The Crypto Fear & Greed Index: What It Measures and What It Does Not

Warren Buffett's best-known principle — "be fearful when others are greedy, and greedy when others are fearful" — describes a pattern that market observers have long discussed. The Crypto Fear & Greed Index is an attempt to put a number on one half of that sentence: the collective emotional state of the market.

This guide explains how the number is built, what its extreme readings have historically coincided with, and — just as importantly — what it cannot tell you.

The DAI Fear & Greed Index is a composite sentiment score from 0 to 100, recalculated every five minutes:

• 0–20: Extreme Fear — selling is broad and indiscriminate
• 21–40: Fear — cautious, risk-off sentiment dominates
• 41–60: Neutral — balanced sentiment, no strong directional bias
• 61–80: Greed — optimistic, risk-on sentiment
• 81–100: Extreme Greed — euphoric, late-cycle behaviour

The score is built from five inputs, each carrying a fixed weight:

One thing follows from this table, and it is the most important thing to know about this particular index: every one of the five inputs is market data. There is no social-media sentiment component, no survey, and no search-volume proxy. Several well-known fear and greed indices include those inputs; this one does not.

That is a deliberate trade-off, and it cuts both ways. The score cannot be moved by a coordinated posting campaign — but it also cannot see a mood that has not yet reached price, volume or dominance. It measures what the market did, not what participants say.

The index is most discussed at its extremes, and the reasoning is always some version of the same argument:

At extreme fear, selling has been broad, leveraged positions have been closed out, and the participants most likely to sell have already sold. At extreme greed, buying has been broad, leverage is elevated, and the participants most likely to buy have already bought.

The contrarian reading is that a crowd at an emotional extreme has, by definition, already acted — and therefore has less left to do.

This is an argument, not a law. It describes a tendency that observers have noted across many markets and many decades. It does not predict any particular move, and it has failed often enough that its failures have their own name: value traps on one side, melt-ups on the other.

You will find articles that attach precise return figures and success rates to "buy at extreme fear" rules. We do not publish those numbers, for two reasons.

The first is methodological. Any such figure depends entirely on choices the author rarely shows: which start date, which holding period, which definition of "extreme", whether the periods overlap, and what happens to the sample when a single cycle is excluded. Change any one of them and the number changes materially.

The second is regulatory. A published success rate attached to a decision rule reads as a performance claim about a strategy. Data Analytic Investments is an information and educational provider, not an investment firm, and does not publish performance claims.

What can be said without inventing precision: extreme readings have historically been uncommon and short-lived, and they have tended to cluster around turning points that were only identifiable afterwards. That last clause is the whole difficulty. A reading of 12 looks like an opportunity in hindsight and like the beginning of something worse at the time.

Sentiment is rarely read on its own. Market participants who use it generally read it against the prevailing market regime, and the combinations are usually described like this:

• Bull trend with fear — commonly interpreted by observers as a temporary sentiment dislocation rather than a structural change
• Bull trend with greed — trend intact, but the sentiment input adds little information
• Bear trend with fear — the difficult case: fear can persist for months without resolving
• Bear trend with greed — commonly read as a counter-trend rally within a broader decline
• Ranging market with either extreme — the case where mean-reversion arguments are most often made

Note that every line above describes how the combination is read, not what follows from it. The regime does not make the sentiment reading correct; it changes how much weight readers place on it.

One frequently discussed pattern is divergence between price and sentiment:

Bullish divergence — price makes a new low while the index makes a higher low. Sentiment is less fearful than the price alone would suggest.

Bearish divergence — price makes a new high while the index makes a lower high. Sentiment is less greedy than the price alone would suggest.

Divergence is a description of two series moving differently. It is not a prediction, and it resolves in both directions.

This is the part of the index most worth understanding, because these errors are common and each one is avoidable.

The index measures the emotional state of the market. It does not measure where price is going. These are different questions, and the index only answers the first.

"Fear is an opportunity" is an argument that assumes the larger structure is intact. In a sustained decline, extreme fear can persist for months while price continues lower. The sentiment reading looks identical in both cases.

Extreme readings persist. An index below 20 can stay below 20 for weeks. Anyone reading the index as a single point in time is reading it wrong; extremes are ranges, and they end when they end.

The index is recalculated every five minutes, but the behaviour it measures — broad selling, elevated leverage, exhausted buyers — plays out over days and weeks. A five-minute refresh does not make it a five-minute indicator. Reading it as a short-term trigger mistakes the update frequency for the timescale of what it describes.

The DAI Fear & Greed page shows the current score, the band it falls into, and the full methodology — each of the five factors, its weight, and what it measures. The score is recalculated every five minutes from live market data.

The page presents the score and how it is built. It does not tell you what to do with either.

View the Fear & Greed page →

Sentiment is one input among many, and a noisy one. The Fear & Greed Index is useful because it is transparent — you can see all five factors and their weights — and because it puts a number on something otherwise discussed only in adjectives.

What it cannot do is tell you what happens next. No sentiment indicator can. The market has been described as a voting machine in the short run and a weighing machine in the long run. This index counts votes. Weighing the evidence is a separate task, and it is yours.

The DAI Fear & Greed page shows the live index alongside market regime context, with sourced data. Educational market observation — not investment advice. View the page →