Market closed · Next open: Thursday 9:30 AM ET
Last update: Nov 4, 2026, 6:23 PM ET

AI Infrastructure

Legence (LGN) Stock: Why a Record Q2 Sold Off — Then Rebounded — on Margins and an Impairment

Legence reported a record Q2 FY2026 before the open on August 13 — revenue up 111% to $1.26 billion, adjusted EBITDA up 114%, a record $5.67 billion backlog that is ~60% data-center, and raised full-year guidance — yet the stock fell about 7.9% that day before rebounding ~6% on August 14. The selloff was about a real gross-margin slip (17.4% from 21.5%), a $41.1 million impairment in its sustainability-services unit, and a premium multiple (~14x FY26 EBITDA) after a hot run-up; the bounce reflected a mostly non-cash loss, solid cash generation, low ~1.6x leverage, and a 26%-above-consensus EBITDA beat. Peers were roughly flat, confirming the move was company-specific.

By Roberto LiccardoPublished (ET)10 min readLGN
Wide view inside a modern data center's mechanical and electrical infrastructure — rows of cooling air-handling units, overhead cable trays and colored power and cooling piping along a service corridor under cool blue lighting — representing Legence's data-center engineering and services business.

Why Legence stock swung on its record second quarter

On Thursday, August 13, 2026, before the open, Legence (NASDAQ: LGN) reported a record second quarter — revenue up about 111%, adjusted EBITDA up 114%, a record backlog, and raised full-year guidance — and the stock fell about 7.9% for the day, trading down as much as roughly 9% intraday on heavy volume [1][4]. The next session it reversed: by early afternoon on Friday, August 14, LGN was around $67, up about 6%, recovering most of the drop [2]. A record quarter sold and then bought back the next day is a puzzle worth unpacking — and the answer is a real margin slip, a genuine but contained impairment, and a stock that had run up hot into a premium valuation.

Legence gives investors exposure to the data-center buildout through physical mechanical, electrical and sustainability (MEP) infrastructure services — designing, building and maintaining the cooling, power and controls that make a data center run. It is a Blackstone-backed, acquisitive consolidator that went public on September 12, 2025, so it trades with the volatility of a young, richly valued name with a thin post-IPO float [3].

The earnings scorecard: strong growth, weaker margins

On the top and cash-earnings lines, the quarter beat and the company raised guidance [1]:

  • Revenue $1.26 billion, up 110.7% (from $598.9 million), about 15% above the ~$1.10 billion consensus; revenue growth excluding the Bowers Group acquisition was about 60% (a non-GAAP measure) [1][5].
  • Adjusted EBITDA $154.6 million, up 114%, about 26% above the ~$123 million consensus and above the $115–125 million guide, at a 12.2% margin — roughly flat year over year (about 12.1% a year ago) but up about 90 basis points sequentially [1][5].
  • Raised full-year 2026 guidance to revenue of $4.7–4.8 billion (from ~$4.1–4.3 billion) and adjusted EBITDA of $565–585 million [1].
  • Record backlog and awarded contracts of $5.67 billion, up 105%. The firmer, contracted figure is $4.02 billion of remaining performance obligations, of which 65–75% is expected to convert within 12 months; data centers and technology are about 60% of backlog, up from roughly 36% a year earlier [1].

What spooked the tape sat lower on the income statement. Gross margin fell to 17.4% (18.5% on an adjusted basis) from 21.5% a year earlier (21.8% adjusted) — a real, roughly 330–410-basis-point compression, worst in the Engineering & Consulting segment (adjusted gross margin 31.1% versus 33.2%) with the Installation & Maintenance segment also under some pressure [1]. And Legence swung to a GAAP net loss of $27.8 million attributable (a $34.6 million consolidated loss), or $(0.37) per diluted share [1].

