Comfort Systems USA FIX
Industrials · Engineering & Construction · Synthos Deep Dive · 2026-07-03
The Overview
> Comfort Systems USA is a plumbing-and-wiring company — but at industrial scale. It installs and services the heating, cooling, air-flow, piping and electrical systems inside big buildings. Right now its hottest business is building the cooling and power systems for AI data centers, the giant warehouses full of computers behind the AI boom.
>
> Is the stock cheap or expensive? Expensive. You're paying about 50 dollars for every 1 dollar the company earned last year — a rich price that only makes sense if it keeps growing fast. The good news is it is growing fast, and faster each quarter.
>
> Our verdict: Buy — but tactically, meaning buy a smaller amount than you would a rock-solid company, because this is a construction business tied to the building cycle. When companies stop building data centers, its growth can reverse.
>
> The one big worry: its work backlog is huge today, but a backlog is a snapshot of orders, not a guaranteed contract for years. If the AI-building spree slows, the growth story cools quickly.
>
> What the three scores mean in plain words:
> - Downside Risk 6/10 (a bit above average risk). The company owes almost no money (safe), but the stock swings hard and it depends on the building cycle.
> - Growth Quality 8/10 (very good). Sales and profits are growing fast and it earns a high return on its money.
> - Exponential Potential 6/10 (moderately high). Unusually, its growth is still speeding up — but it's a contractor, not a software company, so don't expect that forever.
Putting a number on it: our fair-value estimate is $1,850 against a current price of $1,518.73 — real upside if our numbers are right.
Our summary metrics
Net cash & A- balance sheet, but beta 1.67, cyclical construction demand, and ~50× trailing on a datacenter boom.
~29% FY25 revenue growth (51% organic in Q1'26), EPS doubling, 52% ROE, backlog $12.45B (nearly 2× YoY).
Growth is still ACCELERATING (rare) on the AI-datacenter buildout — but it is a cyclical contractor, not a software compounder.
What does “fair value” mean?
Fair value is Synthos’s estimate of what one share is worth today, based on our model of the company’s future cash generation and the risks to it. It is not a price target or a prediction of where the stock trades next quarter — it’s the price where we think risk and reward are balanced. Above it, you’re paying for outcomes better than our base case; below it, the market is offering a margin of safety.
The model’s inputs and weightings are proprietary — they’re the product. What we publish is the output, plus an independent, public-math cross-check further down the page (“Check our number”) so you can judge it for yourself.
What do the 5 tiers mean? (Core · Tactical · Watch · Hold · Avoid)
Exponential Potential
Growth is still ACCELERATING (rare) on the AI-datacenter buildout — but it is a cyclical contractor, not a software compounder.
What could take this further than the base case — and where the market may be underpricing it. Full forward-growth and acceleration math in Deeper analysis, §4 below.
Deeper analysis
Technicals, fundamentals, valuation, and the full expert-claim evidence panel — the detail behind the numbers above.
Reference table
| Street consensus | $1,991.50 (high $2,200 / low $1,800; 5 Buy · 4 Hold · 0 Sell) — context, not our anchor |
| Valuation | ~50× trailing EPS · 40× FY26E · 33× FY27E · 26× FY28E · EV/S 6.0× · EV/EBITDA 35× |
| Technicals | Uptrend intact but cooling — $1,741, −16% off 52-wk high, below 50-DMA, above 200-DMA, RSI 45, +234% 12-mo (SPY +21%) |
| Conviction | Moderate — 0 expert voices in the Synthos KB; call rests on fundamentals, estimates and quant |
| Position sizing | Satellite / tactical, ~1–3% — cyclical beta 1.67, size it like a cyclical |
What the experts actually said 1 traceable claims on FIX · showing the highest-conviction voices
“The AI datacenter buildout is real, not retail hype — Comfort Systems reported its largest-ever backlog, up ~$9-10B (65% y/y), transforming from a midsize contractor into a core AI-infrastructure player.”
Every claim reconciles to a real claim_id in the Synthos knowledge base — this is the evidence the verdict is built on, not vibes. Management (the company itself) is shown but half-weighted; one cautionary voice is included on purpose.
Price & moving averages 12 months · 50 & 200-day averages · 52-week range
Solid line = price · dashed line = 50-day average · dotted line = 200-day average · the two thin horizontal lines mark the 52-week high and low. Price above both averages is an uptrend.
