HLLY — Holley Inc. · 2026-09-07 · Verdict: WATCH · Conviction 3

Price $3.05 (last close 2026-09-04, screen row in universe_under2b.csv refreshed 2026-09-07; no live prices available) · Mkt cap $369.8M · EV ~$828M · EV/EBIT 9.9x (screen, CY2025 EBIT $82.5M) · FCF yield 9.2% (screen) · Net debt $(458.7)M · ADV $2.87M Sources read: 10-K 2026-03-16 (Items 1, 1A, 7), 10-Q 2026-08-05 (Item 2), DEF 14A 2026-03-20, 8-Ks 2026-05-06 (Items 2.02 and 5.07), 2026-05-18, 2026-05-26, 2026-07-14, 2026-08-05, Form 4s (12 filings, trailing 12 months). No call transcript or prepared remarks were filed as an exhibit; the bundle's transcript file is the Q2 press release (EX-99.1), so section 5 works from written management statements only.

1. What the business actually does

Holley designs, manufactures and sells performance parts to car and truck enthusiasts: fuel injection, carburetors, tuners, exhaust, ignition, superchargers and motorsport safety gear (10-K 2025, Item 1). It runs over 65 brands, with the top five at 49% of 2025 sales (Holley EFI 17%, Holley 11%, MSD 10%, Flowmaster 6%, Simpson 5%). Sales go direct-to-consumer and through distribution partners, where e-tailers and warehouse distributors were 54% of 2025 sales, the top ten partners 42%, and the single largest 22%. It employed 1,407 full-time staff, has spent an average of $23.7 million a year on R&D since 2020, and came public through a 2021 SPAC merger with Empower Ltd.

2. Why it looks mispriced, and the honest limit of the edge

The label is a leveraged deleveraging stub with an indifferent shareholder base, not a proprietary insight. Two private equity firms hold 33.6%: Holley Parent Holdings (Sentinel) 20.4% and MidOcean 13.2%, Sentinel's latest 13D/A filed 2025-09-19 (DEF 14A 2026, Beneficial Ownership). Neither is a long-term holder, and that overhang plus $2.87 million of daily volume keeps institutions away. Screens show 5.45x net debt to EBIT on 121 million shares at $3, which excludes the name from most quality filters, while underneath the business earns 41-43% gross margins on about 2% of sales in capex, so every dollar of debt repaid transfers to a $370 million equity. The sponsors sell for fund-life reasons and the screens react to a ratio rather than to cash conversion. That is a real setup, but it is the textbook levered-stub trade, and it only works if EBITDA stops falling.

3. Unit economics and growth

FY2025 sales were $613.5 million, up 1.9%, but price added about $16.0 million while volume subtracted about $4.7 million (10-K 2025, Item 7). Gross margin rose to 43.4% from 39.6%, adjusted EBITDA to $124.0 million from $110.5 million, and free cash flow fell to $34.2 million from $41.8 million.

The 2026 half sharpens the point. Q2 sales grew 3.2% to $172.0 million, made up of $4.7 million from the HRX acquisition and about $10.0 million of price against about $9.4 million of lower volume (10-Q Q2 2026). For the half, volume fell about $18.6 million while price plus HRX added about $18.2 million, leaving sales flat at $319.3 million. Gross margin slipped to 41.0% from 41.7%, SG&A rose to 23.5% of sales from 19.8% on class-action legal fees and consulting, and adjusted EBITDA fell to $33.8 million from $36.4 million, a 19.6% margin against 21.9%. R&D was cut 15.7% for the half to $7.7 million, an odd economy at a company whose stated edge is innovation.

Cash conversion is weaker than the headline. H1 free cash flow of $34.6 million includes roughly $11.1 million of one-time IEEPA tariff refunds received after the Supreme Court's 2026-02-20 ruling (10-Q Q2 2026), so ex-refund it is about $23.5 million on $61.1 million of adjusted EBITDA. The 2025 gap was wider: $124.0 million of adjusted EBITDA became $34.2 million of free cash flow after roughly $50 million of interest, cash taxes and the restructuring and notable items adjusted EBITDA excludes.

4. Balance sheet and capital allocation

At 2026-06-28 there was $527.7 million drawn under the November 2021 Credit Agreement, $69.0 million of cash and $97.5 million available on a $100 million revolver (10-Q Q2 2026). Credit agreement leverage was 3.74x against a 5.00x maintenance covenant (Q2 press release; 10-K 2025, Item 1A). It moved only from 3.75x at year-end 2025, because credit agreement EBITDA fell from $131.3 million to $129.0 million while debt came down. The bundle's excerpts do not disclose the term loan maturity, which is a diligence gap.

Management prepaid $15.0 million on 2026-07-13, taking voluntary prepayments to $115 million since September 2023, funded from free cash flow (8-K 2026-07-14). Because permitted cash in the covenant is capped at $50 million and the company held $69 million, that prepayment does cut covenant net indebtedness, to about $468 million. Reaching the sub-3.5x year-end target from there needs credit agreement EBITDA near $134 million, about $5 million above the current trailing figure.