Chart titled 'Legence (LGN): a record Q2 the market sold, then bought back.' Left panel shows Q2 FY2026 year-over-year growth bars: revenue +111%, adjusted EBITDA +114%, backlog +105%, with a red note that gross margin slipped to 17.4% from 21.5%. Right panel shows August 13 2026 close-to-close moves: Legence down 7.9% while its MEP peers were roughly flat - Quanta about -0.8%, EMCOR about -0.3%, APi Group about -0.2%. A stat strip shows revenue $1.26 billion (estimate $1.10 billion), adjusted EBITDA $154.6 million (estimate $123 million), record backlog $5.67 billion up 105%, raised FY26 revenue guide $4.7-4.8 billion, and a Strong Buy consensus with an average target near $106.
Legence beat on revenue, adjusted EBITDA and backlog and raised guidance, but a gross-margin slip and a $41.1M impairment-driven GAAP loss sent the stock down 7.9% on Aug 13 while its MEP peers were roughly flat — before a ~6% rebound on Aug 14. Sources: Legence Q2 2026 release and 10-Q; StockAnalysis.

The margin slip and the impairment, explained

Two things drove the loss and the sell-off, and it is worth separating them. The gross-margin compression is partly mix — the Bowers acquisition and faster-growing, lower-margin installation work reweighting revenue — but it was not purely optical: adjusted gross margin fell across the business, and Engineering & Consulting saw genuine cost and demand pressure. What kept it from flowing through to profitability was operating leverage below gross profit: selling, general and administrative expense fell to 11.7% of revenue from 12.1%, which is why adjusted EBITDA margin held roughly flat even as gross margin slipped [1].

The GAAP loss, meanwhile, was largely non-cash. It included a $41.1 million impairment — $21.6 million of goodwill and $19.5 million of long-lived assets — taken against a sustainability-services reporting unit in Engineering & Consulting, reflecting genuinely lower new-contract demand and reduced revenue and cash-flow projections for that business, on top of $73.9 million of stock-based and other non-cash compensation and $44.2 million of depreciation and amortization [1]. That impairment is a real negative — a piece of the business is shrinking — but it is a segment issue, not a crack in the data-center demand that drives the backlog. Notably, interest expense was not the problem: it actually fell to $16.9 million from $30.4 million a year earlier [1]. BTIG cut its price target to $100 from $120 (keeping a Buy rating) on the margin concern, even as it acknowledged the beat [6].

Why it matters

Legence is a way to own the AI-infrastructure build-out one step removed from the chips and clouds — the physical plant of a data center. That market is growing fast: data centers and technology went from about a third to about 60% of a record $5.67 billion backlog in a year [1]. The bull case is that this demand has years to run and that Legence, as a scaled MEP consolidator, compounds revenue and EBITDA through organic growth and acquisitions while it de-levers. The bear case is that an acquisitive consolidator at a premium multiple has little room for the gross-margin slip, segment impairment and GAAP loss it just printed. The same tension runs through the AI-infrastructure names we track, from the memory suppliers at SanDisk to the data-center engineering backlog at Sterling Infrastructure.

How the peers and sector traded the same day

The clearest evidence that this was a company-specific reaction, not a sector move, is that Legence's closest data-center-infrastructure peers were roughly flat on the same session while LGN dropped nearly 8% [8]:

Name (ticker)Aug 13, 2026 closeRead-through
LGN — Legence−7.9%Its own Q2 print: margin slip + impairment-driven loss into a rich multiple [1]
PWR — Quanta Services≈−0.8%Electrical/infrastructure services peer; no shared catalyst [8]
EME — EMCOR Group≈−0.3%Closest MEP-contracting comparable; data-center demand intact [8]
APG — APi Group≈−0.2%Safety/specialty services peer [8]

With Comfort Systems, EMCOR and Quanta essentially unchanged while Legence fell nearly 8%, the move was clearly about Legence's specific print — the margin line and the sustainability-services impairment — not the data-center demand picture that supports the whole group [8].

