FIP$4.67-3.9%Cap: $552MP/E: —52w: [==|--------](May 12)
FTAI Infrastructure (FIP) signed an equity purchase agreement on April 29, 2026 to sell Long Ridge Energy & Power to MARA Holdings for $1.52B. The Q1 2026 10-Q (filed May 8) removed going concern language, confirmed Jefferson 2024B refinancing commitments, and posted Railroad Adj EBITDA of $40.2M, up 102% versus Q1 2025 pro forma. The stock closed May 12 at $4.67 — the exact price at scenario authoring March 16. Three structural overhangs compressed to one in seven weeks. The tape did not move.
What the filing says
The Long Ridge transaction (FIP 8-K May 7, FIP 10-Q May 8, MARA 10-Q May 11) is an equity purchase of 100% LLC interests at $1.52B base. MARA assumes or repays existing debt (≈$546M of loans plus $208.9M of electricity swap liabilities), backstopped by a $785M Barclays bridge committed in MARA's 10-Q. Net proceeds to FIP "in excess of $300M." Required approvals: HSR, FERC change-of-control, and an 8.75% Senior Notes consent solicitation. Drop-dead Nov 30, 2026 (extendable to Jun 30, 2027 if regulatory only); $75M termination fee payable to FIP if MARA cannot close.
Capital structure cleared: Term Loan covenant compliance confirmed; only $25.4M of debt due in next 12 months once Jefferson 2024B is addressed. The 10-Q replaces going concern language with "Management's planned actions, including the sale of Long Ridge, are considered probable to be implemented and to provide sufficient liquidity… to meet its obligations as they become due." On Jefferson 2024B ($218M, 10%, July 1 maturity), Nicholson confirmed on the call "we received commitments for the refinancing of a little over $200 million" — terms not yet disclosed; the $255M Backstop bridge remains available as fallback.
Operationally, Railroad doubled YoY to $40.2M Adj EBITDA as W&LE diluted USS concentration from 41% to 22% of segment revenue. Jefferson Terminal printed $14.4M (+81%) with three contracts targeting $50M of incremental annual EBITDA. The bear column: the Ares RailCo Series A preferred accreted to $970.5M from $937.6M, recording a $32.2M quarterly charge ($27M PIK plus $5.2M accretion) on a path to a $1.5B mandatory redemption floor by approximately 2029.
What the market thinks
Spot $4.67 sits at the 19th percentile of the 52-week range ($3.90 to $7.93). Short interest is 23.4% of float at 19.1 days to cover — up from 19.0%/15.0 days at the March worldview snapshot. Shorts added through the May 7 sale announcement. ATM IV is 208% at the 255th percentile, priced for binary distress. Sell-side mean target is $11.67 from three covering analysts; BTIG cut its target from $10 to $9 on May 8 — the sale-announcement day. The CEO's May 2025 open-market purchase ($2.6M at $5.22, code P) is underwater. Back-solving a rough bull/bear binary against current spot puts market-implied P(bear) in the 50-55% band.
Why the gap exists
The interesting fact is that BTIG cut on the sale day. The expected sell-side response to a $1.52B sale plus going concern removal plus refi commitments is to raise targets; one of three analysts did the opposite. That implies they re-modeled the post-Long-Ridge entity and found less value. The most plausible vector is the Ares preferred: $128M annualized charge, mandatory $1.5B redemption floor by approximately 2029, noncompliance triggers not publicly specified. After Long Ridge proceeds clear Term Loan, Ares becomes the largest single capital cost. A model that previously gave full credit to deleveraging may now see less of that delta accruing to common equity. Without BTIG's note text the rationale is unverified — but the cut combined with growing short interest is what keeps the stock at scenario-authoring price while the worldview Log Bayes moved from 2.70 to 7.10.
The alternative reading is slower: structural-distress stigma takes time to clear, sell-the-news positioning unwinds slowly, and FERC clearance plus a clean Long Ridge close compresses the gap. Both readings are consistent with the current tape. The trawl prefers the slower reading because the evidence-weight shift is too large to be priced at zero, but cannot dismiss BTIG without the underlying model.
Risks (ranked)
The Ares preferred is the structural problem. Post-Long-Ridge it is the largest single capital cost — $128M annualized charge against a deleveraged free cash flow base. Mandatory $1.5B redemption by approximately 2029 forces either refinancing, asset-sale-against, or dilution; noncompliance triggers and PIK compound mechanics are not publicly disclosed.
MARA execution is next. The Barclays bridge is committed but not funded; MARA is the cohort laggard (1-year stock -16% versus RIOT +191%, IREN +634%, APLD +740%, KEEL +294%); its "indications of interest" from hyperscalers are IoI not LOI — the same language MARA used for its Starwood JV in February, which still has no named tenant fifteen months later. The $75M break fee provides partial compensation if MARA walks but reputational and re-marketing cost is real.
