Market Sentiment. What the crowd feels, and why it is late.
Sentiment data tries to quantify how the market feels: fear and greed indices, social volume, survey-style measures and positioning-derived readings.
Its value is contrarian more often than directional. Extreme readings tend to appear when a move is already well advanced and the last participants are arriving.
Sentiment is a derivative of price, not a leading indicator
Most sentiment measures are built from things that follow price: social posts, search volume, survey responses, positioning after the fact. People are greedy because the market went up. That makes sentiment a reasonable description of the current state and a poor predictor of the next one, which is why extremes are more useful as a caution flag than as an entry.
Market Sentiment in a strategy
Readings at the top or bottom of their historical range argue for smaller size, not a reversal trade.
Sentiment agrees or disagrees with funding and open interest, and agreement between them is the stronger read.
A strategy can behave differently in sustained fear than in sustained greed.
Sudden jumps in social volume mark where retail attention arrived, which is often near the end.
Let sentiment cut your size, not pick your direction
The common mistake is shorting extreme greed. The defensible use is letting an extreme reading reduce risk on trades you were taking anyway.
- IFThe Fear & Greed reading is above 80, in extreme greed
- ANDFunding is positive and above its own 30-day median, so positioning agrees
- ANDYour strategy has independently produced a long entry signal
Sentiment cannot time anything, so the rule never generates a trade by itself; it only modifies one your strategy already wanted. Requiring funding to agree matters because a single sentiment index is easy to dismiss, whereas sentiment and positioning pointing the same way is a stronger statement that the trade is crowded. Halving rather than skipping reflects that crowded markets frequently keep going, and that being flat in a strong trend is its own kind of loss.
Composite scores built from volatility, momentum, volume and social data. Widely quoted, and methodology differs by publisher.
Platforms such as LunarCrush and Santiment track mention counts and engagement across social feeds.
Long/short account ratios published by exchanges, which are harder to manipulate than social metrics because they reflect real positions.
The weakest source on this list
Sentiment metrics are unstandardised, provider-specific, and easy to manipulate at the social end. Two indices can disagree completely on the same day. Use it as a supporting input that adjusts sizing, and be sceptical of anything that treats it as a primary entry signal.
Build a bot around Market Sentiment tonight.
No screens to babysit: define the conditions and the bot does the rest.
- Extremes as a caution
- Confirming crowding
- Regime context