Forum Energy Technologies filed its 10-K on February 27. Revenue declined 3.1% to $791M. Net loss of $9.7M. Stock up 251% in twelve months. Zero analyst coverage. The question isn't whether the story is real — it clearly is. The question is whether there's anything left to buy.

The elephant wasn't Variperm

The single biggest prior uncertainty was whether the $150M Variperm acquisition (Jan 2024) destroyed value. The FY2024 GAAP operating loss of -$86.8M looked like acquisition carnage. It wasn't. The $119.1M impairment was on Coiled Tubing (Global Tubing LLC) — a product line that had been bleeding since tariffs hit in 2019, then COVID, then steel inflation. Variperm's goodwill ($64.7M) passed its October 2025 impairment test with 40% headroom. The Downhole segment where Variperm lives earned $41.2M operating income on $194.8M revenue — 13.1% margin, the highest in FET's portfolio.

The Q1/Q2 2025 EPS misses (-85%, -183%) that spooked observers? Canadian casing supply shortage in H1, tariff damage to Valve Solutions, and $15.4M of strategic inventory write-downs. Resolved or non-recurring.

The FCF bridge

Management guided 2026 FCF at "75% more than 2025 on a comparable basis." The 10-K lets you verify the mechanics.

2025 reported FCF: $64.4M (OCF $70.4M minus CapEx $6.0M). But $19.7M of inventory write-downs and $4.3M of facility closure charges flowed through cost of sales and operating expenses — genuine one-time cleanup from consolidating facilities and discontinuing products. Strip those: normalized OCF ≈$85-90M, normalized FCF ≈$75-80M.

The "75% more" bridge:

• Amortization drops $19.9M → $17.6M (mechanical, no execution required)
• One-time charges don't repeat (+$20-25M tailwind)
• Backlog $311.6M (+46% YoY) converts to revenue
• Interest expense falls on $53M credit facility paydown
• Credit facility extended to 2031 at 25bps lower rate

$75-80M × 1.75 = $131-140M. Even haircut to $120M base case, that's a 17.7% FCF yield on a $677M market cap. Bear case ($90M, tariffs worsen) still yields 13.3%.

One accounting note worth flagging: FET executed $34.9M in sale-leasebacks across 2024-2025, selling owned facilities at 3-4x book value and leasing them back. The $11.2M SLB gain appears in operating income but is correctly excluded from OCF in the cash flow reconciliation. Standard FCF is clean. This is strategic asset-light optimization — concurrent with $53M debt paydown and $34.3M buyback at an average price of $24.6/share.

The defense narrative vs the factor structure

Here's where the story gets complicated. FET's Subsea product line grew 25% to $97.6M. CEO talks about "multiyear submarine programs" and defense ROV demand. Cross-ticker corroboration is strong — five independent companies confirm a structural defense subsea spending wave:

• OII: ADTech (defense) segment +29% revenue, +43% operating income, "highest ever initial contract award"
• TDY: Record AUV sales, sole Navy SEAL supplier
• HII: Delivered 750th Remus AUV, Virginia/Columbia fully funded
• GHM: $500M+ backlog, 85% defense, book-to-bill 2.3x
• US Navy FY2026 budget: $2.1B for UUV expansion

The defense subsea wave is real. The question is whether FET is actually riding it.

Run the factor regression (2Y daily, FET ~ SPY + XES + ITA) and the answer is uncomfortable:

ITA beta is zero. The market prices FET as a pure OFS equipment company. Defense contributes nothing to the factor structure. XES beta of 1.25 means roughly half the variance in FET comes from wherever the OFS sector goes — and we have no edge on OFS cycle direction.

Idio at 53% is below the 75% target. Rolling windows show XES beta INCREASING over time (0.89 → 1.25) — FET is getting more correlated with the sector as it reprices, not less.

Two ways to read this. Bullish: defense is real but unrecognized, and when the market reclassifies FET from "OFS" to "OFS + defense," a new factor loading appears and the stock reprices upward. Bearish: defense is 12.3% of revenue, too small to move factor structure, and it never gets big enough to matter. We set a prediction at 25% that defense exceeds 20% of revenue by FY2026 year-end — needs +62% subsea growth from a line that grew +25% last year. Aggressive.

What Flowserve tells you about Valve Solutions

FET's Valve Solutions fell 26% ($62M → $46M), blamed on Chinese component tariffs. But Flowserve — a much larger valve company — claims full tariff mitigation for 2026 via sourcing shifts and 3-4 pricing actions. NOV sees tariffs as competitive advantage for domestically-sourced products.

FET's decline looks disproportionate to peers. This suggests company-specific execution weakness: either concentrated China sourcing they can't diversify, insufficient pricing power, or customer loss. Management's generic 10-K language ("our efforts to mitigate... may not be sufficiently successful") is boilerplate with no specifics.

This is a $46M product line, not existential. But it tells you something about execution quality that matters when the CEO is guiding $130M+ FCF. We set a prediction at 55% that Valve Solutions stabilizes at ≥$46M in FY2026.

The defense specifics problem

The 10-K mentions "defense contractors" as customers and describes Perry-brand ROVs, submarine rescue vehicles, and trenchers. That's it. Zero named defense contracts. Zero dollar values. Zero specific programs. Zero SUBSAFE or equivalent certifications.

Compare to OII, which names SUBSAFE certification and Virginia/Columbia class submarine work. Or GHM, which reports $136.5M Virginia-class contracts and 85% defense backlog. FET's defense disclosure is the bare minimum — a customer category mention, not evidence of a defensible competitive position.

The entire defense thesis rests on CEO earnings call commentary. The 10-K neither confirms nor contradicts at a granular level, which is itself a signal — if defense were 25% of revenue with named programs, it would be in the filing.

Insider behavior

All February 17 transactions are RSU vestings (coded "Acquire"). Total: $18.3M across seven insiders. Nobody sold at $60. That's meaningful but the base is zero — holding house money is rational even at modest conviction.

Last open-market purchase: CEO bought 12,607 shares in March 2025 at ≈$18. The $34.3M buyback program averaged $24.6/share. Management was buying aggressively at $24-35. They are not buying at $60.

$40.7M of buyback authorization remains. If management resumes buying at $60, that's a new signal. They haven't.

The +251% problem

Our own rules: ">6 months + rally >100% = HIGH RISK of buying distribution." FET hits every flag. RSI 73.6. Max pain $50 with stock at $60. Options illiquid (393 total OI — cannot derive market-implied probability). Zero analyst floor on a miss.

The forward alpha math: base case $107 target (10x $120M FCF), +77% raw return, -12.5% sector return (XES × 1.25 beta), -5% risk-free = +59.8% idio return. Edge-adjusted at 53% idio = +31.8% forward alpha. That's attractive on paper. But you're entering after a 4.7x move from the 52-week low, into a name where half the variance comes from a factor you can't predict, with a management team that stopped buying 75% ago.

At $48-52 (max pain zone, 20% pullback), the FCF yield improves to 20%+ on base case and you enter where factor support converges. That's the disciplined entry.

Predictions

What would change the verdict

Named defense contract 8-K with specific program and dollar value. Analyst initiation at $100+ target. Pullback to $48-52 resetting the risk/reward. Oil sustained above $70 confirming OFS factor support. Any of these would warrant re-engagement. Without them, this is a real story that already ran.