The Crypto Fear and Greed Index Is 59% Price. That Is Why It Always Agrees With the Chart.

You have seen the dial. A number between 0 and 100, a coloured needle, a word underneath it: fear, neutral, greed. On 29 October 2026 it read 51, which is neutral. A month earlier it read 74, which was greed.

Most people treat that number as an independent read on how the market feels, separate from what the market is doing. Something you can check against the chart.

It is not separate from the chart. It is substantially made of the chart.

The index has six published components. One of them, the survey, is currently paused, and the survey was the only component that actually asked human beings how they felt. Renormalise the five that are still running and you get this: volatility and market momentum together make up 58.8% of the live index.

So a gauge marketed as a sentiment indicator is, by weight, mostly a restatement of recent price action. That does not make it useless. It makes it something different from what most people think they are reading, and the difference changes how you should use it.

Here is the full breakdown, the arithmetic, the disagreement between providers, and the one thing the index is good for.

TL;DR

• On 29 October, Alternative.me read 51 and CoinMarketCap read 46. Same metric, same day, five points apart.
• Six components, one switched off. Surveys carry a stated 15% weight and are paused.
• Renormalised, volatility is 29.4% and momentum/volume is 29.4%. Together 58.8% of the live index.
• Social media is 17.6% and uses Twitter only. Reddit is in the published methodology but not in the live index.
• Alternative.me states its index covers bitcoin only. The thing everyone calls the crypto fear and greed index is a bitcoin gauge.
• Bands: 0 to 24 extreme fear, 25 to 49 fear, 50 neutral, 51 to 74 greed, 75 to 100 extreme greed. A reading of 51 and a reading of 74 are both "greed" and they are not the same market.
• It is backward looking by construction. Every input is a measurement of something that already happened.

1. What the number actually is

The index compresses several market measurements into a single value from 0 to 100, where 0 is maximum fear and 100 is maximum greed. The bands are:

Worth noticing how wide those middle bands are. A reading of 51 and a reading of 74 both print the word "greed" under the dial, and they describe completely different conditions. If you are going to use the index at all, use the number and ignore the label, because the label throws away most of the information.

The all-time range on Alternative.me runs from about 5 to 95. CoinMarketCap's version has run from about 5 to 92. So the extremes do get reached, but rarely, and the index spends most of its life in the middle where the label is least informative.

2. The six components, and the one that is switched off

This is the published methodology, and the second column is where it gets interesting.

Three things in that table deserve more attention than they usually get.

The survey is paused. It carried 15% and it was the only component that worked by asking people a question rather than measuring a price series. The index that is supposed to tell you how the market feels has switched off the part that asked.

Social media is Twitter only. The published methodology describes Reddit sentiment analysis, but Reddit is not in the live index. So the 15% social slice is one platform.

Bitcoin dominance is a component. Bitcoin's share of total market capitalisation carries 10%. That is a structural market measure, and it moves for reasons that have nothing to do with sentiment, as we showed when dominance rose in October while bitcoin itself fell.

3. The arithmetic nobody runs

Here is the part that changes how you read the dial.

If one component carrying 15% is paused, the remaining five have to be rescaled to sum to 100. The live index therefore runs on 85% of the stated weight, redistributed:

Add the first two rows. 58.8%.

Volatility measures how much price has been moving relative to its own 30 and 90 day averages. Momentum and volume measure how much price has been moving in a direction, relative to the same averages. Both are functions of the price series. Neither asks anyone anything.

So nearly three fifths of a sentiment gauge is a derived price indicator. Which explains something people notice and find reassuring: the index almost always agrees with the chart. Of course it does. It is substantially computed from the chart.

The remaining 41.2% is more genuinely about attention and positioning: Twitter volume, Google search interest and bitcoin's structural share. That part is real and it is the part worth watching. It is also the minority.

None of this is hidden. Alternative.me publishes the weights and states that surveys are paused. The point is that almost nobody does the division, and without the division the dial looks like independent evidence when it is mostly a second view of the same data.

4. Two providers, two numbers, same day

If the index measured something objective, every provider computing it would land in roughly the same place. They do not.

On 29 October 2026:

Five points apart. Both land in a neutral-ish zone, so on this particular day the disagreement does not change the conclusion. On other days it has.

On 25 September 2026, CoinMarketCap printed 73 while CFGI printed 55. Eighteen points apart on the same afternoon. One dial said the market was running hot, the other said it was barely warm, and a reader checking either one would have come away with a different picture.