Data summary: last close $1,518.73, 13% below the 50-day average ($1,752), 3% above the 200-day average ($1,476) — a mixed trend. 27% below the 52-week high of $2,067, 117% above the 52-week low of $699.
Bollinger Bands 20-day average ± 2 standard deviations
The shaded band widens when the stock gets more volatile. Riding the upper edge = strong momentum (sometimes stretched); the lower edge = weak / potentially oversold.
Data summary: price $1,518.73 is currently at/below the lower band (potentially oversold) (band $1,531–$1,855).
RSI (14) momentum gauge · 0–100
Above 70 (overbought zone, shaded) = overbought, below 30 (oversold zone, shaded) = oversold. Currently 37.
MACD 12 / 26 / 9 · trend & momentum
The MACD line crossing above the signal line (bars flip to the up color) = momentum turning up; crossing below (bars flip to the down color) = turning down. Bar height = the size of that gap.
Data summary: MACD is currently below its signal line by 16.53, negative momentum.
Relative performance vs S&P 500 & its sector (XLI (sector)), set to 100 a year ago
Solid = FIX · dashed = S&P 500 · dotted = XLI (sector). A rising line means it is beating that benchmark — the sector line shows whether it is a leader or laggard within its own group.
Forward revenue & earnings actual → estimate · "FY" = fiscal year, "E" = estimate
Darker bars = actual results, brighter = analyst estimates. Taller bars to the right = expected growth.
Key stats an RIA wants
1. What it is
Comfort Systems USA (NYSE: FIX) is a Houston-based national provider of mechanical and electrical (M&E) installation, renovation, maintenance, repair and replacement services — HVAC, plumbing, piping, controls, fire protection and electrical work — for commercial, industrial and institutional buildings. It operates a decentralized roll-up of ~197 locations across 143 cities and ~22,700 employees, growing through both organic share gains and bolt-on acquisitions. Founded 1917; IPO 1997. Fiscal year ends December 31. CEO Brian E. Lane.
The reason the stock has 3.3×'d in a year is demand mix: FIX has become a prime beneficiary of the AI-datacenter and advanced-manufacturing (chip fab, industrial) construction wave, where the mechanical and electrical scope per building is enormous.
Revenue mix (FY2025, from filings):
- By segment: Mechanical $6.67B (73%) · Electrical $2.43B (27%). The Electrical segment is the faster-growing piece (up from $1.50B in FY24, +62%), reflecting datacenter power/electrical demand.
- By geography: the FMP geographic segmentation is empty for FIX — the business is essentially 100% United States, which removes FX and foreign-policy risk but concentrates it fully in the US construction cycle.
2. The expert thesis — (none in the Synthos KB)
There is no expert coverage of FIX in the Synthos knowledge base: total_claims = 0, net_bullish_voices = 0. No investor or operator in our tracked panel has published a traceable claim on this name. That is the honest state of the KB, and it is common for mid-cap industrials outside the mega-cap/AI-narrative spotlight.
Consequently this verdict is entirely fundamentals-, estimates- and quant-driven. We do not manufacture conviction we don't have. Where the LLY-style note would cite a dozen voices, here you get the numbers themselves — FMP financials, analyst estimates, the SEC earnings release (§9) — and nothing is attributed to an expert because none exists. Readers who weight Synthos notes by KB breadth should treat this as a breadth-zero name: the thesis stands or falls on the data, not on a panel.
3. Synthos scores & the Bull / Base / Bear cases
The one-glance judgment — three scores, 0–10, each anchored to real metrics (not probabilities we can't honestly calibrate):
| Score | 0–10 | The read |
|---|---|---|
| Downside Risk (lower = safer) | 6 · Moderate-High | Net cash (net debt −$196M, A- rating) and 52% ROE are fortress-like, but beta 1.67, a ~50× trailing multiple, and structural construction cyclicality mean a demand or multiple wobble hurts. A −16% drawdown already off the high shows the volatility. |
| Growth Quality | 8 · High | FY25 revenue +29.5%, Q1'26 organic revenue +51%, EPS more than doubled YoY, ROE 52%, ROIC ~38%, net-cash — high-quality right now. Capped below 9 because the returns are cycle-inflated, not annuity-like. |
| Exponential Potential | 6 · Moderate-High | The second derivative is positive — revenue growth held ~+29.5% (FY25) and re-accelerates toward ~+30.6% (FY26E), with Q1'26 organic revenue +51% YoY and backlog nearly doubled YoY. Genuinely rare. But it is a cyclical contractor at a $61B cap, so this is not a durable software-style exponential. |
The three cases (our own scenario model — assumptions shown; each target is a ~12–18-month fair value). We deliberately do not attach probabilities: the base case is the expected path, so a weighted blend would just restate it with false precision. The cases bound the range; the scores summarize them.