What the Street did with it

The analyst reaction captured the split: trim the target on margins, keep the rating on the growth [6]. BTIG cut to $100 from $120 but reiterated Buy, citing margins despite the ~26% EBITDA beat. Coming into the print the sell side was broadly constructive — Bernstein had initiated at Outperform — and the stock carried a consensus "Strong Buy" across roughly 17 analysts with an average target near $106 (range about $80–$125), well above the ~$67 price [7]. Even the firm most vocal about the margin worry stayed bullish on the business; the debate is about the multiple, not the demand.

What to watch

  • Gross-margin trajectory. Watch whether gross margin stabilizes as the Bowers acquisition laps and installation mix normalizes, and whether SG&A leverage keeps holding adjusted EBITDA margin steady [1].
  • The sustainability-services unit. The $41.1 million impairment flags real demand softness in Engineering & Consulting; watch whether it stays contained or spreads [1].
  • Customer and backlog quality. One Installation & Maintenance customer was 17.2% of Q2 revenue, up from 5.7% a year earlier — watch that concentration, plus the conversion of the $4.02 billion contracted RPO (65–75% due within 12 months) and whether the data-center share keeps rising [1].
  • De-leveraging and organic growth. Revenue ex-Bowers grew about 60%; watch the organic/acquired split and whether net leverage keeps falling from ~1.6x as cash builds. Moves are tracked on the LGN stock page and the earnings calendar [1][7].

Illustrative valuation sensitivity

The scenarios below are a descriptive, author-weighted exercise — not a forecast, target, recommendation, or intrinsic fair-value calculation. Each price is expressed as an EV/FY26-adjusted-EBITDA multiple to make it concrete: implied price = (multiple × ~$575 million FY26 adjusted-EBITDA guidance − ~$749 million net debt) ÷ 108.05 million shares [1]. The weights are the author's subjective assumptions; the downside deliberately sits below the Street's ~$80 target floor to capture a de-rating case [7]. The percentages are illustrative and sum to 100%.

ScenarioIllustrative priceWeightKey drivers
Upside~$125 (≈25x EV/FY26 EBITDA)30%Gross margin stabilizes, the sustainability softness stays contained, backlog converts, and the premium multiple holds — toward the high end of the analyst range [7].
Middle~$100 (≈20x EV/FY26 EBITDA)45%Guidance is delivered and leverage keeps falling, but gross-margin questions keep some discount on the multiple, holding the stock near the consensus target — still above today's price [1][7].
Downside~$50 (≈11x EV/FY26 EBITDA)25%Margin compression persists, the sustainability weakness spreads or a large customer rolls off, and the multiple de-rates toward the lower post-IPO range near the 52-week low — below the Street's target floor [2][3].

Weighting those (0.30 × $125 + 0.45 × $100 + 0.25 × $50) gives an author-weighted scenario value near $95, above the ~$67 quote and today's ~14x forward multiple [2]. The point is the width of the band, from the 52-week-low area to $125: on a fast-growing but currently GAAP-loss-making consolidator, small changes in the margin trajectory move the valuation a lot. This is descriptive analysis of a corporate event, not investment advice.

LGN data snapshot — August 14, 2026 (intraday, session open)

