavy call OI at $4 — gamma dynamics could pull toward $4 into expiry
• April-May 2026: Remaining CRC and Snyder CDU ITC transfers close (≈$25-30M cash)
• May 11: Q1 2026 earnings — the binary. Revenue trajectory post-Australia ramp becomes visible
• Q2 2026: SOSA construction update, potential new EPC contract announcement
• Q3 2026: Stoney Creek project financing confirmation or absence

Conclusion

The Own & Operate pivot is real and has institutional backing. The first two assets are operational. The ITC mechanism works. The capital structure is no longer existential. Four insiders bought in front of this filing.

Against that: 142 employees can't execute a 1.5 GW ambition without something breaking. Australia revenue is a sector boom that Fluence is now in. SBC dilution is 55% of market cap annually. Common equity is last in the waterfall. FEOC risk on the ITC engine is underpriced.

The doorway is genuine. One path works; one doesn't. May 11 is where it collapses to one interpretation. The entry is now — below both institutional anchor prices, on the day the 10-K was filed, with the insider signal still fresh and the F6 window still open.

If May 11 shows the bridge intact, size up. If it shows the gap, exit. Don't hold through a second quarter of ambiguity.

Evidence