| Case | Key assumptions | Fair value |
|---|---|---|
| Bull | AI-datacenter/fab capex boom persists; backlog keeps converting; FY27E EPS beats to ~$60 (vs $53.4 cons) and the market keeps paying a growth ~41×. | ~$2,450 (+41%) |
| Base (our anchor) | Estimates roughly hit — FY26E EPS $43.15, FY27E $53.44; a still-fast but maturing cyclical earns a ~35× forward multiple on FY26E. | ~$1,850 (+6%) |
| Bear | Datacenter capex cycle rolls over; backlog burns without replacement; FY27E EPS stalls near ~$40 and the multiple de-rates to ~26× as the market re-prices it as a cyclical. | ~$1,050 (−40%) |
Synthos fair value = the base case, ~$1,850 (+6%), with the full $1,050–$2,450 span as the honest range. Note the base case sits only modestly above today's price and below the Street's $1,991.50 consensus — we are more cautious than the Street because we discount the cyclicality and the rich multiple more heavily. This is a tracked call — the Forecaster Scorecard grades it once it matures.
4. Exponential Potential
Synthos separates compounders (durable high returns on capital) from exponentials (accelerating, multi-baggers-from-here). FIX is the unusual case of a cyclical that is genuinely accelerating — which is why it scores above a 5 despite not being a software or platform business:
- Forward growth: revenue CAGR FY25→FY28E ~21.7% ($9.10B → $16.39B); EPS CAGR ~32% ($28.88 → $67.26 cons) as operating leverage and scale expand margins.
- Acceleration (the 2nd derivative) is POSITIVE — the rare part: revenue growth +34.9% (FY24) → +29.5% (FY25) → +30.6% (FY26E) on estimates, and within the year Q1'26 organic revenue was +51% YoY. Same-store backlog nearly doubled ($6.89B → $12.21B YoY). Most names we score are decelerating off an inflection; FIX is still climbing.
- Room to run: at $61B market cap in a large, fragmented US M&E services market (tens of billions of addressable spend, plus the datacenter/fab super-cycle), there is real runway — but this is a contractor, so "room to run" is bounded by the building cycle, not a winner-take-most TAM.
- Reinvestment runway: disciplined bolt-on M&A ($280M FY25 acquisitions) plus organic capacity — funded from FCF ($1.03B FY25) with net cash. The reinvestment engine is intact and self-funding.
Exponential Potential: Moderate-High (6/10). The acceleration is real and rare, which lifts it above a mega-cap compounder — but honesty requires flagging that the accelerant is a capex cycle (AI datacenters, fabs), and cycles turn. Own it for the acceleration while it lasts, sized for the fact that it is a beta-1.67 cyclical.
5. Financials (real numbers — FMP annual/quarterly)
- Revenue: FY25 $9.10B, +29.5% (FY24 $7.03B, +34.9% on FY23 $5.21B). Multi-year acceleration at scale.
- Quarterly trajectory (real acceleration): Q1'25 $1.83B → Q2 $2.17B → Q3 $2.45B → Q4 $2.65B → Q1'26 $2.87B (+56.5% YoY, +51% organic per management). Second derivative positive through the latest print.
- Margins (expanding): gross 25.1% TTM (up from ~21% FY24), EBITDA margin 17.1% TTM, operating ~15.7%, net 12.1% TTM (up from 7.4% FY24). Operating leverage is the story — same work, fatter margins as volume scales.
- Earnings: net income $1.02B FY25, nearly double FY24's $522M; diluted EPS $28.88 vs $14.60. Q1'26 net income $370M, EPS $10.51 (vs $4.75).
- Cash flow: operating CF $1.19B FY25, capex −$155M, FCF $1.03B — real cash conversion, not just accrual earnings (income quality TTM 1.36). Aided by large customer deposits/deferred revenue ($2.12B) — a positive working-capital dynamic (customers prepay), but note that inflates cash and can reverse if bookings slow.
- Balance sheet: total debt $786M against $982M cash → net cash −$196M; net-debt/EBITDA −0.4×. A- letter rating. Financially fortress-like; the risk here is demand/multiple, not solvency.
