CNN Fear & Greed Sentiment Analysis 2026-09-14

Data Cutoff and Update Status

Beijing time 2026-09-14 is Monday. Market data covers through the last trading day 9/11; 9/12 (Friday) was a market holiday.

New Changes and Continuations

Core change: The Fear & Greed composite index fell rapidly from 39.1 (Fear) on 9/8 to 33.3 (Fear) on 9/11, and collapsed from 53.7 (Neutral) over 10 days to 33.3, a cumulative drop of -20.4 points. This is a structural cooling of sentiment, not a single-day fluctuation.

Continuing signals:

• Consecutive fear days已达 4 days (9/8–9/11), following earlier episodes of Extreme Fear
• Stock Price Breadth remains in Extreme Fear territory (12.6 points), the weakest sub-indicator
• Credit spreads remain in normal range, forming a clear divergence from sentiment
• Margin debt and ETF inflows maintain red-alert levels at historical highs

Composite Index Trend

![FNG Trend](/charts/fng-trend-2026-09-14.png)

Last 10 trading days:

Key statistics:

• 10-day trend: 53.74 → 33.34, change -20.40 (rapid cooling)
• Sample minimum: 5.17 (2025-11-20), current level still ~28 points from extreme
• Consecutive fear days: 4 days (9/8–9/11)
• Full-sample rating distribution: Fear 93 days / Greed 68 days / Neutral 48 days / Extreme Fear 41 days

Prior close verification: Composite timestamp 2026-09-11T23:59:53+00:00, previous_close = 33.11. Consistent with historical series value of 32.20 on 9/10 (9/11 is the latest trading day). The +0.23 change from 33.11 to 33.34 is an intraday micro-move, not a trend reversal.

Seven Sub-Indicators

![FNG Radar](/charts/fng-radar-2026-09-14.png)

Note: The JSON output only returns safe_haven_demand and stock_price_breadth as independent top-level objects. The remaining 5 sub-indicators’ raw scores and values are embedded within the composite calculation pipeline and not separately exported. Complete per-indicator verification is not possible here.

Verified sub-indicator interpretation:

• Safe Haven Demand (30.6, Fear): Raw value 0.61, in Fear range. Safe-haven asset demand is elevated, consistent with 4 consecutive fear days. Funds are flowing into Treasuries/golds.
• Stock Price Breadth (12.6, Extreme Fear): Raw value 807.56 (down/up stock ratio), Extreme Fear level. This is the single biggest drag on the composite — market breadth is deteriorating, with selling pressure significantly outweighing buying.

Divergence Analysis and Counter-Evidence

Primary divergence: Fear sentiment vs. normal credit

The CNN Fear & Greed composite is in Fear territory (33.3), yet HY OAS is only 2.70% (normal range, <3%). This means:

1. Market sentiment panic has not transmitted to credit markets. Bond investors are not demanding additional risk premium, indicating institutional default-risk pricing remains stable.
2. Breadth deterioration (807.56 put/call ratio) coexists with narrow credit spreads. A possible explanation: speculative Put buying on options (hedging or directional speculation) inflates PCR while actual default probabilities remain low.

Counter-confirmation condition: If HY OAS begins widening rapidly from 2.70% past 3%, sentiment panic receives credit-market confirmation and risk level must be upgraded.

Secondary divergence: Cooling sentiment vs. leveraged/record fund flows

Margin debt at $1.304T sits at historical highs; 2026 YTD ETF net inflows of $8,560B are record-setting. Despite rapidly cooling sentiment, leverage and fund flows remain at extreme levels — a classic “sentiment cools but capital hasn’t exited” signal.

• Historical reference: High margin debt alone is not a crisis signal (see 2021–2022 sideways-with-volatility period), but combined with sentiment cooling, it reduces the market’s cushion. If volatility spikes, leveraged positions can amplify selling pressure.
• Not an automatic crisis call: Record-level margin debt by itself is not a crisis signal, but demands heightened vigilance when paired with sentiment deterioration.

Credit / Rate / Leverage Cross-Verification

![Sub-Indicator Trends](/charts/fng-sub-trends-2026-09-14.png)

Credit/rate interpretation:

• Credit spreads (HY 2.70% / IG 0.80%) are both in normal range — the strongest mitigating signal in this analysis. Bond markets are not pricing in risk alongside equity sentiment deterioration.
• The yield curve: 10Y-2Y is positively +0.33%, ending deep inversion, but 10Y-30Y still inverts at -0.42%. Curve shape is complex. Short-term rates already reflect rate-cut expectations (2Y 4.56% < 10Y 4.95%), but ultra-long end premium remains high.
• Margin debt and IPO/ETF data are structural risk factors but are lagging or cumulative, unable to reflect recent marginal changes.

Conditions for Risk Escalation or Mitigation

Conditions requiring risk upgrade (any single trigger warrants level increase):

1. HY OAS breaks above 3% from current 2.70% → credit markets begin pricing panic
2. Consecutive fear days extend beyond 7, composite drops below 25 → panic sentiment entrenches
3. Stock Price Breadth rises further from 807 → breadth deterioration accelerates
4. Latest FINRA monthly margin debt shows significant QoQ growth → leverage continues accumulating

Conditions confirming risk mitigation (multiple must align simultaneously):

1. Composite closes above 45 for 3 consecutive trading days (back to Neutral)
2. Stock Price Breadth drops below 400 (up/down ratio returns to balance)
3. HY OAS compresses further below 2.0% (credit markets confirm optimism)
4. Consecutive fear days fall to 2 or fewer

Data Limitations

1. Incomplete sub-indicators: This script output only returns safe_haven_demand and stock_price_breadth as independent objects. Raw scores and values for the remaining 5 sub-indicators cannot be directly extracted from the JSON; per-indicator verification is incomplete.
2. Margin debt 5-month lag: FINRA monthly data only updates through 2026-04, unable to reflect September leverage levels.
3. IPO/ETF are YTD cumulative: Does not reflect marginal changes in the past week or month.
4. FRED data as of 9/10: Credit spreads and yield curve data are from the previous trading day.
5. CNN FNG inherent limitations: Composite is a weighted synthesis of 7 sub-indicators with opaque weighting model; score interpretation relies on historical percentiles rather than absolute thresholds.