I'll start by searching for ARM Holdings' ticker and getting the latest option chain data.Now let me process this data. I need to get the current stock price to identify the right strike range. Let me also load recent price data.ARM's latest close is ~$289.73. Now let me compute the premiums as a % of strike for puts in the range you want — strikes between 25% and 50% of current price (75% OTM to 50% OTM), which at ~$290 means roughly strikes from $72.50 to $145, but let me use the exact range: 25% of $289.73 = ~$72.43 and 50% = ~$144.87.

Let me now compute this properly using the option chain data.Good, now let me pivot this into a proper table format for display.Here's the put premium landscape for ARM Holdings — current price ~$289.73, strike range $75–$145 (75% OTM to 50% OTM), across the next 11 months of expirations. Premiums shown as % of strike price (mid-market, bids must be non-zero).

ARM Put Premiums (% of Strike) — Sep 2026 to Jun 2027

Key observations:

• The sweet spot for the best premium-to-time tradeoff is around the $135–$145 strike zone, where Jun 17 2027 puts yield 11.5%–13.5% of strike — roughly 330 days out.
• At the $125–$130 range (~44% of spot), you can collect 3–5.5% by Dec/Jan (5–6 month tenors) and 7–11% out to March/June 2027.
• Shallowest premiums: Sep/Oct expirations are very thin even at the top of the range (0.5%–2.8%), meaning time decay hasn't built meaningful premium yet for these deep-OTM strikes.
• The Jan 15 2027 expiration offers a decent midpoint: 5–6.8% at $125–$145 strikes (~6 months).
• Blanks (—) indicate no bid was present. Many sub-$105 strikes in nearer expirations have no two-sided market.

Let me pull the data needed: shares outstanding, net income (quarterly for TTM), and current price.At $75 per share, here's the math:

TTM Net Income (GAAP):

GAAP P/E (TTM) at $75: 88.6x

Even at a 75% drawdown to $75, ARM would still trade at nearly 89x TTM GAAP earnings — reflecting the thinness of current profitability relative to the revenue base. For context, at the current ~$290 price, the TTM GAAP P/E is roughly 343x.

Let me get more details from the earnings call transcript and the analyst detail page.Here's a summary of ARM's official guidance and analyst consensus:

ARM Holdings — Official Company Guidance (from Q4 FY26 call, May 6, 2026)

Q1 FY27 (quarter ending Jun 30, 2026): - Revenue: $1.26B ± $50M (~20% YoY growth) - Non-GAAP OpEx: ~$760M - Non-GAAP EPS: $0.40 ± $0.04 - Royalty and licensing revenue both expected up ~20% YoY

Full Year FY27 (ending Mar 31, 2027): - ~20% revenue growth across both royalty and licensing - Licensing growth backend-weighted (~60% in 2H) - OpEx growing sequentially by a few percent each quarter, but trailing revenue growth → incremental margin expansion by year-end - AGI CPU revenue: ~$90M in Q4 FY27 (first production silicon revenue)

Long-Term (FY31): - $15B AGI CPU revenue + $10B IP revenue = $25B total - EPS > $9

Analyst Consensus Estimates (from MarketWatch / StockAnalysis / Chartmill):

Q1 FY27 Quarterly Consensus: - EPS: $0.42 (vs guidance $0.36–$0.44) - Revenue: $1.27B (vs guidance $1.21–$1.31B)

Key Commentary from Management: - Data center royalty revenue continues to more than double YoY; expected to double again in FY27 - AGI CPU: $1B guided (constrained by supply), but $2B+ of visible demand across FY27–28 - AGI CPU gross margins: ~30%+ on first-gen products - 50+ ecosystem partners endorsed the silicon strategy (Meta, SAP, Cloudflare, AWS, Google, NVIDIA, OpenAI, etc.) - Arm expects to be the largest CPU architecture by market share by end of decade

Analyst consensus sits slightly above the midpoint of guidance for Q1. For the full year, the Street models ~$6B in revenue and ~$2.17–$2.40 in non-GAAP EPS. ARM reports Q1 FY27 results on July 29, 2026.

