ASX

I have enough to refresh this dossier with trader-grade specifics. Key correction vs. the prior version: the Q1 print already happened (2026-04-29), so there's no binary earnings catalyst in the next 30 days — and the stock is now extremely extended (~104% above its 200-DMA, fresh 52-week highs, P/E ~53), with the "Nvidia Halo Trade" narrative going mainstream. Writing the output now.

{
  "body_markdown": "# ASX — ASE Technology Holding Co., Ltd.\n\n## Current Thesis\nAdvanced-packaging/OSAT is the literal bottleneck of the AI-chip buildout, and ASE is the world's largest OSAT now absorbing TSMC's CoWoS overflow (\"CoWoS-like\" CoW capacity). The fundamental narrative is ACCELERATING but the TAPE is MATURING/late: stock is +120% YTD, +264% over 1yr, trading ~104% above its 200-day MA and printing fresh 52-week highs. We'd be buying the \"Nvidia Halo Trade\" leg (memory/packaging/cooling spend rises every time Jensen scales). The catch: the easy re-rate is done, the narrative is now on Benzinga halo-trade listicles (public), and there's no earnings catalyst for ~2 months. A fresh entry at [entry redacted] is a chase, not a fat pitch — the high-conviction add is a pullback to support.\n\n## Bull Case\n- Q1 2026 (printed 2026-04-29): revenue US$5.508B, +17.2% YoY; net income +87% YoY; EPS NT$3.24 +85% YoY; gross margin 20.1%.\n- ATM (assembly/test) segment +29.7% YoY to NT$112.4B — advanced packaging is carrying the mix, not legacy.\n- LEAP advanced-packaging guidance RAISED on the Q1 call to >$3.5B for 2026 (~10% above prior guide).\n- Q2 2026 guide (2026-04-29): consolidated revenue +7–9% QoQ, gross margin +20–100bps, ATM +9–11% QoQ — sequential acceleration into 2H.\n- Record ~$7B 2026 capex funds VIPack (FOCoS / silicon-bridge, ~200x interconnect density vs organic) — structural share in the CoWoS-overflow trade.\n- Cluster confirmation: NVDA 2026 CoWoS demand ~595k wafers (~60% of global); TSMC expanding CoW orders to OSATs in 2026–2027; ASE/SPIL + Amkor + Powertech share ~80k wafers of NVDA CoWoS overflow.\n\n## Bear Case\n- TAPE is extreme: ~104% above 200-day MA ($19.35), ~43% above 50-day ($24.40), and at the 52-week high ($35.71). Mean-reversion target risk is live.\n- P/E ~53 on a historically 12–18x cyclical OSAT — the multiple already prices the AI-packaging story.\n- \"Nvidia Halo Trade\" is now a Benzinga listicle (2026-05-07) — narrative going public = late-stage per the saturation rule.\n- EPS $0.20 ADR was only inline-to-slight-miss vs forecasts on the Q1 print — the bottom line is NOT outrunning the multiple.\n- Amkor's $2.5–3.0B 2026 capex (~3x prior) = capex arms race; CHIPS/reshoring could tilt US-CSP packaging share toward AMKR.\n- AI-capex beta: any CoWoS double-ordering/digestion scare hits the whole OSAT cluster at once — and ASX gapped +6.65% in a single session (2026-05-22) straight into that risk.\n\n## Setup & Price Structure\n- Price $34.81 (2026-05-22 close), +6.65% / +$2.17 on the day, sitting just under the 52-week high of $35.71.\n- 52-week range $9.23–$35.71; market cap ~$78.3B; P/E ~53.\n- 50-day MA $24.40, 200-day MA $19.35 — price ~104% above the 200-day. Parabolic, deeply extended.\n- RSI was 64.6 on 2026-05-12; after the 2026-05-22 pop to highs it is materially higher (est. low-70s) — approaching but not yet a confirmed blowoff (>88).\n- Structure is a clean uptrend with NO recent pullback to support. The high-conviction entry is a retest of the rising 20-EMA (~[entry redacted] est.), not chasing the high.\n\n## Catalyst Calendar (next 30 days)\n- ~2026-06-10 (est.): May monthly revenue release (Taiwan-listed names report monthly sales ~10th) — nearest real catalyst; watch MoM sequential to confirm the +7–9% Q2 guide.\n- Early June: Computex AI-hardware read-through to packaging/CoWoS demand headlines.\n- Ongoing: TSMC CoWoS allocation / CoW-to-OSAT order-release headlines.\n- NO earnings in window: Q1 printed 2026-04-29; Q2 print is ~late July (est. 2026-07-30) — no binary print for ~2 months.\n\n## What Would Change Our Mind\n- Bull→stronger (re-entry): pullback to a rising 20-EMA (~[entry redacted]) that holds, then a breakout retest — would upgrade conviction and justify sizing up.\n- Bull→broken: weekly close below the 20-EMA (~$30), or a break of the 50-day ($24.40) on volume = trend over, stand aside / cut.\n- Thesis-break: Q2 (~2026-07-30) ATM revenue growth below the guided +9–11% QoQ, or any LEAP/advanced-packaging guide cut = acceleration thesis dead.\n- Cluster-break: AMKR / Powertech / SPIL rolling over, or a CoWoS double-ordering/digestion scare — exit the basket, don't single-name hope.\n\n## Correlation Notes\n- Tightly correlated to AMKR (US OSAT peer), TSM (upstream CoWoS allocator) and its own subsidiary SPIL; secondary beta to NVDA AI-capex.\n- This is a CLUSTER trade — ASX, AMKR and Powertech move on the same CoWoS-overflow headline. Don't double-size: ASX + AMKR is one bet, not two.\n- ADR proxy for Taiwan-listed 3711.TW — carries overnight gap risk to the Taiwan session and NT$/USD translation.\n",
  "current_thesis": "Advanced-packaging/OSAT is the AI-chip bottleneck and ASE is the largest OSAT soaking up TSMC's CoWoS overflow — narrative ACCELERATING on fundamentals but the TAPE is MATURING: +120% YTD, ~104% above 200-DMA, at 52w highs, no earnings catalyst for ~2 months. Fresh entry at [entry redacted] is a chase, not a fat pitch.",
  "invalidation_trigger": "Weekly close below 20-EMA (~$30) or break of 50-day MA ($24.40) on volume; or Q2 (~2026-07-30) ATM revenue growth below guided +9–11% QoQ = advanced-packaging acceleration broken.",
  "current_conviction": "MEDIUM",
  "archetype": 2,
  "catalyst_date": "2026-06-10",
  "themes": ["advanced-packaging-osat", "ai-chip-infra-memory", "ai-chip-asia-proxies"],
  "notes": [
    "Q1 printed 2026-04-29 (rev $5.508B +17.2% YoY, ATM +29.7% YoY, LEAP guide raised >$3.5B). Next earnings ~late July 2026 — NO binary print in the 30-day window.",
    "Cluster trade: size ASX + AMKR as ONE bet, not two. Same CoWoS-overflow headline moves the whole OSAT basket (AMKR, SPIL, Powertech).",
    "Tape is extreme: ~104% above 200-DMA, P/E ~53, at 52-week high. High-conviction entry is a 20-EMA pullback (~[entry redacted]), not chasing the high.",
    "Taiwan-listed 3711.TW reports monthly revenue ~10th of month — recurring soft catalyst; watch for MoM acceleration confirming +7–9% Q2 guide."
  ]
}