Tools

Fear & Greed, with the noise removed

The daily print of the Fear & Greed Index is a coin flip away from changing its label. The seven-day average is not, and it is the version that feeds our index. This page shows both, the whole history against the bitcoin price, and what actually happened around the extremes — including the parts that do not flatter the contrarian story.

Seven-day average

Cool

28.4

The mean of the last seven daily prints through 26 Jul 2026. This is the value that enters the Hodlometer Index.

Percentile (4y)
25th
Median (4y)
49.8
Weight in index
25%

Today's print

Fear

26.0

The raw reading published by alternative.me for 26 Jul 2026, on a 0–100 scale where zero is maximum fear.

History from
Feb 2018
Days collected
3,094

Fear & Greed through 26 Jul 2026; bitcoin closes through 26 Jul 2026. This is a bitcoin-only index — alternative.me publishes one reading, and we apply it to both majors.

Sentiment against price

Seven-day average and the bitcoin price, 2018–2026

Two panels on one time axis: sentiment above, price below in log scale. They are drawn apart on purpose — a shared vertical field would put a linear 0–100 scale next to a logarithmic price one, and where their zero lines meet is arbitrary. Any correlation you read off such a chart is drawn by the layout, not present in the data.

02040608010085 — extreme greed20 — extreme fear $3.0k$10k$30k$100k20192020202120222023202420252026

Fear & Greed, 7-day average (top)BTC daily close, log scale (bottom)

The extremes, counted

Below 20 and above 85

Measured on the seven-day average since 1 Feb 2018.

MeasureBelow 20Above 85
Days 338 10.9% 90 2.9%
Distinct episodes 12 3
Most recent 6 Jul 2026 25 Nov 2024
BTC a year later, full range −58.3% … +1,040.1% −35.5% … +313.0%
… median inside that range +43.9% +94.4%
… higher a year later 134 of 219 days 67 of 90 days

Forward figures compare the bitcoin close on each extreme day with the close 365 days later, so they stop 26 Jul 2025. The windows overlap almost entirely — read them as a description of a few episodes, not as odds.

Recent episodes

Runs of the seven-day average inside an extreme, joined across gaps of up to 30 days.

PeriodExtremeDaysMost extreme
5 Jun 2026 – 6 Jul 2026 Fear 29 10.7
3 Feb 2026 – 17 Apr 2026 Fear 71 7.7
16 Nov 2025 – 23 Dec 2025 Fear 19 12.1
19 Nov 2024 – 25 Nov 2024 Greed 6 87.9
10 May 2022 – 18 Jul 2022 Fear 68 7.4
10 Jan 2022 – 29 Jan 2022 Fear 11 15.9
22 May 2021 – 24 Jul 2021 Fear 25 15.7
11 Nov 2020 – 25 Feb 2021 Greed 83 94.0

Eight most recent of 15 episodes since 1 Feb 2018.

What this means for a holder

The contrarian frame — buy when others are fearful, at the point of capitulation — is the reason most people look at this index, so it is worth checking against the record rather than repeating. In our data the fearful side holds up better than the greedy side, and neither holds up well enough to act on alone.

The seven-day average has spent 10.9% of its life below 20 and only 2.9% above 85. That asymmetry is not a market fact — it is a property of an index that spends most of its time in the lower half. And the greedy sample barely exists: 90 days in total, of which the great majority fall inside one episode in the winter of 2020–21. A median of +94.4% a year after those days does not mean greed is bullish; it means we are measuring one bull market with a very small ruler.

What the index is genuinely useful for is noticing that your own mood has company. Extreme readings cluster — they arrive in multi-week episodes, not as isolated days — so if the number has been in an extreme for a month, that is a fact about the environment you are making decisions in. Whether it should change anything depends entirely on your plan, and a plan that changes when sentiment changes was not a plan.

Two limitations to keep in view. First, one quarter of the Hodlometer Index rests on a method we cannot reproduce: alternative.me publishes the component weights but not the inputs or the calculation, and its survey component is currently paused. Second, this is a bitcoin index. We apply it to ether as well, and when the two diverge — as they have through most of 2026 — this component cannot see the difference.

Fear & Greed vs the Hodlometer Index: not the same thing

Both are one number from 0 to 100, both sit on the same kind of dial, and one of them is a quarter of the other. That is enough overlap to muddle anybody, so here is the difference laid out.