FigureValueAs-of / source
Earnings-day close$63.24 (−7.9%; intraday low ≈−8.8%)Thu, Aug 13, 2026 — Q2 reported pre-open, heavy volume [4]
Intraday quote≈$67.05 (+6.0%)Aug 14, ~12:03 p.m. ET — StockAnalysis [2]
Day's range (Aug 14)$63.35 – $67.94Aug 14 session, still open [2]
Prior-day run-up+5.7% on Aug 12 (into the print; close $68.68)Aug 12 session [4]
52-week range$26.96 – $107.24 (IPO Sep 12, 2025)As of Aug 14 [2]
Market cap≈$7.24B (108.05M shares)Aug 14 — StockAnalysis [2]
Valuation≈14x FY26 adj-EBITDA guide (≈26x trailing); ≈43x forward P/E; trailing P/E n/a (GAAP loss)Aug 14 — StockAnalysis [3]
Balance sheetTotal debt $1,041.4M, cash $292.0M → net debt ≈$749M; net leverage ≈1.6× reported (1.5× pro forma), from ~3× at IPO; H1 operating cash flow $194.5MAt Jun 30, 2026 — 10-Q [1]
Q2 FY2026 resultsRevenue $1.26B (+110.7%); adj. EBITDA $154.6M (12.2% margin); gross margin 17.4% vs 21.5%; GAAP EPS −$0.37 (incl. $41.1M impairment)Reported Aug 13 — release / 10-Q [1]
Backlog / guidanceBacklog + awarded $5.67B (+105%); contracted RPO $4.02B (65–75% within 12 mo), ~60% data-center; FY26 guide raised to rev $4.7–4.8B / adj. EBITDA $565–585MAug 13 — company release [1]
Analyst consensusStrong Buy; avg target ≈$106 (range ~$80–$125); BTIG $100 (from $120), Buy~17 analysts — StockAnalysis / BTIG [6][7]

The BestStocks event page corroborates the Aug 14 gain (about +5.8%, near $67). Aug 14 prices are intraday snapshots that moved through the session, not closing prices; the $63.24 Aug 13 figure is a settled close.

The setup at a glance

QuestionAnswer
What happened?Legence reported a record Q2 FY2026 before the open on Aug 13 (revenue $1.26B, +111%; adj. EBITDA $154.6M, +114%; raised guidance) — yet the stock fell ~7.9% that day, then rebounded ~6% on Aug 14 [1][2].
Why did a record quarter fall?Gross margin fell to 17.4% from 21.5%, it took a $41.1M impairment in its sustainability-services unit and posted a GAAP loss (−$0.37), and it trades at a premium (~14x FY26 EBITDA, ~43x forward P/E) after running up ~5.7% into the print [1][3].
Why did it bounce back?The loss was mostly non-cash (impairment, stock comp, D&A) and cash generation is solid ($194.5M H1 operating cash flow); SG&A leverage held adjusted EBITDA margin as gross margin slipped; net leverage is a low ~1.6x and falling; and the beat and raised guidance reasserted [1].
Is the impairment a red flag?A contained one: the $41.1M charge reflects softer demand in Legence's sustainability-services reporting unit — a real negative for that business, but separate from the data-center demand that drives the backlog [1].
Was it a sector move?No — peers were roughly flat (Quanta ≈−0.8%, EMCOR ≈−0.3%, APi Group ≈−0.2%) while LGN fell ~8%, confirming a company-specific reaction [8].
What do analysts think?Consensus "Strong Buy," average target ~$106 (range $80–$125, above the price); BTIG trimmed to $100 on margins but kept Buy [6][7].

Legence (LGN) stock FAQ

Why did Legence (LGN) stock fall on August 13, 2026?

Legence reported a record Q2 FY2026 before the open — revenue up 110.7% to $1.26 billion (about 15% above consensus), adjusted EBITDA up 114% to $154.6 million (about 26% above consensus), and raised full-year guidance — yet the stock fell about 7.9% (and as much as ~8.8% intraday). The selloff was driven by a gross-margin slip to 17.4% from 21.5% a year earlier, a $41.1 million non-cash impairment ($21.6 million goodwill and $19.5 million long-lived assets) taken against its sustainability-services business, a GAAP net loss of $27.8 million (−$0.37 per share), and a premium valuation after the stock had run up about 5.7% into the print. BTIG cut its price target to $100 from $120 on margin concerns while keeping a Buy rating.

Why did LGN rebound the next day?

