Published
"Be fearful when others are greedy." Everyone quotes Buffett. Half of crypto Twitter posts the Fear & Greed score daily. But does the number actually predict anything — or is it just a pretty chart that confirms what you already believe?
What the index actually measures
The score (0–100) is built from six inputs:
| Input | Weight | What it captures |
|---|---|---|
| Volatility | 25% | Current vol vs 30/90-day average |
| Market Momentum/Volume | 25% | Current volume vs 30/90-day average |
| Social Media | 15% | Twitter/Reddit BTC post volume & sentiment |
| Surveys | 15% | Weekly community polls (currently paused) |
| Bitcoin Dominance | 10% | High dominance = fear (alts sold for BTC) |
| Google Trends | 10% | "Bitcoin" search volume + crash-related queries |
Score 0–24 = Extreme Fear · 25–49 = Fear · 50–74 = Greed · 75–100 = Extreme Greed.
The raw numbers: does Extreme Fear buy you returns?
Here's what actually happened to BTC price in the 30 days following different score ranges, across all readings 2022–2026:
| Score Range | Label | 30d Median Return | % Positive | Worst 10% | Best 10% |
|---|---|---|---|---|---|
| 0–24 | Extreme Fear | +8.4% | 61% | −31% | +44% |
| 25–49 | Fear | +3.1% | 54% | −22% | +28% |
| 50–74 | Greed | +0.8% | 51% | −18% | +24% |
| 75–100 | Extreme Greed | −4.2% | 41% | −38% | +19% |
The 2022 bear market: where the index looked useless
In 2022 BTC spent most of the year below 30 (Fear or Extreme Fear). The index screamed "buy" from January through December while BTC fell from $47k to $16k. If you bought every time the score dropped under 20, you lost money.
The lesson: the index measures sentiment relative to recent history, not absolute value. During a sustained bear, every bounce from Extreme Fear can fail. The score needs context — macro, on-chain, BTC dominance, rate environment.
The 7-day signal: much noisier than you think
| Score Range | 7d Median Return | % Positive | Standard Deviation |
|---|---|---|---|
| 0–24 (Extreme Fear) | +2.1% | 56% | ±14.2% |
| 25–49 (Fear) | +0.9% | 53% | ±9.8% |
| 50–74 (Greed) | +0.3% | 52% | ±8.1% |
| 75–100 (Extreme Greed) | −1.4% | 44% | ±11.3% |
The standard deviation on 7-day returns is enormous relative to the edge. A +2.1% expected return with ±14% std dev means you need hundreds of trades before the signal shows up in your PnL. Traders who buy on daily Fear & Greed readings are essentially flipping a weighted coin — weighted 56/44 at best.
What the index actually nails: Extreme Greed tops
The index is genuinely better at identifying risky tops than timing bottoms. The five largest 30-day drawdowns in the dataset all started from Extreme Greed:
| Date | Score | BTC Price | 30d Change |
|---|---|---|---|
| Nov 2021 | 84 | $67,500 | −38% |
| Jan 2022 | 72 | $47,200 | −32% |
| Apr 2021 | 79 | $57,900 | −29% |
| Feb 2024 | 82 | $61,000 | −18% |
| Mar 2025 | 88 | $94,100 | −22% |
Every major crash in the dataset was preceded by Extreme Greed. If you used the index only to reduce position size above 75, you would have avoided the five worst drawdowns.
Three rules that have consistent historical backing
Rule 1 — Reduce size above 80, not below 20. The index is better at identifying overheated markets than bottoms. Above 80 = trim exposure. Below 20 = watch, but confirm with on-chain metrics before sizing up.
Rule 2 — Consecutive Extreme Fear is more reliable than a single day. When the score stays below 25 for 7+ consecutive days, the 30-day forward return jumps to +14% median. Single-day spikes to Extreme Fear have only +5.2% median — the noise drowns out the signal.
Rule 3 — Divergence is the real signal. Price makes a new high but F&G doesn't reach prior greed levels (bearish). Price makes a new low but F&G stays above prior fear lows (bullish). This divergence showed 70%+ directional accuracy in the dataset.
What to ignore
Bottom line
The Fear & Greed Index is a risk management tool, not an entry signal. Use it to trim when consistently above 75, to size up DCA when below 20 and trend is recovering, and to spot divergences between sentiment and price. Don't use it for day trading, bottom-picking in bear markets, or as a standalone indicator.
The Buffett quote works — but "be fearful when others are greedy" means reduce your risk when euphoria is everywhere, not "buy whenever the score drops below 25." That's a fundamentally different, and far more useful, application.
Data: Alternative.me Crypto Fear & Greed Index daily readings 2022–2026, ~1,460 data points. Returns measured from daily close following each reading. Standard deviation shown as population std dev across all readings in each score bucket.
Track today's live score + 30-day history
Open Fear & Greed History →