What the crowd feels, read against them
A quick read on whether the crowd is fearful, balanced, or euphoric right now — and why the uncomfortable side is usually the historically rewarding one.
What sentiment analysis is trying to catch
Sentiment indicators exist because markets are made of people, and people become most confident near the point of maximum risk. The classic contrarian reading — fear near bottoms, euphoria near tops — is well documented in Bitcoin's history, and sentiment measures are the attempt to make that legible in numbers rather than vibes.
Sentiment is also the noisiest indicator family. It reacts within hours, reverses just as fast, and the underlying data sources are more vulnerable to manipulation than a blockchain or a price feed. That combination is why sentiment works better as a composite than as any single reading, and why this lab lets you see what each source contributes.
The sources behind the composite
Survey and index measures, chiefly the Fear & Greed Index, blend volatility, momentum, volume and search interest into a single published number. Widely watched, and useful for exactly that reason — it is what a large share of the market is looking at.
Derivatives positioning reads sentiment from where money actually sits rather than what people say. Funding rates on perpetual futures show which side is paying to hold its position, which is a considerably harder signal to fake than a poll.
Attention measures track how much of the public is paying attention at all, using audience growth on the largest crypto channels. Retail attention is cyclical and it tends to arrive late, which makes it useful for locating the crowd within a cycle.
How to read a sentiment composite honestly
Sentiment is a condition, not a trigger. Extreme greed can persist for months in a strong uptrend, and extreme fear can persist through a long decline — anyone who sold the first greedy reading of a cycle would have missed most of it. The useful pattern is divergence: sentiment moving one way while price moves the other. Each factor here is normalized 0–10 against its own observed history, so a reading describes where that source sits relative to itself, not against an absolute standard. Several of the series are short, and their normalization will shift as history accumulates.
Common questions
Do I need an account to use the Sentiment Lab?
The live readings and the composite score are visible to everyone. A free account unlocks the research controls — adjusting weights, toggling indicators, and replaying the composite through history.
Is sentiment part of the published risk model?
The risk model has a momentum and sentiment category, but it uses a narrower, longer-history set of inputs than this lab exposes. The newer sentiment sources here are published as research rather than folded into the composite.
Can sentiment data be manipulated?
Some of it, yes — which is why the sources here lean toward measures backed by money or platform-reported audience data rather than post counts and engagement metrics, where automated accounts distort the picture badly.
Does extreme fear mean it is time to buy?
Not on its own. Extreme fear has coincided with major bottoms and it has also persisted for months while price fell further. It is one input among several, and the risk model treats it that way.
Related on Alphabit
- Fear & Greed Index — the best-known sentiment gauge
- Social sentiment tracker — YouTube attention data
- Market Lab — the same bench for price technicals
- On-Chain Lab — the same bench for blockchain factors
Data is provided for research and education. Nothing here is financial advice — see the disclosure.