On August 14 the stock recovered roughly 6% to about $67. The GAAP loss was mostly non-cash — the $41.1 million impairment, plus $73.9 million of stock-based compensation and $44.2 million of depreciation and amortization — while the business generated $194.5 million of operating cash flow in the first half. Adjusted EBITDA still beat consensus by about 26% because SG&A leverage (11.7% of revenue, down from 12.1%) offset the gross-margin slip, net leverage was a low ~1.6x (down from about 3x at the IPO), and the raised guidance and record data-center backlog reasserted. Together those gave buyers a reason to fade the initial drop.

What was Legence's Q2 impairment, and is it a red flag?

Legence took a $41.1 million non-cash impairment — $21.6 million of goodwill and $19.5 million of long-lived assets — against a sustainability-services reporting unit in its Engineering & Consulting segment, reflecting lower new-contract demand and reduced revenue and cash-flow projections for that business. It is a genuine negative for that unit, but a contained one: it reflects weakness in the sustainability-consulting business rather than in the data-center demand that drives Legence's record backlog. The thing to watch is whether that softness stays isolated or spreads.

Why did Legence's gross margin fall, and does it matter?

Gross margin fell to 17.4% (18.5% adjusted) from 21.5% (21.8% adjusted) a year earlier — a real compression, worst in Engineering & Consulting (adjusted gross margin 31.1% versus 33.2%), partly from a mix shift toward lower-margin installation work and partly from cost and demand pressure. It did not flow through to adjusted EBITDA margin, which held roughly flat at 12.2%, because SG&A fell to 11.7% of revenue from 12.1% — operating leverage below gross profit. The key question is whether gross margin stabilizes as the acquisition mix normalizes; it is the number the bear case rests on.

What does Legence do, and is it an AI-infrastructure stock?

Legence provides mechanical, electrical and sustainability (MEP) infrastructure services to data-center buildouts and other complex facilities — designing, building and maintaining the cooling, power and controls that make a data center run. It is effectively a 'picks and shovels' way to own the AI-infrastructure build-out one step removed from the chips and clouds. Data centers and technology grew to about 60% of its record $5.67 billion backlog, up from roughly 36% a year earlier. Legence is a Blackstone-backed, acquisitive consolidator that went public in September 2025, so it trades with the volatility of a young, richly valued name.

View all research →
An abstract visualization of a falling stock chart over a search and AI motif in orange-red tones, representing Reddit's crash after a Q3 revenue beat was overshadowed by a Google AI search-traffic warning.
Stock AnalysisOctober 30, 2026 · 12 min read

Reddit (RDDT) Crashes ~30% After a Q3 Beat on a Google AI Search-Traffic Warning

Reddit crashed about 30% to roughly $90.79 on October 30 — a striking reaction, because the quarter itself was strong: revenue of $895 million rose 53% year over year, a clear beat, with net income of $283 million, diluted EPS of $1.40 and adjusted EBITDA of $402 million. The stock cratered on what management said about the future. On the earnings call, CEO Steve Huffman warned that search referrals fell more than a million per week on average in October as Google answers more queries with AI, and that daily active users (around 130 million) could fall a low-double-digit percentage in the fourth quarter if the trend held. Investors also fixed on a 2% year-over-year decline in US daily actives — Reddit's most valuable market, where revenue per user is more than five times the international level — and on fourth-quarter revenue guidance of $1.00–$1.03 billion that came in below expectations; Citizens downgraded the stock before the open. The one bullish counterweight was monetization: revenue per user still jumped 36% globally. The move was company-specific — Reddit's roughly 30% drop was about four times the moves of social peers Snap and Pinterest (each down ~7%), on a day the broad market fell about 2%. After the crash Reddit trades at roughly 14 times forward earnings with about $3 billion of net cash — not expensive for 50%-plus growth, if the growth is durable. The whole debate is whether rising revenue per user can keep outrunning a top of the funnel that AI search is draining.

