SNDK-01 — Multiple Storage Stocks Have Flown to the Sky! Is This Cycle Still Going? Looking at Optical Modules and Memory. Opportunity Always Exists — Just Need the Awareness

Date: 2026-03-17 Tickers: SNDK (SanDisk), WDC (Western Digital), STX (Seagate), plus optical/networking adjacent (LWLG, LASR, VIAV, ANET, CRDO, CIEN) Source: https://www.youtube.com/watch?v=reN8CbO3bN8

Summary

• Storage stocks have melted up massively in 2025-2026. SNDK: $40s → $700+ (~14x). WDC (parent before SNDK spinoff): $40s → $200s (5-6x). STX: $90s → $400 (4-5x). All driven by AI-era memory/storage demand explosion.
• SNDK Q4 2025 (calendar) earnings exploded: revenue +62% YoY, gross profit +400%+, operating income +400%+, net income +600%+. The cost-line breakthrough (margins crossed past fixed cost = profit detonation). Gross margin: 30%→50%. Net margin: -110% → +26% net margin in 5 quarters.
• Forward guidance Q1: revenue $4.4-4.8B, gross margin 65-67% (up another 15+ percentage points). EPS could hit $12-14 (vs ~$3-4 currently). At that earnings rate, forward PE compresses fast.
• 2025 fiscal revenue ~$7.3B → 2026 fiscal ~$15B (2x). EPS expected to grow from $2.76 to ~$40. Massive operating leverage.
• Cycle warning: Semiconductor industry is brutally cyclical. SNDK's worst quarter (Q1 2025 fiscal) had -110% net margin. The boom started H2 2025. Historical industry boom cycle ~3 years before destocking pain. We're 1 year in.
• Insider/institutional read: Institutional 81%, retail 12%, insider very low (typical for spinoff). FMR top holder, slight reduction. Vanguard reduced 3%. BlackRock reduced 40% in Q4 2025 (most aggressive trim). Multiple smaller institutions adding (rank 5-10+). Mixed signal.
• Speaker's view: stocks already very high — not buying here. May see consolidation/pullback. Long-term order pipeline is real but valuation is extended. Lesson: catch the bottom of cycles, not the top. Speaker also lists optical/networking names worth watching (LWLG, LASR, VIAV, ANET, CRDO, CIEN) and notes the wafer/foundry sector as next potential beneficiary (HBM memory needs 4-6x more wafer area than DRAM).

Translation

In 2025-2026, memory stocks have rallied 5x to 15x. Today let's look at the storage sector with SanDisk (SNDK) as the example, walk through the financials, and think about cycles.

I'm X. All my videos are personal investment notes. Manage your own risk.

Storage sector winners

I came to this sector late. By the time I noticed, prices were already in the sky.

• SNDK (SanDisk): $40s → $700+ in 2025 (mostly H2 2025) — ~14x
• WDC (Western Digital): $40-50 (H1 2025) → $200+ now. WDC was SanDisk's parent before the 2024 spinoff. 5-6x
• STX (Seagate Technology): $90s (H1 2025) → ~$400 now. 4-5x

All are storage-focused. SNDK is the most extreme runner because it spun off later and started from the lowest base.

SNDK business and structure

SNDK = flash storage (SSDs, NAND), the memory that retains data when power is off. Used to be one company with WDC; spun off in 2024. Both companies were losing money for years pre-AI-boom; now AI demand has flipped them.

Mcap now: ~$100B. Pre-spike was ~$10B (small-mid). Currently still net negative on annual basis but Q4 turned profitable.

Headcount: 11,000+. Closely held shares: 6%.

Financials trajectory

Quarterly revenue trajectory (calendar): - Q1 2025: $1.7B (still YoY decline) - Q4 2025: ~+62% YoY

Q4 2025 = breakthrough quarter: - Revenue +62% YoY - Gross profit: +400%+ YoY - Operating income: +400%+ - Net income: +600%+ YoY

The growth rate divergence (revenue 62%, net income 600%) is operating leverage — they crossed the cost-coverage threshold and incremental revenue mostly drops to bottom line.

Margin trajectory (annual): - Pre-2025: gross margin ~30%, net margin slightly negative - Q1 2025 (worst): net margin -110% - Q2 2025: still struggling - H2 2025: cycle inflection - Q4 2025: gross margin ~50%, net margin ~26%

Forward guide Q1 2026: - Revenue: $4.4-4.8B - Gross margin: 65-67% (up another 15-20 percentage points) - Approaching NVIDIA-level margins - EPS forecast: $12-14

Annual: - 2025 fiscal: ~$7.3B revenue - 2026 fiscal: ~$15B revenue (2x) - EPS: $2.76 → ~$40

Share count is stable (no dilution since spin-off), so EPS growth is real, not diluted.

Why this is the most important lesson

Look at the deep red period before this run-up. In Q1 2025, SNDK was at -110% net margin — losing more than its revenue. The stock was at $40s, no one wanted it.

Then the AI memory demand exploded. By Q4 2025, the cycle flipped. Margin transformation was extreme.

The lesson: semiconductor cycles always end up in dramatic boom periods after dramatic bust periods. What's hard is identifying the cycle bottom and being brave enough to buy when reports look most ugly.

Cycle implications for here / now

We're ~1 year into a cycle that historically lasts ~3 years before the next bust. So 2 more years of potential upside before destocking pain. But: - A lot is already priced in - BlackRock (a major institutional holder) trimmed 40% in Q4 2025 — they've been wrong before but they're not selling without a thesis - Forward PE looks low only because earnings are spiking — if earnings retrace, valuation snaps tighter

Geographic mix

2025 globalization improved: - China: ~25-30% - US: ~15%+ - Hong Kong + EU + Other Asia: ~15% each

More balanced vs prior China-heavy mix. Diversification reduces single-market shock risk.

Business mix

• Cloud (data center): growing fastest, +64% QoQ in Q4 2025
• Edge (terminal inference, future AI): +21% QoQ (steady)
• Consumer: +39% QoQ

Edge growth still building — speaker thinks edge inference is the next big infrastructure phase (after data center buildout).

What to do now

Speaker isn't buying SNDK at these levels — too extended. May see consolidation or pullback. The story has moved from "discovery" to "extension."

Watch list — adjacent under-explored sectors: - Optical / networking: LWLG, LASR (laser/photonics), VIAV (network security/photonics), ANET (Arista — hasn't run yet), CRDO, CIEN - Wafer / foundry: HBM (high bandwidth memory) requires 4-6x more wafer area than DRAM. Demand for more advanced wafer production is rising. Wafer sector could be next.

Reflection

The optical/networking sector started up only in H2 2025 — 2 years after ChatGPT moment. Not all "AI beneficiaries" inflect at the same time. Some lag dramatically. Looking at where we still see deep cycle troughs now gives clues for where the next AI-tailwind unlock will be.

Strategy: find sectors at cycle bottom. Build conviction position. Hold through the storm. Trim small portions on overshoots, accumulate on underwater periods. Cycle awareness is the meta-skill, not picking individual stocks.

OK that's the share for today. Wishing everyone financial freedom soon. Thanks. See you next time. Bye-bye.