A $25 million buyback was authorized 2026-05-26, with $2.0 million used and $23.0 million left at quarter-end (8-K 2026-05-26; 10-Q Q2 2026). Shares outstanding still rose about 0.6% year over year to 121.2 million, so equity compensation is out-diluting the repurchase. Insider Form 4s over twelve months show zero open-market purchases, ten grants and six tax-withholding dispositions.

5. What they said versus what they did

CFO Weaver argues Q2 adjusted EBITDA was "approximately flat year-over-year" once the prior year's non-cash tariff capitalization benefit is removed, and reiterated the guide (Q2 press release, CFO). CEO Stevenson said in May that "our share price does not fully reflect the strength of our operating model" (8-K 2026-05-26). Debt really has come down $115 million and the 2024 material weakness was remediated as of 2025-12-31 (10-K 2025, Item 1A). Against that, bonuses run on adjusted EBITDA targets, and for 2024 the board found the threshold was not met and paid a discretionary 50% award anyway (DEF 14A 2026); the CEO's salary rose from $700,000 to $800,000 on 2025-12-10 with the stock near $3. At the May 2026 meeting, sponsor director James Coady drew 29.4 million withheld votes against 59.3 million for, and the share-plan increase drew 27.6 million against on 88.6 million voted (8-K 2026-05-06, Item 5.07). The general counsel departed 2026-05-15 (8-K 2026-05-18). Debt discipline is real; pay discipline is not.

6. Valuation

Guidance is $610-640 million of core sales, $127-137 million of adjusted EBITDA, $42-47 million of interest and $15-20 million of capex (Q2 press release). Net of H1's actual $61.1 million, that requires H2 adjusted EBITDA of $65.9-75.9 million against $60.3 million in H2 2025, so 9% to 26% growth after a first half down 4.1%, at a margin near 23% against 19.1% in H1. Management's own 10-K says first-half sales are normally the higher half (10-K 2025, Item 1). The case is that hockey stick.

Two-year horizon to end-2027. Base: adjusted EBITDA settles at $130 million, 7.0x, about $60 million of cumulative free cash flow to debt, so EV $910 million less about $399 million net debt, roughly $4.30 a share. Bear: discretionary demand rolls over, adjusted EBITDA falls to $105 million at 6.0x with little paydown, about $1.55, with covenant leverage near 4.2x and interest coverage 2.3x. Bull: the step-up proves structural, 2027 adjusted EBITDA $145 million at 8.0x, about $6.55. Weighted 50/30/20 gives about $3.95 against $3.05, roughly 29% upside.

Reverse DCF: at $3.05 the roughly $828 million enterprise value against 2025 unlevered free cash flow near $95 million (adjusted EBITDA $124.0 million less $12.3 million capex less about $17 million cash taxes) implies a perpetual growth rate of about negative 1.5% at a 10% cost of capital, so the price already assumes the business slowly shrinks forever.

7. Catalysts and timeline

Q3 results in early November 2026 are the single test of the second-half inflection. Year-end leverage below 3.5x would be confirmed in the FY2026 10-K around March 2027. Further portfolio divestitures or bolt-ons, use of the remaining $23.0 million authorization, and resolution of the securities class action are secondary. A sponsor block sale would clear the overhang but hurt the price first.

8. Pre-registered kill criteria

Risks: 5.4x net debt to EBIT on discretionary demand, a largest distribution partner at 22% of sales, tariff volatility, and a putative securities class action naming the company and current management (10-K 2025, Item 1A). 1. Q3 2026 adjusted EBITDA below $33 million, or any cut to the $127-137 million range. 2. Credit agreement total leverage above 3.5x at 2026 year-end, per the FY2026 10-K. 3. Q3 and Q4 2026 price-versus-volume bridges both showing volume down year over year. 4. Diluted share count up year over year at both of the next two reporting dates while the $23.0 million authorization goes unused.

9. Verdict and summary

WATCH, conviction 3. Holley is a 41-43% gross margin, low-capex portfolio of genuine enthusiast brands whose equity is a thin stub on about $459 million of net debt, so at $3.05 you buy a $370 million equity attached to an $828 million enterprise, and the reverse DCF says the price already assumes cash flow shrinks about 1.5% a year forever. The deleveraging is real, $115 million repaid since September 2023 out of free cash flow, but covenant leverage has barely moved, 3.75x to 3.74x, because EBITDA fell as fast as the debt. Every dollar of growth this year came from price and the HRX acquisition while volume fell about $18.6 million in the first half, and guidance now requires second-half adjusted EBITDA up 9% to 26% at a 23% margin in the seasonally weaker half. Management points to $8.3 million of cost savings, divested low-margin brands and new retail placements as the bridge, and it may work, but no insider bought a share in the open market in twelve months, the buyback is smaller than the equity-compensation dilution, and about a third of voting shares went against the sponsor director and the share-plan increase. Probability-weighted value is roughly $3.95 against $3.05, not enough margin of safety for this much leverage on discretionary demand. The Q3 print in early November settles it.

Research for discussion, not investment advice. Positions and sizing are the reader's decision.