When: RSI >50

Then: Third signal confirming bullish momentum; all three required

When: All three technical conditions met simultaneously

Then: Re-engage with long exposure; treat rallies as distribution opportunities until then

HOLD - preserve capital until technical confirmation (break above 506 with volume and rising RSI, or capitulation flush)

• 50 SMA has been declining for three straight weeks from roughly 514 down to 506, signaling intermediate-term trend change rather than healthy pullback
• Four of five momentum and trend indicators (lower highs, lower lows, declining 50 SMA, declining 10 EMA, MACD negative, RSI below 50) pointing same direction
• Support levels at 450, then 429, then 200 SMA at 335 if bear case plays out
• Beta of 2.49 means position sizes must be controlled; 8% move from 465 to 429 requires 8.5% round trip plus bottom timing
• Forward EPS estimate of 15.45 implies ~300% growth from TTM 3.91; a 15% miss puts EPS at 13.13 with potential PEG expansion above 1.5
• Stock is 20% off highs but still 39% above 200 SMA, leaving room for mean reversion to 380 or lower
• September is historically worst month for Nasdaq-100; no urgency to be early with high-beta name that just lost 50 SMA
• Institutional flows showing rotation away: ARK trimming, Barron's favoring Nvidia, Zacks preferring Broadcom, CNBC framing AMD as 'chipping away'
• Current drawdown of 20% from highs is insufficient for 'sellers exhausted' thesis; exhaustion occurs after significant moves lower
• Tariffs specifically target data center servers and gaming consoles—AMD's two largest revenue lines and biggest profit pools

→ vs aggressive: Pattern recognition argument is survivorship bias—prior 10/50 EMA touches occurred within intact uptrend with rising 50 SMA; current setup shows actively declining 50 SMA, which is early-stage trend reversal, not healthy dip

→ vs aggressive: Squeeze resolution depends on price action, not narrative; MACD negative, RSI below 50, lower highs/lower lows, and drying volume all suggest lower path of least resistance

→ vs aggressive: Tariff exposure is company-specific, unresolved, and directly hits data center and gaming revenue—the company's largest profit pools—even if headlines eventually fade

→ vs aggressive: Add-on-at-429 strategy ignores September seasonality headwind, 2.49 beta exposure, and potential flush to 380 (18-20% loss) before bottom is confirmed

→ vs aggressive: Forward P/E 30 and PEG 1.03 pricing bull case without reserving for estimate cut cycle into Q3; forward 15.45 EPS implies heroic 300% growth that is vulnerable to guidance reset

→ vs aggressive: 'Sellers exhausted' argument is dangerous—sellers exhaust at bottoms AFTER significant price decline, not before; stock still 39% above 200 SMA with room to mean-revert

→ vs aggressive: September seasonality dismissed as 'timing consideration' but represents structural headwind that should keep conservative allocation patient for 3-4 weeks

→ vs aggressive: Competitive narrative ignores institutional signals—ARK selling, Barron's favoring Nvidia, Zacks preferring Broadcom, CNBC framing AMD as 'chipping away'

HOLD with tightened stop at $429; trim existing positions in the $480-490 zone; do not initiate new positions at $465; await re-entry triggers at $506 breakout or $430-440 flush

• Revenue grew 50% YoY in Q2 2026 with gross margins expanding from 39% to 54%
• Free cash flow is $8.84 billion annualized with net cash position grown from $2 billion to $8.83 billion in twelve months
• Forward P/E of 30 with PEG of 1.03 represents reasonable valuation, not heroic; forward EPS estimate of $15.45 is supported by underlying business momentum
• 50 SMA has been declining for three straight weeks; MACD negative for a full month; RSI at 44.83 is neutral but trending down
• Symmetric risk profile: $506 reclaim level is 9% away; downside to $429 capitulation low is 8% away
• Realistic bear case is $429-$450 zone (3-8% downside), not the 20-30% drawdown to the 200 SMA at $335
• September is historically weak seasonal window for Nasdaq-100 with elevated beta risk (beta 2.49)
• AMD has concentrated product exposure in data center servers and gaming consoles—categories specifically named in tariff policy scope
• For existing holders: maintain position size but tighten stop; trim 15-20% at $480-490 resistance on any rally
• For new capital: wait for daily close above $506 with volume and RSI crossing above 50, OR flush to $430-440 zone on volume spike

→ vs aggressive: The 'every buying opportunity since 2023' comparison ignores critical context—moving average slopes and MACD/RSI conditions at those times were fundamentally different from current deteriorating trend conditions

→ vs aggressive: The 'who's left to sell' argument is flawed—absence of buyers in a high-beta name with headline overhang creates downside vacuum, not upside catalyst; markets go up when there's a bid, and there's no bid right now

→ vs aggressive: Tariff dismissal ignores AMD's concentrated exposure in data center and gaming consoles—exactly the categories named in Lutnick policy scope—while Nvidia and Broadcom have more diversified product exposure

→ vs conservative: Fundamental picture is too dismissively treated—50% revenue growth, 54% gross margins, $8.84B FCF, and net cash position growing to $8.83B represent genuine structural strength supporting the forward EPS estimate

→ vs conservative: Not all four momentum indicators are bearish—RSI at 44.83 is neutral (not bearish) and signals the move has been absorbed; MACD histogram at -0.24 is essentially flat, indicating momentum exhaustion, not building downside

→ vs conservative: Seasonal argument cuts both ways—September weakness means lower expected returns floor, not guaranteed losses; waiting for confirmation above $506 risks missing a 5-10% squeeze move with worse entry and worse momentum

→ vs conservative: The 20-30% drawdown scenario to 200 SMA at $335 requires breaking $429 capitulation low, $400 round number, and multiple support levels—this requires fundamental AI capex thesis breakdown, not just continued technical deterioration; probability is much lower than implied