Four Ways to Sell Hardware Into the Boom
- ETN — Eaton. A diversified electrical giant. Switchgear, UPS, transformers, busway — the power distribution and protection backbone of a data center — plus aerospace and broad industrial end markets. Huge, direct data-center exposure, but far from a one-theme company.
- GEV — GE Vernova. The power-generation and grid-equipment arm spun out of GE. Gas turbines, grid transformers and switchgear, plus a loss-making wind business. It sells the machines that make the electricity and the gear that moves it — the supply side of AI's power problem.
- VRT — Vertiv. The closest pure-play in this whole series. Data-center thermal management (cooling) and power management (UPS, PDUs, switchgear), sold almost entirely into hyperscale and colocation build-outs. When you hear "AI needs liquid cooling," this is the name.
- FPS — Forgent Power Solutions. A recently-IPO'd small-cap (public since early 2026) making custom electrical distribution gear — transfer switches, switchgear, e-houses, transformers. Data centers are its single biggest end market (~42% of revenue). Explosive growth, thin trading history.
The spectrum runs from diversified (ETN, GEV) to concentrated (VRT) to small-and-speculative (FPS). As with the contractors, where a name sits on that spectrum explains both its selloff and its risk.
Where the Prices Are
As of the August 24 close:
| Ticker | Price | Off 52wk high | vs 200-day | RSI(14) |
|---|---|---|---|---|
| ETN | $408.67 | −14.5% | +7.7% | 42.6 |
| GEV | $942.10 | −21.2% | +8.3% | 39.2 |
| VRT | $254.97 | −32.9% | +0.6% | 39.7 |
| FPS | $32.19 | −51.2% | n/a‡ | 34.9 |
‡FPS IPO'd in early 2026 — no 200-day average yet.
The tell repeats. The diversified anchor (ETN) fell the least (−14.5%, still above its 200-day). The pure-play (VRT) fell twice as hard (−32.9%). The tiny newcomer (FPS) got cut in half — though for FPS the cause is as much a post-IPO share-supply overhang as it is the theme. When the market de-rates a narrative, concentration is beta, and beta cuts both ways.
What the Quarter Actually Said
All four grew fast and raised guidance. The differences are in the quality of the print.
| Rev (YoY) | Adj EPS | Orders / backlog | Fwd P/E | |
|---|---|---|---|---|
| ETN | $8.53B (+21%) | $3.15 (beat) | $24.1B (+43%) | ~30x |
| GEV | $11.11B (+22%) | $2.47 (miss) | $176B (+88% ord.) | ~68x |
| VRT | $3.27B (+24%)* | $1.52 (+60%, beat) | ~$15B backlog | ~38x |
| FPS | $379M (+103%) | +190% NI | $1.98B (+157%) | n/m† |
*VRT revenue missed the Street by ~$105M despite +24% growth. †FPS's P/E is not meaningful — minimal GAAP earnings against a high market cap; figures are its most recent reported quarter (fiscal calendar offset). GEV EPS missed by ~19% even as revenue beat and guidance rose.
Eaton was the cleanest: a beat, a record $24.1B electrical backlog up 43%, book-to-bill above 1.2x, and raised full-year EPS guidance — all with the diversification of an aerospace and industrial business behind it. Vertiv beat on EPS (+60%) and raised guidance above consensus, but revenue came up ~$105M short — a timing miss on lumpy large projects that spooked a stock priced for perfection. GE Vernova is the paradox: orders exploded +88% to a staggering $176B backlog, and it raised revenue and free-cash-flow guidance sharply — yet it missed EPS by 19%, because the wind segment keeps bleeding and the earnings haven't caught up to the order book. Forgent put up the fastest growth of anyone in this series (+103% revenue, backlog +157%), but it's newly public, barely profitable on a GAAP basis, and under a share-supply cloud.
