Conviction: Medium Status: Researching (not held)

Prices as of the Friday 2026-07-24 close. Prior note: the 2026-06-22 entry used the 2026-06-18 close (06-19 Juneteenth holiday, 06-22 unpublished).

2026-07-31 tape note: MPWR $1,316.18 (+2.67% vs $1,282.01), RSI 45.7 (warmed from 41.8). 7D -5.80% (moderated from -7.32%), 30D -4.79% (deepened from -2.19%), 3M -18.35% (deepened from -15.91%). From the 52-week high, the gap is -23.21% (narrowed from -25.21%). Per the tape: trend "weak-down", regime "pullback", golden cross intact. No new fundamentals re-researched. Per the full-scan summaries (2026-07-31 close). 2026-08-08 tape note: MPWR $1401.54 (+6.49% vs $1316.18), RSI 53 (warmed from 45.7). 7D -1.72% (moderated from -5.80%), 30D +6.54% (turned positive from -4.79%), 3M -12.32% (moderated from -18.35%). From the 52-week high, the gap is -21.51% (narrowed from -23.21%). Per the tape: trend "weak-down", regime "pullback", golden cross intact. No new fundamentals re-researched. Per the full-scan summaries (2026-08-07 close). 2026-08-14 tape note: MPWR $1402.01 (+0.03% vs $1401.54), RSI 52.7 (cooled from 53). 7D +0.03% (turned positive from -1.72%), 30D +3.65% (moderated from +6.54%), 3M -9.41% (moderated from -12.32%). From the 52-week high, the gap is -20.51% (narrowed from -21.51%). Per the tape: trend "strong-up", regime "pullback", golden cross intact. No new fundamentals re-researched. Per the full-scan summaries (2026-08-14 close).

---

Editorial Note

The 25-year supporting actor that finally stepped into the lead role got a tape confirmation this week. As of this 2026-06-22 refresh: price $1,563.70, RSI 50.8 — back at the neutral midline after a strong week. The 7D -5.27% print is the trailing-window optics from a low base earlier in the period, but the 06-18 session was a +8.0% single-bar move (close $1,563.70 vs $1,448.21 the prior day) — MPWR ripped on Friday alongside the broader silicon/WFE/memory bid. The stock has now reclaimed SMA20 from below, sitting essentially at SMA20 ($1,575.93, -0.8%) and +1.1% above SMA50 ($1,546.42). 30D is -1.64% (roughly flat — still digesting) while 3M has accelerated to +46.3% (vs +40.79% on June 16) — the Vera Rubin socket-share thesis keeps strengthening on the longer frame even as the stock chops near its moving averages. -12.56% from the 52-week high — same as three weeks ago, so the Friday rip recovered the recent dip but has not yet repaired the larger drawdown. The Vera Rubin thesis is intact; this week's silicon-rotation bid pulled MPWR back to its moving averages from the lower bound of the entry zone.

2026-07-11 tape check: The silicon-rotation bid fully reversed. MPWR is back to $1,352.74, RSI 45.2, and has fallen back BELOW both SMA20 ($1,421.36, -4.8%) and SMA50 ($1,516.83, -10.8%) — a wider gap below the moving averages than even the June 16 reading. 30D has deteriorated sharply to -14.77% (from -1.64%) and 3M has flipped from +46.3% to -1.27%. Drawdown from the 52-week high widened to -21.08% (from -12.56%), and price is now BELOW the $1,450-$1,580 entry zone entirely, sitting only ~6.8% above the $1,267 stop-loss level. Nothing in the Rubin socket-share fundamentals has changed (thesis flag still reads intact) — this looks like the silicon/WFE/memory basket giving back its June rip, the same rotation dynamic flagged in the 06-22 note, just running in reverse. Watch the $1,267 stop closely; it is no longer a distant hypothetical.

2026-07-17 tape check: The slide has continued at a calmer pace. MPWR $1,312.00, RSI 43.9 (down modestly from 45.2), still below both SMA20 ($1,365.55, -3.9%) and SMA50 ($1,492.11, -12.1%) — the SMA20 gap has actually narrowed slightly. 7D -3.01%, 30D -15.97% (deepened further), 3M -11.86% (worse than -1.27%). -23.46% from 52wk high (widened from -21.08%), and price is now only ~3.5% above the $1,267 stop-loss level — closer than two weeks ago. thesis flag: intact unchanged; no new fundamentals re-researched. The $1,267 stop remains the level to watch.

