When: Close >$172.80 with MACD positive for 2+ sessions

Then: Add back toward benchmark weight

When: Flush to $145-$155 on capitulation volume >5M with RSI <30 and MACD positive divergence

Then: Add back on oversold signal

When: Q4 2025 earnings confirming forward EPS >$8.00 and FCF >$2.0B

Then: Upgrade toward Hold

When: Q4 2025 earnings showing FCF <$2.0B or PFAS charge >$200M

Then: Downgrade toward Sell

SELL with trim toward 50%, 157.50 as reassessment level, reject HOLD

• Deterioration phase with MACD negative for 24 sessions and 50 SMA rolled over; calling it neutral hides asymmetry
• Asymmetric risk worse than framed: 12% upside to 185 requires reclaiming 50 SMA at 171, Bollinger mid at 172, and 10 EMA at 166 in sequence against deteriorating macro
• Downside to 138 (2025 spring low) is clean measured move target if 200 SMA fails
• 2-3% residual position too meaningful given FCF funding capital returns through balance sheet drawdown, negative tangible book, and dividend already cut by more than half once
• 157.50 stop is approximately 4.4% below current price or 2 ATR (ATR 3.59); in hawkish Fed regime with rising recession odds, ATR can expand and stop can be overrun in single session
• Debt service channel risk: 12.6 billion debt against 85 billion market cap is 15% enterprise-level leverage; refinancing costs rise at 19-year high 10-year yield with Fed on hold
• Deleveraging becomes mechanically harder when borrowing costs rise and FCF compressed at 1.4 billion; dividend most vulnerable line item in 2027 if management must choose
• Second dividend cut would be sentiment event, not just cash flow event; residual position would absorb that shock
• Capital preservation is not just about this trade; freed capital should be earmarked for defensive positioning, higher-quality balance sheet names, or waiting for FOMC clarity
• Conservative path: act now with discipline, accept missing upside if support holds, prioritize certainty of risk reduction over uncertainty of waiting for confirmation that may arrive too late

→ vs neutral: HOLD at 158-159 support zone asks to absorb overnight and event risk while hoping for technical pattern the market may never deliver in time; history does not protect capital, and with 19-year high in 10-year yield and no Fed cuts priced in, gaps are common and stops do not work in gaps

→ vs aggressive: 2-3% residual position too meaningful given FCF funding returns through balance sheet drawdown, negative tangible book, and dividend already cut by more than half; residual would absorb second leg down and dividend cut shock

→ vs aggressive: 157.50 hard stop is technically defensible but inappropriate given hawkish Fed regime with rising recession odds where ATR can expand quickly; should be treated as soft alarm level for real-time reassessment rather than hard line

→ vs aggressive: 12% upside to 185 assumes full bull case but requires tape to reclaim 50 SMA at 171, Bollinger mid at 172, and 10 EMA at 166 in sequence against deteriorating macro; conviction outruns risk discipline

SELL with 30-40% trim, hard stop at $157.50, reassess at next earnings/confirmed close below $159/retest of 50 SMA above $171

• 200 SMA at $159 has institutional significance with Bollinger lower band at $158.94 creating confluence that concentrates orders
• AI/data-center thesis is a multi-year secular tailwind for industrial exposure through cooling, filtration, and electronics materials
• Institutional sentiment is mildly bullish anchored by fresh CEO event on September 15 about data-center exposure and 2027 target progress
• 3M deleveraging is positive: 25% debt reduction from peak, debt at $12.6 billion against $5.9 billion in cash, current ratio 1.24
• 30-40% trim balances protection and optionality better than 50% trim; marginal protection from larger trim is small but marginal cost in missed upside is meaningful
• Hard stop at $157.50 with 2-3% residual cap addresses gap risk through position sizing rather than stop softening
• Defined reassessment window (next earnings, confirmed daily close below $159, or retest of 50 SMA above $171) prevents residual from becoming a forgotten position

→ vs aggressive: Dismissing 200 SMA as 'just a number' ignores institutional and algo behavior; confluence matters because it concentrates orders at specific prices

→ vs aggressive: 'Priced in' argument is unfalsifiable—you can always claim any good news is priced in after the fact, which is narrative convenience not analytical rigor

→ vs aggressive: Rotation thesis has a hole: Dow, S&P, and Nasdaq all sold off on September 15, so freed-up capital has nowhere obvious to outperform MMM

→ vs conservative: Pushing trim to 50% reduces risk-adjusted upside without meaningfully reducing risk; 30-40% better balances protection and optionality

→ vs conservative: 'Soft reassessment level' at $157.50 is decision fatigue dressed in responsible language; hard stops remove discretion at the exact moment when discretion is most dangerous

→ vs conservative: The 2-3% residual cap is precisely the answer to gap risk; softening exits to chase perfect exit means never having one