When: MACD histogram turning positive

Then: Second of three required confirmations for re-entry to Hold

When: RSI sustained above 40 on weekly basis

Then: Third of three required confirmations for re-entry to Hold

HOLD existing position without adding capital; maintain stop at $86.50; trim if price reaches $92-$93

• SO's $77 billion debt compares to a mortgage analogy that breaks down due to refunding risk, active rate base flux, and 93% odds of zero Fed cuts in 2026
• Interest expense ran $778-$895 million per quarter as of Q2 2026 and could step higher if long-end rates back up 100 basis points on refinancing
• Construction in Progress jumped from $7.69 billion to $11.87 billion in one year, meaning the $77 billion debt is not static but growing in lockstep with capex
• Issuing equity at 18x earnings to fund assets earning 10-11% regulated returns is dilutive in year one and dependent on multiple expansion
• The 18.6% forward EPS growth is analyst consensus, not a mechanical output, and is vulnerable to revision if rate cases disappoint or construction costs overrun
• PEG ratio of 2.08 exceeds the sector's historical ~1.5, indicating the market is paying a premium for growth not yet delivered
• MACD flatlined at -1.21 with histogram at zero is not a bullish momentum signal but momentum that has stalled
• RSI at 37.77 is approaching but has not triggered oversold territory; a lower low below $87.88 would be a breakdown signal
• Death cross (50 SMA crossing below 200 SMA) is a widely-watched institutional signal that triggers systematic selling from trend-following strategies, pension funds, and quant funds
• OpenAI deal concentration creates 3.2 gigawatts of new load in a single counterparty, a new risk vector absent three years ago; Texas freeze on data center hookups is a cautionary precedent
• Stock traded above $95 in late July so market has already priced substantial value for the OpenAI deal; pullback since reflects reassessment of other factors
• Dividend is funded by debt and equity issuance with negative free cash flow, making SO a leveraged equity income vehicle rather than a true bond substitute
• 23-year dividend growth streak is at risk if capex cycle runs into trouble, rate cases come in soft, or interest expense crowds out coverage
• The $92-$93 trim level consensus between conservative and aggressive analysts suggests even the bullish case acknowledges this as a reasonable exit point
• Stop level of $86.50 sits approximately 1.5 ATR below current price and just above the lower Bollinger band at $86.61

→ vs aggressive: Debt analogy comparing SO's $77 billion to a mortgage ignores active rate base flux, annual refunding risk, state commission scrutiny, and 93% odds of zero Fed cuts; debt is not static but growing from $4 billion in new capex financing

→ vs aggressive: Calling 18.6% forward EPS growth 'mechanical' misrepresents it as analyst consensus number vulnerable to revision if rate cases disappoint, regulators push back on allowed ROEs, construction costs overrun, or interest expense increases

→ vs aggressive: Technical setup showing declining volume on consolidation reflects no buyers stepping up at current levels, not accumulation; MACD at -1.21 signals stalled momentum, not bullish divergence

→ vs aggressive: OpenAI deal is already priced into the $95 late-July peak; single-counterparty concentration of 3.2 gigawatts creates a new risk vector with precedent from Texas data center hookup freeze

→ vs aggressive: Bond substitute framing is false because SO has had a 7.3% drawdown in six weeks, free cash flow is negative meaning the dividend is funded by debt/equity issuance, not a bond-like characteristic

→ vs neutral: Death cross should not be dismissed as a lagging indicator; it is a widely-watched institutional signal that triggers systematic forced selling regardless of fundamentals

HOLD; do not add at $88.34; wait to reclaim $89.92 with MACD and RSI confirmation; trim 25-33% at $92-93 (50 SMA at $92.54); stop at $86.50; conditional exit if death cross confirms AND price breaks $87.88 on rising volume

• HOLD is correct; both analysts agree and this is a neutral position, not conviction or timidity
• Debt is not fragile but calling it a 'predictable revenue machine' like a mortgage is wrong—SO debt grows 5-6 billion annually for tripling CIP
• Dilution is transitional, not permanent—depends on whether rate base monetizes at pace market is pricing (18.6% forward EPS growth)
• Death cross is a coin flip; price action (confirmation above/below $87.88 with volume) tells direction rather than narrative
• OpenAI deal is genuine structural positive but upside is measured—stock already pulled back 7% from peak on AI enthusiasm
• OpenAI paying full infrastructure costs means zero demand risk and zero capex risk for SO; concentration risk is to OpenAI's success, not SO's balance sheet
• Dividend is stable (23 consecutive years) but SO behaves like equity, not bond—7.3% drawdown in six weeks disproves bond substitute framing
• Aggressive trim recommended if $88-to-$92 move happens quickly on heavy volume (blow-off top pattern in utilities)
• The streak of dividend payments through every rate cycle and recession is structural, not fragile, but dividend does not protect principal

→ vs aggressive: Debt is not a bond or mortgage—mortgage debt amortizes down while SO debt grows 5-6 billion annually for tripling CIP; 'predictable revenue machine' is marketing rhetoric

→ vs aggressive: Death cross pattern-matching from survivorship bias is dangerous—every utility recovery has had different drivers; 'this time it's the same' gets you caught in 12-month drift

→ vs aggressive: OpenAI deal is now embedded in expectations; stock rallied from mid-$80s to over $95 then pulled back 7%—market already priced the deal and started discounting

→ vs conservative: Dilution is transitional, not permanent—if rate base growth translates to 18.6% forward EPS growth, dilution gets absorbed over 2-3 years

→ vs conservative: Concentration risk framing is wrong—OpenAI paying full infrastructure costs means zero demand risk and zero capex risk for SO; 'concentration' risk is to OpenAI, not SO's balance sheet

→ vs conservative: Negative free cash flow is how every regulated utility in build-out phase operates; dividend is paid from operating cash flow and has continued through every rate cycle for 23 years