Q2 revenue growth of +25.4% YoY is misleading because it came off a Q2 2025 trough of $12.86B—the lowest point in a multi-year decline—not a peak, making the comparison the easiest comp of the cycle

• Q2 2025 base was $12.86B (trough of multi-year decline)
• +25.4% YoY revenue growth

Gross margin expansion to 40.4% represents catch-up, not differentiation—TSMC runs at 53-58% and AMD at ~52%

• Gross margin expanded from 27.5% to 40.4%
• TSMC runs at 53-58%
• AMD runs at ~52%

The $12.74B balance sheet mark-down and -$11.03B net loss contradict the 'regime change' narrative—real businesses don't print cash and lose money simultaneously

• Q2 net loss of -$11.03B
• $12.74B mark-down on the balance sheet
• Operating income was +$1.97B

TTM operating income averages only ~$0.4B, not the $1.97B Q2 print—with Q2 delivering 73% of TTM operating profit, the business is concentrated in a single quarter rather than diversified

• TTM operating income averaging $1.97B, +$0.93B, +$0.55B, +$0.86B = ~$0.4B avg
• Q2 alone delivered 73% of TTM operating profit

Forward EPS of $2.04 requires $10B+ in net income—7x Intel's historical peak of $1.4B in 2021—and TTM net income is currently -$11.3B

• $2.04 forward EPS
• Intel earned ~$1.4B in net income in its BEST full year (2021)
• Current TTM net income: -$11.3B

The Mobileye write-down signals management's private assessment that crown-jewel investments have fallen 30-50%—CFO knowledge versus bull's interpretation

• Mobileye write-down (management chose to mark it down)
• $12B hit on publicly-traded equity stake (Mobileye trades)
• Management telling you value of crown-jewel investments has fallen 30-50%

Q1 2026 OCF of $1.1B versus Q2 2026 OCF of $7.0B represents a 6x swing in 90 days—cash conversion is highly dependent on working capital movements and one-time items, not steady-state

• Q1 2026 OCF was $1.1B
• Q2 was $7.0B
• 6x swing in 90 days

$37.7B net debt divided by $4.87B TTM FCF equals 7.7 years to repay, with suspended dividend (~$6B/year if restored) extending full FCF debt service to 12+ years

• $37.7B net debt / $4.87B TTM FCF = 7.7 years to repay
• Suspended dividend ~$6B/year if restored
• 12+ years of full FCF devoted to debt service

Total debt of $50.5B at 5.5-6.5% refinancing rates creates $2.8-3.3B/year in interest expense against $5-7B in pre-tax operating power—50%+ of operating profit consumed by interest

• $50.5B in total debt
• High-yield refinancing rates (5.5-6.5%)
• $2.8-3.3B/year in interest expense
• $5-7B in pre-tax operating power

Cumulative CapEx over last 5 quarters is $13.1B, with a multi-year $50B+ capex cycle for 18A and 14A node builds that won't generate revenue for 3-5 years

• Cumulative CapEx over the last 5 quarters: $13.1B
• Multi-year, $50B+ capex cycle for 18A and 14A node builds
• Fabs don't generate revenue for 3-5 years

ARM-based server CPUs are already here, not 2-3 years away—Nvidia Grace and Grace-Hopper in production, AWS Graviton gaining share, Microsoft Cobalt shipping

• Nvidia already ships Grace and Grace-Hopper CPUs (in production today)
• AWS Graviton (ARM) gaining share right now
• Microsoft Cobalt is shipping right now

The $12.9B Nvidia-Hugging Face acquisition boxes Intel out of the open-source AI ecosystem and developer mindshare Intel relies on for AI tooling

• $12.9B Nvidia-Hugging Face acquisition
• Hugging Face is where AI models get built and deployed
• Nvidia now owns that pipe

AMD's data-center share has tripled in three years, AMD is at parity or better on process node, and AMD has full-stack accelerators (MI300/MI350)—Intel faces a 2-front war against both Nvidia and AMD

