CNN Fear & Greed Sentiment Analysis 2026-09-29

CNN Fear & Greed Sentiment Analysis — 2026-09-29 (Run Date)

Market Data As-of: 2026-09-28 (Friday close) Run Time: 2026-09-29 06:30 CST

1. New Changes and延续 of Prior Signals

Continuing signals:

• 15 consecutive trading days of Fear rating since 9/8 (15 of 20 trading days in Sep were Fear)
• Brief rebound 9/21–9/25 to 34–37 range; 9/28 fell back to 33.94
• Sub-indicator pattern unchanged: stock strength/breadth persistently extreme weakness, VIX hovering neutral, junk bond demand maintaining Greed

No new data sources. HY OAS, margin debt, IPO/ETF flows carry forward from the prior report with the same cut-off dates (see Section 5).

2. Composite Index Trend

![FNG Trend](/charts/fng-trend-2026-09-29.png)

Complete last 15 trading days sequence:

Key observations:

• Sample minimum: 5.17 (2025-11-20, Extreme Fear). Current 33.94 is 29 points above that extreme.
• 9/15–9/17 hit本轮 fear bottom (27–28 range), rebounded to 36.94 on 9/25.
• 9/28 dropped 3.1 points, reconfirming Fear zone has not fundamentally broken.
• Past 10 trading days: 28.0 → 33.9, overall rise +6.0, but intraday volatility shows fragile rebound.
• Consecutive Fear days: 15 days (since 9/8), the longest consecutive Fear stretch recently.

3. Sub-Indicator Details

![FNG Radar](/charts/fng-radar-2026-09-29.png)

This data source provides 9 sub-indicators (not the standard 7), standardized as follows:

Standardization notes:

• Each sub-indicator Score mapped to 0–100; 0 = Extreme Fear, 100 = Extreme Greed
• VIX is bidirectional: too high = Fear, too low = Greed; 50 = midpoint
• Junk bond demand vs. stock strength/breadth shows extreme divergence (see Section 4)

4. Divergence Analysis and Counter-Evidence

Divergence 1: Stock Strength/Breadth = 0 vs. Junk Bond Demand = 64

Explanation: Equity internals are extremely weak (broad sell-off, breadth collapse), but the credit market is not following into panic. This typically means:

• Institutional investors still seeking yield in credit bonds; no bond-market sell-off
• Stock decline concentrated in specific sectors (tech/growth), not a market-wide liquidation
• HY bond issuers’ fundamentals have not deteriorated to trigger a default wave

Counter-evidence needed:

• Check S&P 500 sector distribution — if decline concentrated in one sector (e.g., semiconductors/software), not systemic
• Track HYG/LQD ETF volume and price — if prices still making highs, the Greed signal is credible
• Monitor default rates (BofA HY Default Index) — if stable, credit bond Greed is rational

Divergence 2: VIX = Neutral (50) vs. Stock Breadth = Extreme Fear (0)

Explanation: VIX at neutral suggests implied volatility has not spiked sharply. Possible reasons:

• Market has “accustomed” to recent weakness; no panic-driven volatility spike
• VIX is less responsive to “slow grind down” and more to突发事件 events
• Options market pricing shows downside protection costs not yet extreme

Counter-evidence needed:

• Check IV Rank — if also low, volatility genuinely not panicked
• Observe 25-delta Put/Call Ratio absolute value — if >0.15, hedging demand is masked by VIX

Divergence 3: Composite 33.94 (Fear) vs. Junk Bond 64 (Greed)

The composite is the average of all sub-indicators. Stock strength/breadth at 0 drags the average down significantly, while junk bond 64 provides upward support. If looking at equity维度 alone, sentiment is worse than the composite suggests.

5. Credit / Rate / Leverage Cross-Validation

![FNG Sub-Trends](/charts/fng-sub-trends-2026-09-29.png)

Cross-validation conclusion:

1. Credit stable. HY OAS at 2.93% below 3% alert threshold, IG OAS at 0.81% normal. Curve 10Y-2Y flipped positive to +0.32%, escaping deep inversion. Consistent with junk bond demand 64 (Greed) — credit market not following equity panic.

2. Leverage dangerous. Margin debt at $1.304T (April 2026 data, 5 months stale) at historical high range. If current levels hold while stock breadth is extremely weak, leveraged positions may be “forced hold” rather than active bullish — this is a lagging but potentially amplifying risk.

3. Record fund inflows. YTD ETF net inflows of $8,560B is a historic record. Massive capital entering without lifting stock breadth suggests:

   • Capital may be concentrated in few large-cap stocks/ETFs (e.g., Magnificent 7), not broadening
   • Or funds entered then got partially trapped, creating “money in, price not rising” stalemate

4. Overall assessment: Normal credit + high leverage + record but concentrated flows = fragile market structure. Not an immediate crisis signal (credit not worsening), but a catalyst for drawdown could see high margin debt amplify declines.

6. Conditions for Risk Escalation or De-escalation

Risk escalation triggers (any one warrants alert):

• HY OAS breaks above 3% → credit spread widening, rising corporate funding costs
• Margin debt monthly data confirms >$1.3T while stocks continue declining → leveraged forced liquidation
• VIX rises from 50 to >60 (Fear zone) → volatility spike confirmed
• Stock breadth stays <5 for 5 consecutive days (expanding Extreme Fear)
• 10Y-2Y re-inverts deeply (<-50bp)

Risk de-escalation signals (not yet confirmed as of this report):

• F&G closes above 45 for 3 consecutive days and holds Neutral zone (40–60)
• Stock strength recovers from 0 to >20
• Stock breadth recovers from 0 to >25
• HY OAS compresses further below 2.5% (enhanced credit risk appetite)

7. Data Limitations

1. Margin debt latest is April 2026 (FINRA monthly release), 5 months stale. Current “historical high zone” status is inferred, not real-time.
2. IPO/ETF flows are YTD cumulative figures, not reflecting this week’s changes.
3. HY OAS as of 09-25, 4 days from run date; material change possible but unlikely in short window.
4. CNN F&G sub-indicators — this data source provides 9 items rather than the standard 7 (no independently named width_of_leader, stock_price_position); internal script mapping may differ slightly from CNN’s official naming.
5. Sample minimum of 5.17 occurred on 2025-11-20, ~11 months ago. Longer historical extremes (e.g., March 2020) are outside this sample and cannot be directly compared.
6. VIX raw value not directly obtained; Score 50 (Neutral) corresponds to VIX approximately 20, an estimate.