Below those: Long Ridge Senior Notes ($600M, 8.75%, 2032) carry CoC put rights — a consent fee above approximately 1% of principal would compress MARA's economics. Jefferson refi terms are unknown until the closing 8-K. Gary Works BF #14 reline (May-August) is modeled at $4-6M of Q2/Q3 Railroad EBITDA impact and was not addressed on the Q1 call.
Catalysts
The seven-day window: Jefferson 2024B closing 8-K. The thirty-day window: FERC application acceptance (clock starts), first Form 4 post-Long-Ridge. The ninety-day window: Jefferson 2024B maturity July 1 (the Jefferson refi resolution test), Q2 print early August (tests BF #14 absorption and Jefferson contract progress). The hundred-fifty-day window: FERC clearance and the Long Ridge close 8-K. Beyond that: rail M&A 8-K Item 1.01 — FIP has four opportunities in pipeline since Q4 2025 and UP has committed TRRA divestiture as a UP-NS merger condition; the Q3 print in early November; and the FY2026 10-K in early March 2027 which tests material-weakness probability.
The cleanest deadline-bound test is whether FIP closes at or above $6.00 on any session on or before September 30, 2026 — the FERC-clearance window resolution, with a 55% probability estimate.
What would change our mind
Toward bullish: CEO or CFO Form 4 code P at the $4-5 range post-sale — the strongest possible insider signal that the dip is the opportunity, given the May 2025 buy at $5.22 was before this de-risking. FERC application accepted within thirty days with no second request. A named hyperscaler tenant for Long Ridge collapses the IoI/LOI gap and validates the $3.13M/MW campus optionality premium. BTIG raising back to $10+ with rationale would suggest the May 8 cut was idiosyncratic.
Toward bearish: BTIG's note rationale revealing an Ares-driven valuation drop the trawl underweighted. MARA insiders selling, or MARA equity issuance to fund the close. FERC second request. Senior Notes consent fee greater than approximately 1% of principal. Q2 Railroad EBITDA below $30M. Any Ares noncompliance language in subsequent disclosures.
Evidence
| Evidence | Source | Credibility | LR |
|---|---|---|---|
| Long Ridge $1.52B equity purchase agreement to MARA, ≈$546M debt repaid/assumed, $75M break fee, mid-Q3 close guided | FIP 8-K 2026-05-07; FIP 10-Q 2026-05-08; MARA 10-Q 2026-05-11 | 0.95 | 2.5 |
| Going concern language removed; "Management's planned actions, including the sale of Long Ridge, are considered probable" | FIP 10-Q 2026-05-08, Note 1 / MD&A Liquidity | 0.95 | 1.8 |
| Jefferson 2024B ($218M, 10%, Jul 1 2026): "We received commitments for the refinancing of a little over $200 million" | FIP Q1 2026 earnings call, Nicholson prepared remarks | 0.85 | 1.6 |
| Term Loan covenant compliance confirmed; only $25.4M debt due in next 12 months | FIP 10-Q 2026-05-08, Note 7 Debt | 0.95 | 1.5 |
| Cross-cohort miner→AI pivot complete: MARA, RIOT, BITF→KEEL all formally repositioned May 2026 | MARA 10-Q 2026-05-11; RIOT 8-K 2026-04-30; BITF→KEEL 8-K 2026-05-11 | 0.95 | 1.5 |
| Railroad Q1 Adj EBITDA $40.2M vs $19.9M PY pro forma (+102%); USS concentration 41% → 22% | FIP 10-Q 2026-05-08, segment results | 0.95 | 1.4 |
| Long Ridge $3.13M/MW = 3.6-4.2× recent gas-fleet M&A comps ($738-985K/MW NRG/Vistra/Calpine); DC campus optionality premium | MARA 10-Q 2026-05-11; NRG Jan 2026 close; Calpine/Vistra disclosures | 0.92 | 1.4 |
| Rail M&A wave three-driver thesis; UP-NS merger refile, UP committed TRRA divestiture | FIP Q1 2026 call; STB filings; NSC Q1 2026 | 0.85 | 1.4 |
| Counter-tape: worldview Log Bayes 2.70 → 7.10; stock unchanged at $4.67; short interest 19.0% → 23.4% through positive news | FIP filing trawl analysis 2026-05-12; market data | 0.85 | 1.3 |
| Total equity -$303M (worse than -$146M Dec 2025); accumulated deficit -$625.9M; GAAP loss driven by interest + preferred | FIP 10-Q 2026-05-08, balance sheet | 0.95 | 0.85 |
| Ares preferred $970.5M carrying ($32.2M Q1 charge: $27M PIK + $5.2M accretion); $1.5B mandatory redemption floor by ≈2029 | FIP 10-Q 2026-05-08, Note Preferred Equity | 0.95 | 0.8 |
| BTIG cut FIP price target $10 → $9 on May 8 (sale-announcement day); peers at mean $11.67 | BTIG analyst note 2026-05-08; public price target tracker | 0.75 | 0.7 |
// comments (0)