When: Capitulation below $430 with RSI <30 + panic volume

Then: Deploy bulk of reserve at maximum conviction

HOLD

• 200 SMA extension of 49% above mean is a vulnerability indicator, not a bullish signal — maximum drawdown risk equals the distance to mean reversion
• AMD fell 70% from its 2021 peak despite being in a secular uptrend, demonstrating that semi uptrends break
• Current AI infrastructure trade is the most concentrated in market history, creating fragility rather than invincibility
• RSI below 30 historically produces durable bottoms; late-July sell-off lacked capitulation volume per the technical report
• PEG of 1.01 assumes stable, predictable growth — semiconductor inventory cycles average 3-5 years and growth can turn negative when cycles turn
• AMD's debt profile transformed: from ~$4.3 billion debt and $13 billion cash at Q2 2026, to ~$9 billion debt and ~$9 billion cash after $5 billion bond issuance
• 2.49 beta means the stock will underperform in any risk-off environment by more than double the index move
• Customer concentration is a structural risk — every marquee customer that goes exclusive with Nvidia is a direct hit to AMD's pricing power and market share
• 26% drawdown from $580 occurred within a bullish macro environment, indicating the catalyst is AMD-specific, not macro-driven
• 1.5x benchmark position cap is appropriate given the 2.49 beta and concentration risk

→ vs aggressive: AMD at 49% above the 200 SMA is not a leading stock indicator — it's maximum drawdown risk, similar to Cisco at 70x earnings in 2000 or AMD itself at 40x forward earnings in 2021 before falling 70%

→ vs aggressive: RSI below 30 waiting is not mysticism — it respects documented technical patterns of durable bottoms; the aggressive approach of buying at 47 RSI risks watching the stock drop another 20%

→ vs aggressive: PEG of 1.01 is mathematically correct but practically meaningless — it assumes stable growth, but when semiconductor cycles turn, growth goes from 50% to negative, not to 20%

→ vs aggressive: Investment-grade bond issuance is not 'graduating to the major leagues' — it eliminates AMD's net cash position and increases sensitivity to AI capex disappointment if hyperscaler spending slows

→ vs aggressive: SpaceX-Nvidia exclusivity is not noise — every exclusive arrangement with a major customer directly threatens AMD's AI accelerator market share and pricing power

→ vs aggressive: Falling yields and benign inflation are conditions that historically precede cycle peaks, not beginnings; the 26% drawdown from $580 happened within bullish macro, proving the risk is AMD-specific

HOLD core; add small tranche at 469-475 with confirmations; trim at 505-515

• Core position is sound given secular AI trend, 50% revenue growth at scale, and 14-point gross margin expansion
• Tiered entry: deploy ~25-33% of planned add at current levels if 469-475 zone holds with constructive reversal candle, RSI turns above 50, and MACD histogram crosses positive
• Reserve majority of add capital for capitulation trigger at 430 with RSI below 30 and panic-volume confirmation
• MACD histogram narrowing to -1 signals momentum deceleration that often precedes reversals before RSI hits extreme levels
• Trim into strength at 505-515 relief rally rather than adding; elevated beta (2.49) makes trimming the asymmetric move
• Hard ceiling at 1.5x benchmark weight non-negotiable given beta, leverage transformation, and customer concentration risk
• Trailing stop at approximately 2x ATR (~$65) sets hard floor at ~$420 for any new tranche
• Consider protective puts given relatively low implied vol vs elevated realized vol; options market is calm enough to hedge effectively
• Conservative analyst's RSI below 30 prerequisite is technically rigid; other confirming signals can precede reversals
• Aggressive analyst's dismissal of 50 SMA breakdown as noise is dangerous; price sitting 5.5% below dynamic support is a structural break
• Bond issuance represents fundamental transformation from ~$10B net cash fortress to break-even balance sheet with rising capex
• AMD fell 70% from 2021 peak; survivor bias in aggressive analyst's narrative is intellectually dishonest
• Macro environment (falling yields, record equity highs, benign inflation, 8% recession odds) is structurally supportive but AMD-specific factors drive the thesis
• AMD has already absorbed a 26% drawdown; selling into this consolidation phase would be an error

→ vs aggressive: Dismissing the 50 SMA breakdown as short-term noise is dangerous; price sitting 5.5% below dynamic support is a structural break, not noise

→ vs aggressive: Claim that you'd miss 40% of the rebound waiting for RSI below 30 is an assumption, not a fact; bottoms sometimes form without capitulation

→ vs aggressive: Framing bond issuance as purely strategic asset acquisition ignores the fundamental transformation from $10B net cash to break-even balance sheet

→ vs conservative: RSI below 30 as a hard prerequisite for entry is technical rigidity; MACD histogram narrowing signals momentum deceleration that often precedes reversals

→ vs conservative: 'Complacency precedes the peak' is correlation masquerading as causation; markets bottom in benign environments too

→ vs conservative: Trigger conditions too restrictive; waiting for panic-volume washout below 430 would require ~15% additional decline, likely implying a fundamental thesis break