When: Below $440 on capitulation volume

Then: Add 25% sizing

HOLD

• MACD histogram printed 8 consecutive sessions of sequential decline from plus 11.29 on August 7th to minus 0.61, representing textbook momentum decay rather than normal digestion
• 50 SMA below 200 SMA indicates structural damage from November drawdown has not been repaired; absence of structural confirmation is itself a risk
• Polymarket data shows 48% probability MSFT trades at or below 472.50 by week's end versus only 7% probability of hitting 525, indicating downside skew not upside skew
• Polymarket pricing 38% probability of 10-year breaking 5% by year-end and 85% probability of zero Fed cuts in 2026
• Quarterly capex doubled year-over-year to 35.8 billion representing 110% increase in capital intensity; 4-quarter average FCF in 2026 materially below 2025
• PEG at 1.61 already pricing in AI success; trailing P/E of 26.8 and forward P/E of 20 with compressed FCF differs materially from Amazon 2014 comparison
• ATR of 13.54 supports 455 stop (1.5-2 ATR buffer) rather than aggressive analyst's proposed 430-440 stop which would absorb 12-15% drawdown
• 30-year yield hit 19-year high; stagflation-lite backdrop pressures long-duration assets and compresses multiples
• Adding at 478-482 zone would mean buying near 52-week highs before binary catalyst clears; relative strength during selloffs often precedes leading lower
• Conservative tiered add zones are 468-475, 455-465, or below 440—not 481—after catalyst clears on confirmed price action

→ vs aggressive: MACD histogram rolled over with 8 consecutive sessions of decline from plus 11.29 to minus 0.61—this is regime shift momentum decay, not noise or 'digestion'

→ vs aggressive: Polymarket shows 48% probability of trading below 472.50 by week-end versus only 7% probability of hitting 525—this is downside skew, not upside skew; the asymmetry math uses base rates the market disagrees with

→ vs aggressive: Capex doubled from 17 billion to 35.8 billion in 12 months (110% YoY); FCF compression is a fact, not an opportunity, especially with softer Anthropic prints and enterprise IT budget pressure

→ vs aggressive: Stop at 440-430 is 3-4 ATR below price, accepting 12-15% drawdown before action—that is hoping, not risk management for a conservative mandate

→ vs aggressive: Adding at 481 into a name near 52-week highs with MACD rolling over, binary catalyst in 8 days, 10-year pressing 5%, and Polymarket pricing 48% downside is catching falling knives, not compounding capital

→ vs aggressive: Relative strength during sector rotation tells about positioning, not direction—mega-cap names that hold up in selloffs often lead lower when relative bid disappears

→ vs aggressive: Amazon 2014 comparison ignores that MSFT is at 26.8 trailing and 20 forward P/E with compressed FCF—not the valuation setup Amazon had

HOLD core position; trim 10-12% into Jackson Hole; maintain 455 stop; pre-commit to tiered adds at 468-475 if support confirms

• Current price: 481, ATR: 13.54
• Stop: 455 (1.9 ATR below current price) - conservative analyst correct, aggressive 440 stop rejected as 'hope not risk management'
• Trim: 10-12% into Jackson Hole (moderate vs conservative's 25% or aggressive's 0%)
• Add strategy: Tiered adds at 468-475 zone if price action confirms support - reject adding at 481 and reject waiting for 455-465
• MACD: 8 consecutive sessions histogram decline from plus 11 to minus 0.61 with price closing below 10 EMA for first time in rally
• Relative strength: MSFT held up during brutal risk-off week with 19-year highs in long-end yields and geopolitical escalation - meaningful signal not to be dismissed
• Valuation: 20x forward earnings on 31% grower with 40% net margins, PEG at 1.61 - fair not cheap, supports HOLD not aggressive buying
• FCF: 35.8B quarterly capex is known strategic cost; 30M Copilot seats, 100B Azure run rate, Snowflake and IREN momentum suggest utilization tracking ahead
• Polymarket: 48% probability MSFT trades at or below 472.50 this week is a coin flip, not confirmation of bearish asymmetry (52% probability does NOT)
• Jackson Hole base case largely priced - trim modest to avoid friction cost if neutral/dovish and forced to chase 500

→ vs aggressive: Stop at 440 is 8.5% drawdown before acting - ATR-calibrated 455 is correct, 440 is a margin call waiting to happen, not 'room to breathe'

→ vs aggressive: MACD rollover from plus 11 to minus 0.61 over 8 sessions with price below 10 EMA is not 'normal digestion' - that's momentum leadership rotating

→ vs aggressive: Adding at 481 into a stock that just lost its 10 EMA with MACD rolling over is poor timing regardless of structural thesis

→ vs conservative: 25% gross exposure reduction into a catalyst where base case is already largely priced will lock in guaranteed friction cost with no clear risk-adjusted improvement

→ vs conservative: Polymarket 48% probability at or below 472.50 describes a coin flip on direction over a short window, not definitive bearish asymmetry - 52% probability it does NOT

→ vs conservative: Relative strength during brutal risk-off week is a concrete signal about institutional positioning - dismissing it as 'irrelevant' is inconsistent reasoning