An abstract visualization of a falling stock chart over an automotive and circuit-grille motif in steel-blue and red tones, representing General Motors' drop after a Q3 beat-and-raise was overshadowed by 2027 cost and EV concerns.
Stock AnalysisOctober 20, 2026 · 10 min read

General Motors (GM) Falls ~8% Despite a Q3 Beat and a $20B Buyback

General Motors fell about 8% to roughly $75.77 on October 20 — a striking reaction, because GM beat on third-quarter earnings, raised its full-year profit outlook and announced a $20 billion buyback. Adjusted earnings of $3.70 a share topped estimates on revenue of about $50 billion, and GM lifted the midpoint of its full-year adjusted-EPS guidance to about $13.25 (from $13.00) and raised its free-cash-flow outlook. The stock fell anyway on what the guidance implied, not the quarter. The full-year outlook implies a softer fourth quarter — only about $2.7 billion of operating profit versus $4.1 billion in the third, as a Silverado/Sierra redesign temporarily cuts truck deliveries by roughly 35,000 units — the raised $13.25 EPS midpoint still landed just below the roughly $13.41 the Street expected, and management flagged about $2 billion of 2026 cost inflation plus a conditional scenario of more in 2027. Two things that look like culprits mostly were not: the 62% drop in EV sales was already known and reflects a distorted 2025 comparison, and tariff costs are now expected near the low end of an existing range. Crucially, GM did not cut its 2027 outlook — it still expects next year to improve if conditions hold. The move was clearly company-specific — Ford fell only about 1%, Stellantis was flat and the S&P 500 rose — so this was a repricing of GM's forward earnings, not an auto-sector sell-off. The valuation captures the tension: GM trades at barely five to six times the forward earnings it guides to, extraordinarily cheap, because the market prices automakers for the downturn, not the peak. A $20 billion buyback into that multiple is powerfully accretive if earnings hold — but whether they hold in 2027 is exactly the question. Analysts keep a Buy consensus and an average target near $104.

An abstract visualization of a rising stock chart over a molecular antibody and DNA-helix motif in cool teal and violet tones, representing Summit Therapeutics' jump on a new ivonescimab kidney-cancer trial.
Stock AnalysisOctober 19, 2026 · 10 min read

Summit Therapeutics (SMMT) Jumps ~6% on a New Ivonescimab Kidney-Cancer Trial

Summit Therapeutics rose about 6% to roughly $15.57 on October 19 after announcing it had opened a global Phase II trial of ivonescimab — its closely watched lead drug — in previously untreated advanced clear cell kidney cancer, with partner Akeso. The key nuance: this is the opening of a mid-stage study, not a data readout, so it is an incremental catalyst that expands the story rather than one that proves it. Summit is effectively a one-drug company: ivonescimab is a PD-1/VEGF bispecific antibody, licensed from China's Akeso, that outperformed Merck's Keytruda in a Chinese lung-cancer trial — nearly doubling progression-free survival and, in updated data, delivering a significant overall-survival benefit (30.8 vs 22.6 months) — though the global trial behind its U.S. filing has yet to formally confirm that survival benefit in Western patients. With no product revenue and a market value above $12 billion, essentially all of Summit's worth is the market's bet on that drug, so a new indication is a fresh call option on the thesis. Two bigger items frame it: AstraZeneca agreed in late September to invest about $2 billion in Summit equity (a ~12% stake), transforming the balance sheet and removing the financing overhang, and the FDA has a November 14, 2026 decision date on ivonescimab in EGFR-mutated lung cancer — a near-term, binary catalyst that matters far more than a trial start. Monday's move was partly company-specific and partly a tailwind: biotech broadly rallied, with the sector ETF up about 2.5%, but Summit outperformed. The risks are a binary valuation on one molecule, an unconfirmed Western survival benefit, and a stock down roughly 47% from its high. Analysts carry a Buy consensus and an average target near $30, inside a wide range from about $12 to the mid-$40s.