GTX

Current Thesis

Boring turbo cash-cow trying to mint an AI-adjacent narrative leg: Garrett is pushing its high-speed turbomachinery IP into data-center / industrial oil-free centrifugal cooling (Ingersoll Rand multiyear deal 2026-05-12, Trane Technologies collab, China Refrigeration Expo 2026-04-08 portfolio launch, 7–500 ton range, >10% energy savings). The problem for our playbook: the binary that mattered — Q1 CY2026 — already fired on 2026-04-30 (EPS $0.49 vs $0.41 est, revenue $985M vs $913M est, FY guide raised to ~$3.75B). The stock popped to RSI ~85.7, sell-side clustered PT raises mid-May (Stifel $36, JPM $33, BWS $32), and Benzinga flagged it an overbought "ticking bomb" 2026-05-12. We'd be buying a MATURING story after the catalyst and after the analysts caught up — the exact opposite of the 3-6-weeks-early entry this account exists for.

Bull Case

• Q1 CY2026 beat + raised guide (2026-04-30): EPS $0.49 vs $0.41 consensus; net sales $985M vs $913M est, +12% YoY (+6% cc); Adj EBIT margin 15.3%. FY26 guide raised to $3.6–3.9B sales / $300–360M net income / $355–475M adj FCF.
• Genuine narrative bridge to AI infra: oil-free centrifugal compressors target data centers, BESS and industrial cooling — the one credible path from "ICE turbo supplier" to an AI-power/thermal multiple. Two name-brand validators in <5 weeks: Trane and Ingersoll Rand (2026-05-12).
• Aggressive buyback compounder: $250M 2026 authorization; $87M repurchased in Q1, $163M remaining as of May 2026; >$100M total returned in Q1 incl. dividend. Shrinking float on a high-FCF base.
• Sell-side leaning bullish: Stifel Buy $36, JPM Overweight $33 (both 2026-05-14), BWS $32 — Strong Buy consensus across the bullish cohort.

Bear Case

• We're late, not early. Catalyst (Q1) is spent; RSI hit 85.7 on the print and the analyst upgrades are already published. Narrative-momentum edge requires front-running sell-side, not riding it down the back half.
• Theme tag is unstable / forced. This dossier has cycled GTX through 6 themes in 3 weeks (autonomous-vehicles → industrial-power-ai → ev-auto-supply → consumer-discretionary → small-cap-ai-momentum). When the classifier can't find a home, the "AI momentum" frame is manufactured, not real.
• Cooling revenue is option value, not a number. Data-center compressors are expo/partnership-stage; core P&L is still cyclical light-vehicle + commercial turbos. This is a value/buyback story wearing an AI costume.
• Cheap-and-cyclical = value-trap risk in this book. ~8–9x earnings, auto-SAAR sensitive. Multiple is low because the tape and structural ICE decline cap it — exactly the "cheap multiple, no narrative velocity" profile the playbook calls a trap.
• Overbought flag on record: Benzinga "ticking portfolio bombs," 2026-05-12.

Setup & Price Structure

• Last ~$25.87 (June 2026). Sits below the bullish PT cluster ($32–36) but above Deutsche Bank's Hold [entry redacted] — i.e. priced near the skeptics' fair value, not a coiled breakout.
• RSI cooled from 85.7 (late April, post-print blowoff) to ~54 by mid-May — momentum has already bled off; no fresh thrust.
• No clean higher-low breakout-retest in play; the move we'd be chasing is the April earnings gap that's now ~5 weeks old. Entering here is buying chop, not strength.

Catalyst Calendar (next 30 days)

• No dated binary inside the window (by ~2026-07-04). Q2 CY2026 earnings is the next real catalyst, est. ~late July 2026 — outside 30 days.
• Possible un-dated tape movers: incremental data-center / HVAC cooling order or partnership PRs (Trane/Ingersoll Rand follow-through). Watch, don't pre-position.
• Ongoing buyback execution ($163M remaining) — a slow bid, not an event.

What Would Change Our Mind

• Re-entry (bull): a fresh, dated data-center-cooling revenue/order catalyst OR a clean weekly higher-low + breakout-retest above the April highs with RSI re-thrusting from the 50s — then a LOW→MEDIUM probe is defensible.
• Stay out / invalidate: weekly close below the ~$24 20-EMA shelf (Deutsche PT / Q1 breakout base); FY26 sales guide cut below the $3.6B floor; Q2 (~late July) shows zero industrial-cooling traction, confirming "AI" is costume not cash.

Correlation Notes

• Auto-supply beta: moves with BorgWarner (BWA), Aptiv (APTV), Phinia (PHIN) and global light-vehicle SAAR / commercial-vehicle cycle.
• Data-center thermal cohort: the aspirational comp set — Vertiv (VRT), Trane (TT), Johnson Controls (JCI), Modine (MOD). GTX trades on auto multiples today; the bull thesis is a partial re-rate toward this group.
• Small-cap buyback-compounder factor: float-shrink + high FCF — correlates with rates/risk-on for small caps, not with the AI-mega-cap complex.