6. Valuation — priced in or room?
There is no way to call FIX cheap on trailing numbers (~50× EPS, 6.0× sales, 35× EV/EBITDA, 22× book). The FMP letter model scores price-to-earnings and price-to-book a 1 out of 5 even as the business quality scores 4–5. The bull's defense is that EPS is growing faster than the multiple: on live consensus the forward P/E compresses to 40× (FY26E) → 33× (FY27E) → 26× (FY28E) — the rich multiple works itself off if estimates hit. The PEG is favorable on trailing growth (~0.47) but far less so on forward (~2.1), which is the honest tell: you are paying up for growth that the Street expects to decelerate from here even as it stays high.
Street targets (context, not our anchor): consensus $1,991.50, high $2,200, low $1,800 — notably the entire Street range sits above today's $1,741 price, an unusually bullish setup. Our base FV of ~$1,850 is deliberately below consensus: we apply a cyclical-contractor discount and refuse to underwrite the peak multiple into perpetuity. Not a value buy; a fast-cyclical-at-a-full-price buy.
7. Technicals (from the tech block)
- Trend: up but cooling. $1,741 sits below the 50-DMA ($1,883) but well above the 200-DMA ($1,307) — a short-term pullback within a powerful long-term uptrend. MACD +7.4 (still positive).
- Location: −16% off the 52-week high ($2,066), but +230% off the 52-week low ($527) — the max drawdown from peak is −16%, so this is a leadership name that has given back some froth.
- Momentum: RSI(14) 45 — neutral, neither overbought nor oversold. The stock has worked off the overbought condition that accompanied the run.
- Relative strength (the tell): FIX +234% 12-mo vs SPY +20.6% and QQQ +30.3%; +21.9% 3-mo vs SPY +13.7%. Extreme outperformance — which cuts both ways (leadership, but crowded and volatile).
- Read: the −16% pullback and neutral RSI mean you are not chasing an overbought top; a break back above the 50-DMA (~$1,883) would re-confirm the uptrend, while a loss of the 200-DMA (~$1,307) would be the cyclical-warning tripwire.
8. Moat & competitive position
FIX's "moat" is real but modest and cyclical, not structural: (1) scale and geographic density — 197 locations let it staff large, multi-site datacenter and fab projects that smaller local contractors cannot; (2) skilled-labor access — in a trade-labor-short market, having 22,700 trained workers is itself a barrier; (3) execution reputation and prefabrication/off-site capabilities that improve margins and speed. But this is a fragmented, competitive contracting industry with no pricing-power monopoly — the durable edge is operational, and the current super-normal margins are cycle-inflated.
Peer set (market cap): EMCOR $34.5B (the closest large-cap M&E comp), Quanta Services $100B, MasTec $29.5B, APi Group $18.2B, Dycom $13.1B, MYR Group $6.7B, TopBuild $9.9B, Arcosa $7.1B, Granite $6.4B, Primoris $4.8B, Construction Partners $6.1B, Ameresco $1.3B. FIX and EMCOR are the mechanical/electrical pure-plays most levered to the datacenter buildout; FIX commands a premium multiple on its faster recent growth — justified only if that growth persists.
9. Management, capital allocation & guidance
- Capital allocation: disciplined and shareholder-aligned — $280M of bolt-on acquisitions, $216M of buybacks, and a modest dividend ($2.60/share, ~0.15% yield, payout ~6%) in FY25, all self-funded from $1.03B FCF while holding net cash. Buybacks reduced share count (35.35M diluted, down from 35.78M). This is a sensible, opportunistic allocator, not an empire-builder.
- Insider activity: recent Form 4s are a mix of routine director equity awards (2026-05-18) and modest director sales (Myers 6,700 shares at $1,954 on 2026-06-24; Hardy 342 shares at $1,900) plus a gift — normal diversification at elevated prices, no alarming cluster of discretionary executive selling in the sampled window.
- Management's own guidance (half-weighted — their own self-interested words): the SEC 8-K (Q1'26 earnings release, filed 2026-04-23) confirms the hard numbers — net income $370.4M / $10.51 diluted, revenue $2.87B, organic revenue growth +51% YoY, operating cash inflow $388.8M, and backlog $12.45B (vs $6.89B a year prior; same-store backlog $12.21B). CEO Brian Lane's forward language: "Considering recent bookings, underlying persistent demand, and our strong pipelines, we are optimistic about our prospects for the next several quarters." Management provides no specific numeric guidance (no revenue/EPS target range) — the release is qualitative optimism plus a record backlog. Treat the tone as management talking its own book; the backlog figure is the load-bearing, auditable fact.
