When: Daily close >$256.19 (50 SMA) on >35M shares with contracting MACD histogram and RSI >50

Then: Close out 50% of put coverage, pause trim execution

When: Post-Q3 earnings beat with AWS growth ≥17.5% YoY AND constructive 2026 capex/OI guidance

Then: Close 50% of puts, consider 1% add

When: Flush to $238-241 (200 SMA retest) on volume climax with bullish reversal candle (RSI <35, MACD histogram narrowing)

Then: Close out 50% of puts, consider 1-1.5% reload

When: Anthropic S-1 day-one valuation ≥$1.5T

Then: Pre-earnings re-rating catalyst — close 50% of puts, consider add

When: $260 confirmed breakout with MACD/RSI confirmation

Then: 1% add, close 50% of puts

When: $238-241 flush on volume climax with bullish reversal

Then: 1.5% reload, close 50% of puts

When: Weekly close <$240 on >60M shares confirms 200 SMA structural break

Then: Trim additional 20% of original position, exercise puts

When: Q3 earnings miss with AWS growth <16% YoY OR 2026 capex guidance walked back to <$150B

Then: Trim additional 20%, escalate put coverage to 85-90%

When: 30Y Treasury yield breaking above recent 2002 highs combined with hawkish Fed rhetoric

Then: Trim additional 20%, escalate put coverage

HOLD

• CapEx of $131.8B in 2025 and $44.2B in Q1 2026 drove FCF negative by $18.17B in a single quarter - this is speculation, not risk management
• Forward EPS compression to $10.48 from trailing $12.44 represents a ~16% earnings decline that cannot be dismissed as 'accounting noise'
• Q2 2026 EPS of $5.75 likely includes a one-time gain of $15.68B that may not be recurring, flattering the trailing picture
• Multiple binary catalysts (Q3 earnings, Prime Day, Anthropic IPO) in a short window compounds risk rather than reducing it
• Prediction market data shows 91% probability above $245 but only 4% odds of $260 by Oct 5 and 7% odds of $300 by end of October - indicating range-bound, not asymmetric upside
• Technical setup is neutral: RSI at 48.56, ATR contracting to $5.63 (lowest in 30 sessions), VWMA still declining - volatility contraction is not a directional signal
• Fortress balance sheet ($143B cash, $441.9B equity) protects against permanent loss but not a 15-20% drawdown; Beta is 1.44 means 40-50% larger moves than S&P 500
• Google-Anthropic dynamics show Google at 58% and climbing in October with Anthropic falling - near-term pressure on AWS-Anthropic economics until December recovery

→ vs aggressive: FCF being framed as 'moat being constructed in real time' is a thesis, not a fact - Amazon's balance sheet makes the burn survivable but doesn't prove the CapEx will generate expected returns

→ vs aggressive: The ~16% forward EPS decline cannot be dismissed as 'accounting noise' or 'depreciation timing quirks' - dozens of analyst models have priced this in

→ vs aggressive: Prediction market data was read selectively - the 91% floor was highlighted while ignoring the 4% odds of $260 and 7% odds of $300, which collapsed from 50% one week ago

→ vs aggressive: MACD crossover below zero is 'notoriously prone to false signals' and was incorrectly read as a bullish trigger when RSI at 48.56 is neutral

→ vs aggressive: VWMA still declining (sellers in control on volume-weighted basis) was skipped over entirely as an 'inconvenient data point'

→ vs aggressive: Google at 58% and climbing in October with Anthropic falling represents real near-term AWS pressure - dismissing it as temporary means paying full price today for a December recovery that may not materialize

Maintain HOLD with trigger-based framework: add tranche on confirmed breakout above $260 with RSI above 55, trim on test of $241 with bullish reversal, downgrade below $245; prepare to add tactically on positive Anthropic IPO catalyst within 30-60 days.

• Amazon CapEx spending $131.8 billion in 2025, Q1 2026 burned $44.2 billion with FCF negative $18.17 billion
• Forward earnings expected to compress 16%, forward P/E expanded to 24x vs 20x trailing
• Prediction market: 91% probability AMZN holds above $245 through October 5, only 4% odds of hitting $260
• $300 October strike collapsed from 50% to 7% in one week on prediction market
• RSI at 48.56 is neutral, not bullish; MACD crossover at -2.20 is early signal prone to failure
• VWMA declining from $260.38 to $250.65 indicates volume-weighted sellers remain in control
• Beta of 1.44 supports 15-20% drawdown risk in risk-off scenario
• December Polymarket shows 55% odds Anthropic re-takes AI model lead
• Operating cash flow of $139.5 billion annually provides capacity to absorb CapEx cycle
• 200 SMA still intact, structural uptrend unbroken despite medium-term bearish trend
• Break above $256.54 on 50 SMA reclaim with volume confirmation is high-probability entry signal

→ vs aggressive: Treating unproven AI infrastructure as moat being constructed in real time creates blind spots; asking to commit capital at $251 betting on returns 2-3 years out while stock below 50-day MA is leap of faith, not thesis

→ vs aggressive: Prediction market shows 91% probability of $245 floor but also only 4% odds of $260 and $300 strike collapsed to 7%; this is range trade setup with hard floor and soft ceiling, not asymmetric upside

→ vs aggressive: Forward EPS depreciation drag is recurring and predictable, not accounting noise; dismissing consensus forward estimate because it doesn't fit narrative is intellectually inconsistent

→ vs aggressive: Cherry-picking bullish technicals (MACD crossover, RSI recovery, Bollinger squeeze) while skipping VWMA decline from $260.38 to $250.65 shows confirmation bias

→ vs conservative: Binary catalysts cut both ways; base case is mixed execution with upside and downside, not uniform disappointment risk

→ vs conservative: Beta concern of 1.44 is valid but position sizing already accounts for 10-15% drawdown tolerance; risk is managed through sizing, not through action

→ vs conservative: Google vs Anthropic model ranking is near-term sentiment headwind, not structural thesis breaker; AWS-Anthropic commercial relationship persists regardless of standalone model performance

→ vs conservative: Anthropic IPO in 30-60 days at $2 trillion valuation could be meaningful mark-to-market event; trader should have plan to add on pullback from IPO profit-taking