• AMD's data-center share has tripled in three years
• AMD at parity or better with Intel on process node
• AMD has full-stack accelerators (MI300/MI350)

Dell earnings showed server growth driven by GPU-rich AI configurations with more Grace-Hopper, more NVDA-CPU integration—AI server architecture is shifting away from x86 Intel

• Dell server growth was strong, driven by GPU-rich AI configurations
• More Grace-Hopper, more NVDA-CPU integration, more ARM-based auxiliary compute
• CPU in those servers is increasingly not Intel

The stock at $90 represents a 35% drawdown from the $142 peak—corrections of this magnitude in high-beta names (β=2.23) typically take 6-9 months to base, not 1-2 weeks; we're only 9 weeks in

• Stock at $90 (35% drawdown from $142 peak)
• β = 2.23
• 9 weeks into the correction
• Corrections of this magnitude typically take 6-9 months to resolve

50 SMA at $102.15 is 12% overhead, requiring a 13% rally from $90 to $102 with declining volume from the blow-off top

• 50 SMA at $102.15
• 12% overhead
• Declining volume from blow-off top

MACD at -3.31 with signal at -3.28—'imminent' crossover has meant 4-6 weeks, histogram flat-negative for entire correction, momentum bottoming not turning

• MACD at -3.31 with signal at -3.28
• Histogram flat-negative for entire correction
• Momentum bottoming, not turning

ATR compressed from $10.65 to $5.22—compressed ATR after parabolic move historically resolves with volatility expansion event, typically gap-downs on macro or earnings catalysts

• ATR compressed from $10.65 to $5.22
• Compress ATR after parabolic move historically resolves with volatility expansion event
• Gap-downs on macro or earnings catalysts, not gap-ups

Coatue's ~$1.1B position equals 0.22% of Intel's $506B float—hedge fund satellite position, not high-conviction institutional validation, and could represent entire accumulation not start of flood

• Coatue's 12.08M shares at ~$90 = ~$1.1B position
• Intel's market cap: $506B
• Coatue stake: 0.22% of float

The $200 PT requires $20 EPS in 2031 (~50% annual EPS growth from current run-rate)—five-year EPS forecasts at growth semis are wrong roughly 70% of the time

• $200 PT requires $20 EPS in 2031
• ~50% annual EPS growth from current run-rate
• Five-year EPS forecasts at growth semis are wrong roughly 70% of the time

Polymarket shows Fed cut by Sept 2026 = 0%, Dec 2026 = 6%, rate HIKE odds >60%—not priced for hawkish Fed with inflation re-accelerating

• Fed cut by Sept 2026 = 0%
• Fed cut by Dec 2026 = 6%
• Rate HIKE odds >60%

Every 100 bps yield increase = 5-7% multiple compression for INTC at 46.9x forward P/E with $37.7B net debt at floating exposure—multiple compression hasn't happened yet but bill is coming due

• 46.9x forward P/E
• $37.7B net debt at floating exposure
• Every 100 bps yield increase = 5-7% multiple compression

The stock already moved from $24 to $142 (5.9x return in less than 12 months)—easy money made, early buyers sitting on 100-300% gains are systematic sellers on strength, every rally into $100+ is distribution not accumulation

• Stock moved from $24 to $142 (5.9x return in less than 12 months)
• Early buyers sitting on 100-300% gains
• Every rally into $100+ is distribution event, not accumulation

Bear base case: forward EPS comes in at $0.80-1.20 (not $2.04), Q3 is flat-to-down sequentially, foundry fails to announce major Tier-1 wins in 2026, 200 SMA at $73.57 tested within 3-6 months

• Forward EPS: $0.80-1.20
• Q3 is flat-to-down sequentially
• Foundry customer pipeline fails to announce major new wins in 2026
• 200 SMA at $73.57 tested within 3-6 months