10. Catalysts & what to watch
- Next earnings: 2026-07-23 (Q2'26; Street EPS $10.38, revenue ~$2.97B). The key lines: backlog (is it still growing, or peaking?) and organic revenue growth and margin sustainability.
- Backlog conversion & bookings: the single most important number — a flattening or declining backlog is the earliest tell that the cycle is turning.
- Datacenter/AI capex commentary: hyperscaler and chip-fab capex plans (the demand engine) — any broad capex-digestion signal is a warning.
- Margin durability: whether the expanded ~25% gross / 17% EBITDA margins hold as the mix normalizes.
- M&A cadence: continued disciplined bolt-ons vs. overpaying at peak valuations.
Thesis tripwires (what would change the call): two consecutive quarters of declining backlog; organic growth decelerating below ~15%; gross margin compressing back toward 21%; or a broad hyperscaler capex-cut signal. Any of these flips this from Tactical-Buy toward Watch.
11. Key risks
- Cyclicality (structural, the big one): construction demand is cyclical and FIX's current growth is riding an AI-datacenter/fab capex super-cycle. Backlog is a demand snapshot, not a multi-year contracted annuity — it can shrink fast if capex digests. Beta 1.67 reflects this.
- Valuation / de-rating: ~50× trailing leaves no cushion; a growth or margin disappointment could compress the multiple and the price simultaneously (the classic cyclical double-hit).
- Customer/end-market concentration in datacenters: the very tailwind driving the beat is a concentration risk — a small number of hyperscaler buildouts drive an outsized share of incremental demand.
- Skilled-labor availability & wage inflation: the binding constraint on growth and a margin risk.
- Fixed-price contract / estimation risk: the company's own 8-K flags "incorrect estimates for bidding a fixed-price contract" and backlog "failing to translate into actual revenue or profits" as material risks.
- No expert coverage: unlike a conviction name, there is no independent panel corroborating this thesis — the call rests entirely on the data.
12. Verdict, position sizing & monitoring
Buy — Tactical. FIX is a genuinely exceptional operating story right now: FY25 revenue +29.5% to $9.10B, Q1'26 organic revenue +51%, EPS doubling, backlog $12.45B (nearly 2× YoY), 52% ROE, net cash, $1.03B FCF, and — rarest of all — growth that is still accelerating. The A- balance sheet and self-funded capital allocation are excellent. But it is a cyclical mechanical/electrical contractor trading at ~50× trailing near a cycle peak, with zero expert corroboration in our KB and a beta of 1.67 — so it earns a tactical, satellite-sized Buy, not a core allocation.
- Sizing: satellite / tactical, ~1–3% of a portfolio — size it as the cyclical it is. The −16% pullback and neutral RSI make current levels a more reasonable entry than the $2,066 high; scaling in (starter now, adds on backlog-confirming prints or a hold of the 200-DMA) beats a lump.
- Monitoring: re-underwrite on the §10 tripwires — backlog direction is the master signal. Formal re-score each earnings print.
- Single biggest risk: the AI-datacenter/fab capex cycle it rides cools; backlog is a demand snapshot, not a contract annuity. This verdict is logged as a tracked Synthos call as of 2026-07-03 at $1,741.30.
Provenance & disclosures
- Traceability: 0 KB claims, breadth 0 — there is no expert coverage of FIX in the Synthos knowledge base, and the note says so plainly. Every number here traces to FMP financials/estimates or the SEC 8-K (filed 2026-04-23). Fabricated conviction is structurally impossible (no claim_ids to cite, and none are cited).
- Data as-of: fundamentals 2026-03-31 (Q1'26) · estimates & prices 2026-07-02/03 · SEC guidance release 2026-04-23. Forward figures are analyst consensus (FMP), labeled as estimates.
- Management caveat: the §9 guidance is management's own earnings-release language, half-weighted by design; management provided no numeric guidance, only a record backlog and qualitative optimism.
- Not investment advice. Independent research, educational and informational only, never personalized. Hypothetical/forward figures are labeled; the only performance numbers Synthos will headline are the live, real-money Flagship's.
- Version: 2026-07-03. Prior versions available via the deep-dive version dropdown ("based on the info at the time").