Sentiment Extremity, Volatility, and Liquidity Withdrawal
Extreme fear and extreme greed may share a microstructure feature: elevated uncertainty, greater market-making risk, and reduced liquidity. Direction and intensity require separate tests, with controls for volatility embedded in sentiment indices.
Direction and extremity are distinct hypotheses
Treating low readings as buys and high readings as sells is a directional hypothesis. Testing distance from 50 is an intensity hypothesis: it does not predict direction but asks whether sentiment tails coincide with wider spreads, greater slippage, or higher subsequent volatility.
If liquidity worsens at both fear and greed extremes, a simple linear regression can cancel the effects. Analysis should include absolute deviation, percentile bins, or nonlinear terms and report both tails separately.
The endogeneity of volatility
Popular Fear & Greed indices already include price volatility, momentum, or volume. When extremes coincide with volatility, one cannot claim that sentiment caused liquidity withdrawal; both variables may describe the same state.
A stricter test compares extreme and neutral sentiment within realized-volatility bins or controls for volatility and volume. If the effect disappears, the conclusion should be narrowed to a regime indicator rather than an independent causal factor.
From research result to risk management
Even if sentiment extremes reliably predict poorer execution, the practical response may be fewer market orders, lower leverage, and stricter slippage checks—not an automatic contrarian position. A signal can improve execution without forecasting direction.
Future work can combine extremity with funding percentiles, OI change, and DVOL. Each added condition reduces the sample and increases overfitting risk, so sufficient observations and an out-of-sample segment remain essential.