Good catch — there's an important distinction here. Let me clarify what management actually said:

Royalty revenue has two very different components growing at very different rates:

1. Data center royalty — This is the Neoverse/CSS-driven royalty from hyperscaler chips (Graviton, Axion, NVIDIA Grace/Vera, etc.). Rene Haas explicitly said this is doubling year-on-year and he expects it to double again in FY27. This is their fastest-growing segment and will soon be their largest.

2. Total royalty — This includes smartphones, automotive, IoT, networking, and data center combined. Total royalty grew only 11% in Q4 FY26 (tough MediaTek comp) and is guided to ~20% for FY27.

So the "doubling" is being diluted by the slower-growth segments. The smartphone/mobile royalty business — still a very large portion of total royalties — faces unit headwinds (management said mobile unit growth "flipped to negative" last quarter). Lower-end smartphones where Arm has weaker royalty rates are the weakest spot. Memory royalties are also soft.

The net effect: data center doubles (+100%), but it's being dragged down to ~20% blended by slower/negative segments.

Rough math to illustrate: If data center is, say, 25% of royalties and doubles, that's +25 points of growth. If the other 75% is flat to slightly down, you land around +20% blended — which matches the guidance.

The bullish thesis here: if and when data center becomes the majority of royalties (which Haas said it soon will be), the doubling math starts to dominate the blended number and total royalty growth should accelerate meaningfully above 20%. The question is just timing — when does the mix shift far enough.

The exact split isn't disclosed in standard financial filings. Let me search for it specifically.Here it is — from the Q3 FY26 earnings call transcript, CFO Jason Child gave the most explicit breakdown:

ARM Royalty Revenue Split (from Q3 FY26 call, Feb 2026):

Key CFO commentary: - Data center was "just approaching double-digit" at start of FY26; by Q3 FY26 it was "well into double-digit… teens to probably getting closer to 20%" - Smartphones still the dominant segment at 40–45% of royalty revenue - Data center expected to match or exceed smartphones within 2–3 years - On smartphone unit risk: a 20% drop in mobile units would only translate to ~2–4% impact on smartphone royalties (because the low-end, low-royalty phones are the ones that get cut), and ~1–2% impact on total royalties — more than offset by data center growth

So the math becomes clear:

Data center (~18% of royalties) doubling = +18 points. But smartphones (~42%) now growing at maybe mid-single digits (unit headwinds offset by ARMv9/CSS rate uplift) = +3–5 points. Automotive/other (~40%) growing low-double digits = +4–6 points. Blend = roughly low-to-mid 20s — exactly the FY27 guidance.

The real inflection comes when data center crosses ~30%+ of royalties: at that point, even without accelerating its doubling rate, it alone delivers 30 points of growth.

Actually, no — the company hasn't given formal full-year FY27 revenue guidance yet. What they've given is Q1 FY27 guidance (~20% growth), and then qualitative commentary about the full year. Let me be precise:

What management actually said on the Q4 FY26 call (May 6, 2026):

• Q1 FY27: Revenue $1.26B ± $50M (~20% YoY), with both royalties and licensing "up around 20%"
• Full year FY27: Royalty growth "in the roughly 20-ish percent range" and "license and royalty… both will be relatively close to the 20s percent range"
• Licensing will be backend-weighted (~60% 2H / 40% 1H)
• OpEx growing sequentially a few percent each quarter, trailing revenue

So management's directional signal is roughly 20% revenue growth for FY27, but it's qualitative framing, not formal guidance. They haven't given a specific FY27 revenue range — something they typically update as the year progresses.

Analyst consensus is slightly higher:

So the Street is roughly at 21–23%, in line with management's "20-ish percent" framing. But the formal FY27 number likely gets firmed up when ARM reports Q1 on July 29 — that's when they typically provide the full-year outlook.

