When: Close below $290 on volume ≥4M shares

Then: Full exit review

When: Berkshire trims AXP in 13F mid-November

Then: Full exit or tactical sell consideration

When: Forward EPS revision below $19.00

Then: Full exit or tactical sell consideration

When: Close above $334 (death-cross reclaim) on volume ≥4M

Then: Tactical add consideration (requires 2 of 4 triggers)

When: MACD histogram contracts with close above 10 EMA AND RSI >40

Then: Tactical add consideration (requires 2 of 4 triggers)

When: Q3 clean print with all three metric categories benign

Then: Tactical add consideration (requires 2 of 4 triggers)

When: Fed dovish pivot compressing credit-cycle risk premium

Then: Tactical add consideration (requires 2 of 4 triggers)

Trim 25 to 40 percent now; hard stop at $290 on residual; cap residual exposure below 2 percent of portfolio; add back only on confirmed technical and credit improvement.

• Agrees credit cycle direction is a legitimate concern (8 percent loan book expansion funded by 7x net debt increase with 12 to 24 month lag) but critiques treating it as certainty rather than probability
• Forward EPS of 20.18 is likely optimistic; EPS compresses to 17 or 18 as bear case but wide range of outcomes exist between 17 and 20.18
• Forward P/E of 15.2x on potentially overstated EPS implies 17 to 18x P/E if EPS marked down 10 to 15 percent for normalized credit losses
• Moat factors (IBD trust ranking, 190 million merchant locations, 34 percent ROE, Buffett presence) provide valuation floor of 14 or 15x vs 9x for subprime lender if earnings compress
• September 22nd session printed 5.1 million shares on the way down indicates institutional distribution not quiet drift
• Late-cycle slowdown (8 percent recession probability) vs recession distinction means trim is precautionary not panic
• Business Savings and AI payroll launches are near-term rounding errors but provide future re-rating optionality on residual position
• Buffett presence is quality signal on 5 to 10 year horizon not short-term protection; using it to argue against trim misapplies long-term signal to near-term risk

→ vs aggressive: Credit cycle presented as certainty rather than probability; EPS compresses to 17 or 18 is bear case not base case; moat dismissed entirely is mistake when it provides valuation floor

→ vs neutral: Wait for confirmation approach is worst option as confirmation in credit-led downtrend frequently arrives with 5 to 10 percent gap lower on earnings miss; adding on weakness before Q3 is adding into uncertainty rather than adding with information

Trim 35-40% now; hard stop residual at $290; reserve further trims at $331 only on confirmed technical reversal; cap residual below 2% of portfolio; wait for Q3 earnings to clarify before adding back

• Trim 35-40% (higher end of 25-40% range) given severity of technical breakdown below 50-day SMA at 331 and 200-day SMA at 332
• Hard stop at $290 is appropriate for residual position
• Residual cap below 2% of portfolio is appropriate given concentrated single-stock, single-sector consumer credit exposure
• Further trims at $331 only valid with confirmed technical reversal: declining volume on bounce, RSI divergence, and reclaim of 10 EMA at $310
• Add back only when BOTH technicals and credit metrics confirm a base, not one or the other
• Aggressive analyst correct on directional risk but wrong on certainty of timing; 41% D/E ratio (net debt $13.88B vs equity $34.3B) is within normal range for diversified lender, not a wipeout scenario
• Conservative analyst correct on optionality framework but wrong on residual size given stock is 25-27 points below key moving averages with accelerating negative MACD on volume
• Credit cycle probability framing preferred over certainty framing; structure should work across moderate and severe cycle outcomes
• Portfolio concentration argument supports trim independently of credit cycle materialization

→ vs aggressive: Presenting credit cycle as certainty rather than probability; 7x net debt increase (1.87B to 13.88B) implies leverage event the absolute numbers (41% D/E against 34.3B equity) don't support; framing Q3 earnings as binary bad/worse outcomes is too bearish and presents bear case as base case

→ vs conservative: Too comfortable with residual size given stock is 25-27 dollars below 50 and 200-day SMAs with accelerating negative MACD; moat limits but doesn't prevent multiple compression in credit shock (AXP traded single-digit forward P/E in 2008-2009); relief rally to 331 on heavy volume would be a lower high, not a buy signal