Fear & GreedHodlometer
What it measures The mood of the crowd today: how frightened or how excited people are right now. How hot the market is next to its own last four years — a reading for somebody who holds across cycles rather than trades the week.
What it is built from Volatility, market momentum and volume, social media, surveys (currently paused), bitcoin dominance and Google Trends. The Mayer Multiple (30%), the distance to the 200-week average (30%), Fear & Greed itself (25%) and perpetual funding rates (15%).
Who makes it alternative.me. We read their number and cannot rebuild it: the weights are published, the inputs and the arithmetic are not. We do, from data anyone can pull. Every step is written out on the methodology page.
Where the 0–100 is cut At alternative.me’s own lines — 25, 45, 55, 75 — with their labels, from Extreme Fear to Extreme Greed. Into five equal fifths — 20, 40, 60, 80 — labelled Deep Freeze through Overheated.
Which number is shown first The seven-day average, with today’s raw print beside it. One loud day can flip the daily label on its own. One value per closed daily candle, UTC. It is slow by construction: every input is ranked against four years before it is averaged in.

They meet at one quarter. Fear & Greed is one of the four inputs to the Hodlometer Index and carries 25% of the weight — one leg of a four-legged stool. It does not go in raw: we average a week of prints and then ask where that average sits in the last four years, so what actually enters the index is a rank, not the print alternative.me published this morning. The other three legs are price and positioning, and between them they carry the remaining 75%.

Neither one is a signal. A frightened crowd has come before recoveries and before much deeper falls; a hot index reading has come before a top and before another year of climbing. One describes today’s mood, the other describes the whole cycle, and both stop there.

The dials on this site share one shape on purpose: the same arc, the same five colours running cold to hot. Where those colours are cut is a property of the number being shown, not of the dial: a reading of 30 is “Fear” on alternative.me’s scale, while 30 on ours falls in the second fifth, “Cool”. Same instrument, different scale printed on it.

Longer versions of both halves: how the four components are combined into one number, and today’s index reading on the front page.

Read the full guide

Fear & Greed for Long-Term Holders

What the index actually measures, why the seven-day average is the honest version, and how contrarian readings hold up against our data.

Frequently asked questions

What is the Crypto Fear & Greed Index?

A 0–100 sentiment gauge published daily by alternative.me. It blends volatility (25%), market momentum and volume (25%), social media activity (15%), surveys (15%, currently paused), bitcoin dominance (10%) and Google Trends (10%) into one print. Zero is maximum fear, one hundred is maximum greed.

What is the Fear & Greed reading today?

The print for 26 Jul 2026 is 26 (Fear). The seven-day average, which is the number we actually use, is 28.4 — the 25th percentile of the last four years.

Why a seven-day average instead of today's number?

Because the daily print is noisy enough to flip classifications on a single green candle, and a holder deciding anything on that basis is trading, not holding. A week of readings keeps the signal and drops most of the flicker. It is also what feeds the Hodlometer Index, so the number here and the component score on the methodology page are the same number.

Is the Hodlometer Index the same as the Crypto Fear & Greed Index?

No. Fear & Greed is published by alternative.me and measures how the crowd feels today; the Hodlometer Index is ours and measures how hot the whole market is against its own last four years. Fear & Greed is one of four inputs to it, worth 25% of the weight, and it enters as the seven-day average ranked against those four years rather than as the raw daily print. The other three quarters are price and positioning, which is why the two numbers can sit far apart on the same day — they are compared line by line above.

Does the index cover Ethereum?

No. alternative.me publishes one reading, and it is a bitcoin index — there is no separate ether print. We apply it to both majors, which is a simplification we would rather state than hide: when ether diverges from bitcoin, this component does not see it. The two price-based components on this site are calculated per asset precisely to compensate.

Is extreme fear a buy signal?

Our own data does not support treating it as one. Since 1 Feb 2018 the seven-day average has been below 20 on 338 days across 12 episodes. A year after those days bitcoin was anywhere from −58.3% to +1,040.1% — that spread is the finding. Inside it the median was +43.9% and 134 of the 219 measurable days ended higher. The windows overlap almost completely, so this describes a handful of episodes, not a probability, and it is not a rule you can apply to the next one.

How often does this page update?

Daily. alternative.me publishes a new value each day at 00:00 UTC and we pick it up in the same run that recomputes the index. The Fear & Greed print for today often arrives before our own daily candle closes; we hold it back rather than show it, so every page on this site reports the same reading for the same last closed UTC day.

How this page is calculated

  • Source. The daily print comes from alternative.me, which publishes one value per day for bitcoin. We store every reading from 1 Feb 2018 onwards and never rewrite history.
  • The seven-day average is a calendar mean over the trailing seven days, requiring at least four prints in the window — the same function that normalises this component for the index, so the two cannot drift apart.
  • Extremes are defined on the average, not on the daily print: below 20 and above 85. Consecutive extreme days are joined into one episode when the gap between them is 30 days or less, otherwise one long bear market would count as dozens of separate signals.
  • Percentiles use the mid-rank convention inside a trailing 1,460-day window, exactly as on the methodology page.
  • Dates can differ by a day. Sentiment for today is published before today's daily candle closes, so this page may run one day ahead of the index reading on the home page.

Data through 26 Jul 2026 · JSON API · All tools