SPXC

Current Thesis

The breakout the last dossier was waiting for has FIRED. Two weeks ago (2026-05-21) SPXC sat at ~$205, below its 200-day and stuck mid-range — a "wait for it" name. As of 2026-06-04 it's $234.39, having ripped ~14% in two weeks, reclaimed the 200-day (~$210), and pushed to within ~5% of the $246.68 52-week high. The fuel is the 2026-04-30 Q1 beat+raise where management hiked 2026 data-center cooling growth from +50% to +70% and reported HVAC backlog +67% YoY. This is the textbook momentum-realignment setup: ACCELERATING narrative + confirming price + sell-side cluster. Strength IS the setup — we buy the confirmed breakout, we do not sit it out for being "extended."

Bull Case

• 2026-04-30 Q1 beat+raise: Adj EPS $1.69 vs $1.56 est (+22.5% YoY); revenue $566.8M vs $558.5M est (+17.4%). FY26 guide raised: Adj EPS $7.75–$8.15, revenue $2.575–$2.645B, Adj EBITDA $600–$625M.
• DC cooling guide hiked +50%→+70% on the 2026-04-30 call; BMO Capital (Daniel DiCicco) confirmed early May the data-center segment now expected +70% YoY, ~12% of total sales. The narrative is accelerating, not maturing.
• HVAC backlog $755.3M vs $451.3M YoY (+67%; +38% organic). OlympusMAX is the most successful product launch in company history — integrated cooling engineered for high-density GPU racks, $50M of 2025 bookings plus multi-year customer commitments.
• M&A scaling the platform (Q1 2026): closed Thermolec (~$35M rev, Montreal electric duct heating) and Air Enterprises + Rahn Industries (~$300M deal, ~$80M combined rev, serving healthcare/institutional/data-center air handling).
• Capacity build: $100M+ expansion incl. a 459,000 sq ft Madison, AL facility; ~$550M incremental data-center revenue potential by 2028, ~$700M total incremental capacity.
• Sell-side cluster confirms: JPM Overweight PT raised $260→$270 (2026-05-06); Truist Buy $261 (2026-05-04); Wells Fargo OW $240 (2026-05-01); BMO raised on DC growth. Consensus PT ~$249–$273, revised up ~14% over 3 months.
• Price confirmation: +4.2% on 2026-06-02 to $230.08, broke above the 200-day, stock +57% over the trailing year.

Bear Case

• Buying into resistance: at $234.39 (2026-06-04) we're only ~5% below the $246.68 52-week high after a ~14% two-week rip. A fresh entry here chases right into the prior high with no clean pullback entry and wider defined risk.
• Valuation full: P/E 45.86 TTM (2026-06-04), ~29x the FY26 EPS midpoint (~$7.95). GuruFocus GF Value $176.70 — stock ~33% above "fair." Thin margin for error if DC growth normalizes.
• Margin pressure is real: HVAC margin pressured by start-up costs/inefficiencies on the TN/KS/AL capacity ramps (flagged on the 2026-04-30 call). Growth is being bought with near-term margin — an execution slip on the ramp is the bear trigger.
• Second-order dependency: SPXC moves on hyperscaler-capex sentiment it doesn't control. If VRT/ETN/peers guide data-center cooling down, SPXC re-rates with the group regardless of its own backlog. "Peak data center" fear is already circulating in the commentary (TIKR, May 2026).
• No catalyst for ~8 weeks: next print is ~2026-07-30. Post-rip dead-tape / mean-reversion risk while it digests near the highs.

Setup & Price Structure

• Spot $234.39 (2026-06-04, 11:18 EDT), ~flat on day; 2026-06-03 range $230.02–$244.27; 2026-06-02 +4.2% to $230.08.
• Regime flip: reclaimed the 200-day (~$210) — the exact long trigger the prior dossier flagged. Two weeks ago (2026-05-21) price was $205 below the 200-day; now confirmed uptrend.
• 52-week range $152.80–$246.68; price in the upper ~third, ~5% off the high, +57% over 12 months.
• 50-day rising through ~$210–$215; price extended above it. RSI elevated (~70s post-rip) but not yet the RSI>88 blowoff that would matter for an a2 name.
• Entry tactics: best R/R add is a pullback that HOLDS the $210–$215 breakout shelf / 50-day on lighter volume. A weekly close to NEW highs >$246.68 on expanding volume = continuation trigger toward JPM $270. A naked chase at $234 into resistance has the breakout in its favor but limited room before it must make a new high.

Catalyst Calendar (next 30 days)

• NONE hard in the 30-day window. Q2 2026 earnings ~2026-07-30 (est.) — the next binary, outside the window.
• ~June 2026 (est.): industrials/electrical-equipment conferences — watch for data-center-cooling commentary and any DC capex tone shift.
• Peer prints/guides (VRT, ETN, nVent, AAON) through June — SPXC trades with the cohort on hyperscaler-capex sentiment.
• Continued analyst PT revisions (JPM/BMO/Truist/WF clustered early May) bleeding into June would confirm narrative acceleration.

What Would Change Our Mind

• Failed breakout (exit): weekly close back below $210 (the reclaimed 200-day) = breakout failed → exit, do NOT average down below the shelf.
• Narrative break at Q2 (2026-07-30): DC growth guide cut below +70%, HVAC backlog declines QoQ from $755.3M, or HVAC margin compression worsens beyond ramp start-up costs.
• Theme flip: peers (VRT/ETN) guide DC cooling down or "peak data center" becomes consensus → theme → SATURATED → trim.
• Bull-confirm (size up): weekly a daily close below the thesis-invalidation level on expanding volume → new-high continuation toward $270.
• Trim discipline if long: this is an a2 picks-and-shovels name — trim only on RSI>88 with structural cracks (peer underperformance, news drying up), a weekly close below the 20-EMA, or theme→SATURATED. Otherwise HOLD through the rip toward $246+; do not take profits just to feel smart.

Correlation Notes

• Data-center-cooling cohort: trades with VRT (Vertiv), ETN (Eaton), nVent (NVT), Comfort Systems (FIX), AAON — moves on hyperscaler-capex sentiment more than its own backlog day-to-day.
• Second-order to AI capex: tracks NVDA + hyperscaler (MSFT/AMZN/GOOGL/META) capex headlines; a hyperscaler capex cut hits SPXC even with backlog intact.
• Grid/power-AI adjacency: shares the "physical AI buildout" macro with GEV, ETN, POWL. Note: the Serenity DPA list tagged SPXC under "Transformers," but the actual driver is data-center cooling HVAC (OlympusMAX), not power transformers — keep the theme framing accurate.
• Low correlation to software/semis beyond shared capex sentiment; behaves more industrial-cyclical.