2026-07-25 tape check (Friday 2026-07-24 close): The slide has reversed and SMA20 is essentially reclaimed. MPWR $1,333.81 (+1.7% vs $1,312.00), RSI 45.3 (up modestly from 43.9) — now only -0.3% below SMA20 (narrowed sharply from -3.9%) and -9.1% below SMA50 (narrowed from -12.1%). 7D has flipped positive to +1.66% (from -3.01%), 30D has improved to -6.91% (from -15.97%), though 3M has deepened to -18.15% (from -11.86%, a base-effect drag as the June rip rolls further out of the window). -18.49% from 52wk high (narrowed from -23.46%) — price is now ~5.3% above the $1,267 stop-loss level, a bit further away than two weeks ago. thesis flag: intact unchanged; no new fundamentals re-researched. Price remains below the $1,450-$1,580 entry zone, but the bounce off the stop-loss approach is constructive.

2026-07-28 tape check: The reclaim has failed and the stop is closer again. MPWR $1,282.01 (-3.9% vs $1,333.81), RSI 41.8 (down modestly from 45.3) — back to -4.1% below SMA20 (from -0.3%) and -12.2% below SMA50 (from -9.1%). 7D has reversed to -7.32% (from +1.66%), 30D has improved to -2.19% (from -6.91%, base effect), and 3M has deepened to -15.91% (from -18.15%, roughly flat). -25.21% from 52wk high — widened from -18.49%, back near the deepest drawdown on record for this profile. Price is now only ~1.2% above the $1,267 stop-loss level — the closest approach yet. thesis flag: intact unchanged; no new fundamentals re-researched. Watch the $1,267 stop closely into the next session.

---

The Story Right Now

The arithmetic of the last few centimeters is brutal: Blackwell B200 needs ~1,500A on the core rail, Vera Rubin pushes ~2,875A, and to hold 1,500A within 20mV requires a path resistance of ~13 micro-ohms. That's not a taste question, it's physics. NuttyCLD's AI Power Crisis series (six parts, March 30 → May 1) frames MPWR as the protagonist of a 25-year arc that just turned cinematic — the multi-phase VRM, the boring 12V → 0.8V step that nobody cared about for two decades, is now the choke point determining whether NVIDIA's most expensive silicon can actually run at spec. Nutty's intersection metaphor is the cleanest portfolio frame in the whole series: MPS sells the entire traffic-control system at the intersection. Not a single component, not just the controller — the integrated power module, the controller, the DrMOS, the qualification, the supply discipline. That's a different income statement than AOSL ("first into the lane") or Infineon+Delta ("underground entrance").

The lead-change arc is what makes this trade interesting and dangerous at the same time. Hopper had MPS comfortable and dominant. Then GB200 hit 700W+ and the PMIC thermal issues that already existed in Hopper became blocking — Infineon's Vertical Power Delivery (paired with Delta's system integration) won an estimated 60-70% of GB200 PMIC sockets in 2024, with Renesas filling more of the gap. MPS Blackwell exposure cratered, the sell-side repriced, and the stock spent most of 2024 in the doghouse. Then GB300 happened — KeyBanc's September 2025 channel checks said MPS had recovered ~50% B300 HGX share after fixing the thermal architecture, with $900M-cap AOSL emerging as co-leader via OpenVReg reference position. The PT went from $750-ish to $1,050 in one analyst note. By March 2026, KeyBanc raised again to $1,250 on supply-chain work suggesting ~70% Vera Rubin socket share at ASPs ~60% higher than Blackwell — annualizing into 2027 at roughly $420M of incremental revenue and material EPS lift.

Q1 2026 (reported April 30) confirmed the floor. EPS $5.10 (non-GAAP; GAAP diluted was $3.92) vs $4.90 expected, revenue $804.2M vs $782M expected (+26% YoY), and management raised the enterprise data segment growth floor from 50% to 85% YoY — which is the data-center exposure ramping. Q2 guide implies sequential +12% to ~$900M. That print is the first hard read on the Vera Rubin socket-share thesis materializing in the income statement, and it landed clean.

As of this week: RSI 50.8 is back at the neutral midline. The 06-18 session's +8.0% rip pulled the stock off the lower bound of the entry zone and back to SMA20 ($1,575.93), which is now overhead-to-flat rather than the -4.9% gap of three weeks ago. The 3M of +46.3% (up from +40.79%) confirms the broader thesis trajectory is still accelerating. The entry zone from prior analysis ($1,450-$1,580) now brackets current price almost exactly — MPWR sits in the middle of it. The Friday rip is silicon-rotation participation, not a thesis-specific catalyst — the entry setup remains "in the zone, at the moving averages, neutral RSI."