Key nuance: The $1B AGI CPU revenue (spread across FY27–28, mostly FY28) is incremental on top of the base 20% IP/royalty growth. If supply opens up and that $2B demand pipeline converts, FY28 could see meaningful acceleration beyond 20%.

The supply chain is TSMC, not Broadcom.

Here's the breakdown:

Foundry (manufacturing): - TSMC is the exclusive foundry for the AGI CPU. It's built on TSMC's 3nm process (N3), with the first-generation chip using a dual-chiplet design with 136 Arm Neoverse V3 cores. - TSMC's 3nm capacity is the binding constraint — their lines are at 100% utilization, with CoWoS advanced packaging lead times exceeding 14 months. - Future generations are expected to move to TSMC 2nm, but Arm is exploring Samsung Foundry and Intel Foundry for diversification.

Manufacturing partners (broader supply chain): - ASE Holdings — packaging and testing - Quanta Computer, Supermicro, ASRock Rack, Lenovo — server/rack integration and finished systems - Samsung, Micron, SK hynix — memory (HBM and DRAM) - Alchip, Global Unichip, Faraday — design services/ASIC support

Broadcom's role: Broadcom is on the ecosystem support list but is not manufacturing the AGI CPU. Broadcom's relevance here is different — they're the ASIC design partner for Meta's MTIA accelerators (Meta's AI inference/training chips that will pair alongside the AGI CPU in Meta's data centers). Broadcom also has a long-standing relationship with Arm as a licensee.

The bottleneck in summary: It's a pure TSMC wafer allocation problem — same constraint hitting everyone in AI silicon. Arm has $2B+ of demand but can only secure enough 3nm wafers to deliver ~$1B in the FY27-28 window. The constraint cascades through CoWoS packaging, HBM memory, and test capacity as well.

Great questions. Let me pull together what's been disclosed on both fronts.

1. When does explosive growth hit?

The timeline is pretty clear from management's own words:

The explosion is FY29–31, not FY27. FY27 is the "crossing the chasm" year — first silicon revenue, customer validation, supply chain qualification. FY28 is the first real volume year but still constrained. The real hockey stick starts when 2nm allocation opens up, CoWoS lead times normalize, and the second-gen chip ships.

Haas was explicit: they have $2B+ demand for FY27–28 but are deliberately capping guidance at $1B because they can't get the wafers. That's demand, not adoption risk.

2. FY27 margins and GAAP net income

Let me pull the actual data to compare growth rates.Here's the full picture on margins and what FY27 should look like:

ARM Historical GAAP Margins (Annual):

FY24 was the IPO year — massive stock-based compensation crushed operating income. FY25–26 show normalization but operating margins have been compressing (20.6% → 18.5%) as R&D spending ramps aggressively.

What happens in FY27:

Management explicitly guided to incremental margin expansion by year-end FY27. Key drivers:

So net-net: GAAP net income should grow faster than revenue in FY27. If revenue grows ~20% to ~$5.9B and net margin expands from 18.4% toward 20% (closer to FY25 levels), you get net income in the $1.15–$1.25B range — roughly 27–38% growth, well ahead of the 20% revenue growth.

The gross margin will dip modestly (AGI CPU at 30% GM vs 92%+ for IP/royalties), but operating leverage on the SG&A/R&D base more than compensates. The real margin story is FY29+ when AGI CPU scales and its gross margin improves toward 35% as second-gen chips arrive on 2nm.

Fair point — SBC isn't going away. But it matters where it shows up and what's driving the FY24 distortion. Let me be precise:

FY24 wasn't "normal SBC" — it was the IPO.

ARM went public in September 2023. FY24 (ended March 2024) absorbed a massive one-time SBC charge from the IPO — roughly $500M+ in stock-based compensation that hit operating expenses. That's why operating income collapsed from $678M in FY23 to just $99M in FY24 while revenue actually grew 21%.