The Verdict — Who I'd Back
My pick: Eaton (ETN). It's the best risk-adjusted way to own the hardware layer. You get large, direct data-center exposure — the $24B backlog up 43% is real — but wrapped in a diversified electrical-and-aerospace franchise that doesn't live or die on a single theme. It's the cheapest of the large-caps here (~30x), it raised guidance, and it fell the least, which in this tape is a feature: it's the name you can hold through the capex fear without being forced out. The one real caveat is balance-sheet: the $9.6B Boyd Thermal acquisition levered it up and will weigh on reported EPS through amortization for a while. Worth it for the thermal-management expansion, but watch the debt.
Best pure-play / highest torque: Vertiv (VRT). If you want undiluted exposure to data-center cooling and power — the single hottest sub-theme as liquid cooling goes mainstream — VRT is it, and it's now 33% off its high. The revenue miss reads as project-timing noise, not demand softness: EPS still jumped 60% and management raised the full-year guide above consensus. At ~38x it's not cheap, and as the purest play it will be the most volatile in both directions. This is the aggressive pick — the one that pays the most if the buildout keeps compounding and hurts the most if it stalls.
Great order book, wrong price: GE Vernova (GEV). The $176B backlog and +88% orders are genuinely staggering, and the power-generation thesis (AI needs electricity; gas turbines and grid gear supply it) is as durable as anything in this series. But you're paying ~68x for it — the richest multiple across all four articles — while the company is still missing earnings and subsidizing a loss-making wind unit. The story is right; the entry price demands flawless execution that hasn't shown up on the bottom line yet. I'd want it cheaper, or I'd want to see the EPS actually convert.
The speculative small-cap: Forgent (FPS). The growth is real and enormous — revenue doubling, a book-to-bill above 2x, a backlog up 157%, all pointed at data centers. But it's a young public company with thin trading history, non-meaningful earnings, real margin drag from a rapid capacity build-out, and a technical overhang from a secondary offering and lock-up expiry that has pushed the stock down for reasons unrelated to the business. Highest optionality, highest risk, least mature data. A watchlist name and a small, deliberate trade at most — not a core position.
The Headwinds That Could Break This
- AI-capex digestion. Same root fear as the rest of the complex. Hardware orders are a direct read on hyperscaler spend; a pause shows up here fast. The diversified name (ETN) bruises, the pure-plays (VRT, FPS) break.
- Order-to-earnings lag. GEV is the cautionary tale — a record backlog is not the same as delivered profit. Supply-chain constraints (gas turbines especially), mix, and execution can hold EPS well behind the order book.
- Valuation compression. At 38–68x, VRT and GEV have the most multiple to lose if the market keeps de-rating growth. Multiple compression can swamp fundamental gains for a year or more.
- Balance sheet / integration. Eaton's Boyd Thermal leverage and its pending Mobility spin-off add complexity and interest expense; VRT is digesting several thermal acquisitions.
- Technical share supply (FPS-specific). Post-IPO secondary offerings and lock-up expirations can pressure a stock independent of fundamentals — a live factor for Forgent right now.
Bottom Line
The hardware layer has the cleanest demand signal in the whole build — orders are exploding — but the market is being ruthless about price and about the gap between orders and earnings. I'd anchor with Eaton (ETN) for diversified, reasonably-priced exposure that I can hold through the fear; pair it with Vertiv (VRT) as the high-torque pure-play if I want maximum leverage to data-center cooling and power; wait for a better price on GE Vernova (GEV) despite its jaw-dropping backlog; and treat Forgent (FPS) as a speculative flier, sized accordingly. Picks and shovels are a great business in a gold rush — as long as you don't overpay for the shovel.
Prices and technicals pulled live from Yahoo Finance as of the August 24, 2026 close (market closed). Q2 2026 fundamentals from company releases and BusinessWire/StockTitan/24-7 Wall St./Zacks coverage; Forgent (FPS) figures are its most recent reported quarter on an offset fiscal calendar and its earnings base is early/thin. Forward P/E figures are approximate. Levels and multiples are point-in-time and will move.