When: Price at $440–$450 if 50 SMA is tested

Then: Stage deeper value adds

When: Friday NFP is binary catalyst

Then: Plan adds for after the print, not before

HOLD with tightened stops at 10 EMA (499.83) or Bollinger lower band (474.59), stage adds at 474-483 and 440-450 only with confirmation, wait for NFP print and Azure disclosure to clarify path

• RSI bearish divergence confirmed: lower high of 73.5 on Aug 28 vs 77.7 on Aug 10 while price made higher high of 513.53 vs 505.11
• MACD momentum sell signal: MACD line peaked at 30.25 on Aug 11, histogram now at negative 1.85, signal line crossed above MACD Sep 1-2
• EPS decelerating: Q4 2025 EPS 5.16 down to Q1 2026 EPS 4.27 to Q2 2026 EPS 4.81, challenging 31% growth consensus
• PEG ratio of 0.7 based on forward P/E 21.6 and 31% growth is a point estimate not capturing capex risk, AI ROI risk, or margin trajectory
• Operating margins compressed 380 bps from peak 48.9% in Q3 2025 to current 45.1%
• Free cash flow declining from 25.6B to 19.6B while CapEx doubled to 35.8B quarterly; Net PPE grew 107B year-over-year
• Azure dollar disclosure may already be priced at 510; narrowed definition makes historical comparisons harder
• 43M insider sale at all-time highs is a yellow flag regardless of 10b5-1 mechanical nature
• Recession odds at 8% and Mag 7 at records is exactly when next drawdown surprises consensus
• Polymarket Fed cut odds at 1% for September; dovish narrative is 2-4 months away, not next week
• 2x ATR stop of approximately 23.68 should represent no more than 1% of portfolio risk
• Asymmetric risk math: 20% drawdown requires 25% gain to recover; 30% drawdown requires 43% gain

→ vs aggressive: RSI cooling from 77 to 66 is not a 'healthy exhale'; it is bearish divergence with each price high met with weaker momentum thrust, precisely when distributions begin

→ vs aggressive: MACD line still positive at 16.57 is 'technically true but strategically meaningless'; waiting for it to go negative means waiting until damage is done

→ vs aggressive: PEG 0.7 calculation ignores that 31% growth is consensus already priced in; if Q3 2026 misses, the math inverts overnight

→ vs aggressive: Adding at 495 with stop at 499 provides only 4 dollars of risk; that is not a stop, it is a transaction cost

→ vs aggressive: 35.8B quarterly CapEx is not a moat; Net PPE grew 107B and must generate returns; AI monetization at scale unproven

→ vs aggressive: Operating margin trajectory matters more than level; 380 bps compression in 9 months toward 40% means severe multiple compression even at flat revenue

→ vs aggressive: Azure dollar disclosure already priced at 510; if it were a major re-rating catalyst, much is already reflected

→ vs aggressive: Insider sale size of 43M at all-time high zone carries information value; timing of 10b5-1 plan initiation is discretionary regardless of mechanical execution

→ vs aggressive: Dovish Fed narrative is 2-4 months away; being fully positioned into binary NFP catalyst with no confirmation is not prudent sequencing

→ vs aggressive: HOLD is not intellectual cowardice; momentum-investor framing systematically underweights permanent capital loss risk

HOLD with tactical enhancement: maintain core exposure and stage conditional entries at 516 breakout (add tranche, stop at 499) or 474-483 pullback (value entry with stop below lower Bollinger band), sizing positions so 2x ATR equals 1% of portfolio

• Forward PEG of 0.7 is mathematically correct but relies on continued 31% earnings growth assumption that may not hold given quarterly EPS volatility: Q4 2025 EPS 5.16, Q1 2026 EPS 4.27, Q2 2026 EPS 4.81
• Dollar disclosure has bounded marginal value for sell-side modeling; actual significance depends on whether Azure growth accelerates or decelerates
• Current RSI at 66.4 remains firmly in bullish territory; RSI divergences in strong uptrends commonly resolve through consolidation rather than pullback
• Operating margin at 45% represents elite performance; 380 basis points of compression from peak reflects strategic AI infrastructure investment, not deterioration
• CEO insider sale of 43 million at all-time highs joins the mosaic of information but does not drive portfolio action
• Polymarket recession odds at 8% and falling; macro tape is risk-on; catalyst respect via position sizing, not avoidance
• Strategy solves tension between chasing and waiting by defining conditional entry triggers in advance and letting market come to us

→ vs aggressive: PEG calculation of 0.7 is mathematically correct but selectively deployed; forward EPS of 23.57 bakes in 31% earnings growth that quarterly trajectory (5.16, 4.27, 4.81) does not support as smooth compounding

→ vs aggressive: Azure dollar disclosure called major re-rating catalyst overstates marginal information value; multiple has stayed elevated with percentage-based guidance for years

→ vs conservative: RSI divergence argument has merit but is applied too rigidly; current RSI at 66.4 is still bullish, and in strong uptrends divergences commonly resolve through consolidation or breakout rather than pullback

→ vs conservative: Operating margin compression framed as deterioration is misleading; 45% is elite performance and 380bps from peak represents cost of strategic AI platform investment

→ vs conservative: Waiting passively for 474 price target that may never come sacrifices time value; conditional orders solve this by defining triggers in advance