FY25–26 are the normalized run-rate. SBC didn't disappear; it just reverted to ongoing levels:

The GAAP vs Non-GAAP gap (roughly 20 percentage points in FY26) is largely ongoing SBC — and that's structural. It won't vanish.

So to your question — will GAAP net income grow in line with revenue?

Not exactly. Ongoing SBC will keep GAAP margins below non-GAAP. But the FY27 dynamics still point to GAAP net income growing faster than revenue:

• FY26 GAAP net margin: 18.4%
• FY27: revenue ~$5.9B at ~20% growth
• OpEx growing slower than revenue by year-end → operating leverage
• AGI CPU adds ~$90M revenue at ~30% GM — small dilution this year
• Likely GAAP net income: $1.1–$1.2B (22–33% growth vs 20% revenue growth)

The structural SBC load means GAAP margins won't snap back to 25%+ anytime soon. But the directional improvement from operating leverage is real — expenses grow slower than revenue in FY27 for the first time in the recent investment cycle. That's the margin story, SBC included.

Here's the analyst landscape on ARM's valuation — the consensus is remarkably split:

How Analysts Justify (or Don't) ARM's Market Cap

Current price: ~$267 | Market cap: ~$285B | Trailing P/E: ~315x | Forward P/E: ~120x

The Bull Case: "You're buying FY31, not FY27"

The bulls (Rosenblatt $270, Jefferies $290, KeyBanc $430, UBS $470, BofA $460) argue ARM should be valued on FY31 numbers, not FY27:

At the FY31 target: $9 EPS × 50x = $450. $9 EPS × 40x = $360. Current ~$267 is ~30x FY31 EPS — the bulls say that's reasonable for a company with 97% gross margins and a licensing moat.

The bull math: the stock isn't trading at 315x trailing — it's trading at ~30x terminal earnings with a 5-year growth bridge. The $25B FY31 target implies a 38% revenue CAGR from FY26.

KeyBanc's $430: assumes AGI CPU hits $15B, IP hits $10B, terminal multiple 45-50x. UBS $470 & BofA $460: similar but with higher probability weighting on management execution.

The Bear/Neutral Case: "The price forecloses the return"

HSBC (downgraded to Hold, PT $315): Stock surged 122% since March. At 139x FY26 / 95x FY27 earnings, "the narrative is overhyped." Even hitting targets barely justifies the current price.

A.L. Capital (Avoid, DCF $100–$386): P/E 394x is 1,132% above sector median. DCF midpoint $232 vs $331 — negative margin of safety at -27%. Beta of 3.77 means extreme drawdown risk (CVaR -25% in a bad month).

Market Supply Chain (detailed scenario model):

The killer observation: even in the bull case, a $348 entry delivers only ~9% annualized — barely compensating for the risk. The entry price itself forecloses most of the return.

Simply Wall St DCF: fair value ~$63/share (69% overvalued).

What's Actually Priced In

The most cited framework across analysts is Morgan Stanley's probability-weighted approach:

Weighted outcome: ~$6–8B — materially below management's $15B. MS fair value: ~$150.

The Key Risks Analysts Flag

1. Supply, not demand: $2B orders but only $1B wafers secured — the missing $1B is the real near-term test
2. FTC antitrust investigation: could force royalty rate reductions across the $6B+ licensee base
3. RISC-V: Google and Alibaba experimenting with custom RISC-V server cores
4. Qualcomm dispute: unresolved architectural independence claim could eliminate a top-5 licensee
5. Rate sensitivity: at 120x forward P/E, a 100bp rise in the 10-year compresses DCF value by 15–25%
6. AGI CPU margins: at ~30% gross margin vs 97% IP margin, the blended margin will decline meaningfully as AGI CPU scales

Consensus Summary

The real answer: ARM's valuation only works if you believe management's FY31 targets — and the stock is priced as if the bull case has ~70% probability, while most analysts weight it at 30%. The July 29 Q1 report — specifically data center royalty re-acceleration and any supply chain updates — is the next major test of which side is right.