---

Quick Snapshot

---

Action Matrix

---

Price Data

---

Company Overview

One-Liner

Fabless analog/mixed-signal chipmaker that designs the power management ICs and integrated voltage regulator modules feeding everything from GPUs and EVs to consumer electronics and data-center AI sockets.

Business Model

Key Segments (FY2025)

---

Competitive Analysis

Competitive Moat

Competitor Map (the five-company intersection)

Moat Width: Wide. Moat Trend: Widening (Rubin socket share gain at higher ASP) but architecture-cyclical — the moat resets every NVIDIA generation. This is not yet a market with a final winner. It is a market where the definition of winning keeps changing from one generation to the next.

---

Management Assessment

Capital allocation: Conservative — organic growth, zero long-term debt, $3.5B equity, debt/equity = 0, 28% Q4 2025 dividend hike. Red flag: insider net -$751.9M trailing (mostly automated 10b5-1 sales into the parabolic).

---

Financials

Quality: FCF positive ($93M Q1); GAAP profitable; zero long-term debt; net cash; quarterly $2/share dividend funded organically.

---

Valuation

Analyst targets: KeyBanc $1,250 (-20%), Wall St avg $1,294 (-17%, below current), high $1,500 (-4%), Rosenblatt $1,575 (+1%), low $1,000 (-36%). The dispersion is the tell — consensus PT is below current price; today's price already prices the Rubin upside. Fair value zone: $1,200-$1,400 for a starter; $1,600-$2,000 if Rubin ASP exceeds 60% premium AND Feynman socket position holds.

---

Bull / Bear

Bull: (1) Rubin socket-share dominance is incumbency, not luck — ~70% socket projection is a long-cycle asset; (2) ASP expansion >> unit growth (+60% ASP structural, Morgan Stanley: Kyber rack power-solution value >10x vs GB200); (3) cross-cycle EV+AI exposure; (4) fortress balance sheet, founder-led compounder DNA.

Bear: (1) priced for perfection at forward P/E ~52x — Q2 miss triggers repricing; (2) Rubin slips to 2027; (3) architectural disintermediation (VPD or direct conversion); (4) insider selling cluster (-$751.9M) signals a tactical top; (5) auto cyclical drag. Thesis killers: AOSL/Infineon takes >50% Rubin socket share; Feynman 2028 socket goes to a competitor; op margin below 22%.

---

Catalysts

---

Key Changes vs 2026-07-28

---

2026-06-22 Rotation Refresh

MPWR is the power-delivery silicon expression of this week's down-stack rotation. The tape bid silicon-equipment (AMAT/KLAC/LRCX) and memory (MU/MRVL) hard while de-rating AI-monetized-in-SaaS, and MPWR's +8.0% single-session rip on 06-18 was participation in that silicon-complex bid — it re-rated with WFE/memory, not on a Vera-Rubin-specific catalyst. That's the important nuance for this name: where BE (+19.3%/7d, the cohort's electrons leader) and VRT (power/cooling recovery) moved on the watts/electrons side of the rotation, MPWR moved on the silicon side — it's the bridge between the two, the chip that turns watts into the voltages NVIDIA's silicon actually needs. The market is paying for "watts + silicon" and de-rating "apps," and MPWR sits at the intersection of watts and silicon, which is exactly the NuttyCLD "intersection" thesis on the tape. Durable read: the move was rotation-driven, not thesis-confirming — the thesis-confirming prints are still the Q2/Q3 Vera Rubin commentary. So this week pulled MPWR's technicals from "below SMA20, lower bound of entry zone" to "at SMA20, mid-zone, RSI neutral" — a firming, not a breakout, and the cheap-entry window from June 16 has narrowed but not closed. If the silicon basket rotates back out, MPWR gives back this week's rip with it; the floor is the still-accelerating 3M trend and the intact Rubin socket thesis.

---

Research Log

---

Open Questions

• How much of the Q1 +85% enterprise-data guide is Rubin pre-build vs Blackwell sustain?
• What's the Feynman 2028 socket-share read? (TSMC A16 Super Power Rail interaction with VRM topology)
• AOSL co-leader trajectory on Rubin — does the $900M-cap small-co retain reference position?
• Should we open a vrm-pmic watchlist (MPWR, AOSL, VICR, IFNNY, NVTS, POWI, RNECY, ON, ADI, TXN)?
• The -$751.9M insider net — does a parsed view (10b5-1 vs discretionary) change the signal?

Sources

• Tape data (price/RSI/SMA/returns): full-scan pre-computed summaries as of the 2026-07-31 close — *.
• Prior-cycle history: this file's dated update log (refreshed each full scan).