The reason is that these are different indices wearing the same name. Alternative.me uses the six components above and updates daily. CoinMarketCap builds its version from different inputs including implied volatility from options and the put-call ratio. CFGI runs ten indicators, computes them per asset, and refreshes every 15 minutes.

Three different input sets, three different universes, three different update clocks. There is no reason for them to agree and no sense in which one is correct.

The practical rule is the one we apply to any index series: name your source and never mix providers inside a comparison. If you are tracking a change over a week, use the same provider for both ends. A five point provider gap is larger than many weekly moves.

5. It is a bitcoin index

This one is stated plainly by the publisher and still surprises people.

Alternative.me's own page says the index covers bitcoin only, with separate indices for large altcoins described as coming later. Forbes, reviewing the same index, notes that it focuses mainly on bitcoin and so may not reflect the wider crypto market.

So the gauge that half the industry quotes as "crypto sentiment" is measuring one asset. And one of its five live components is literally bitcoin's dominance over everything else.

If you hold a diversified basket, this matters. The index can read greed while the part of the market you actually own is going nowhere, because bitcoin's volatility and momentum are driving nearly 59% of the reading and your altcoin positions are driving none of it. We went through how far bitcoin's behaviour diverges from the rest of the market in our comparison of the major index products.

6. The history that predates the index

A small thing with large implications if you are backtesting against this series.

Alternative.me's index launched in February 2018. Yet historical tables for the index show values for years before that, including readings attributed to 2011 and 2014.

Those earlier values are reconstructed. They were computed afterwards, by applying today's methodology to historical data, for a period when the index was not being published and when several of its inputs, Twitter sentiment analysis in particular, did not exist in their current form.

Backfilled history is normal practice and it is not dishonest. But if you are testing a rule like "buy when the index drops below 20", you need to know which part of your sample is a live record and which part is a reconstruction built with full knowledge of what happened next. It is the same problem we described for index charts in general in our guide to reading them.

7. How to actually use it

Given all of the above, the index still has one genuine use and several false ones.

What it is good for: spotting the extremes. When the reading goes below 20 or above 80, it is telling you that volatility, momentum, social volume and search interest are all pointing the same way at once. That is a real condition. It is crowding, and crowding is information. The index spends most of its time in the middle, so these moments are rare and worth noticing.

What it is not good for: the middle. Between roughly 35 and 65 the index is noise dressed as a signal. A move from 48 on 28 October to 51 on 29 October crosses from "fear" to "greed" on the label while changing essentially nothing about the market.

Four rules if you are going to watch it:

1. Use the number, not the label. The bands are too wide to carry meaning.
2. Stick to one provider. They disagree by more than most weekly moves.
3. Remember it is bitcoin. If you hold a broad basket, this is not your gauge.
4. Treat it as description, not prediction. Every input measures something that already happened. The index tells you where the market has been, and the fact that extremes have historically clustered near turning points is an observation about the past, not a forecast.

The honest summary: it is a well-constructed crowding indicator with a misleading name, published by several firms who compute it differently, running with one of its six components switched off.

Our live sentiment gauge, with the universe and calculation documented, is on the fear and greed page, alongside bitcoin dominance and our broad market indices. The construction rules we use are on the methodology page. If you want to test a sentiment rule against a basket of your own rather than against bitcoin, you can build and replay one in Cryptoindex Pro. The standing caveat on any backtest: no trading costs, every fill at the printed price, and assembled by someone who already knows how the story ended.

FAQ

What is the fear and greed index? A sentiment gauge that compresses several market measurements into a single value from 0 to 100, where low readings indicate fear and high readings indicate greed. In crypto the best known version is published by Alternative.me, with competing versions from CoinMarketCap and CFGI.

How is the fear and greed index calculated? Alternative.me uses six components: volatility 25%, market momentum and volume 25%, social media 15%, surveys 15%, bitcoin dominance 10% and Google Trends 10%. The survey component is currently paused, so the live index runs on the other five, renormalised. After renormalisation volatility and momentum together account for 58.8%.

What does a fear and greed index of 50 mean? Exactly neutral. The bands are 0 to 24 extreme fear, 25 to 49 fear, 50 neutral, 51 to 74 greed and 75 to 100 extreme greed. Because the greed and fear bands are each roughly 25 points wide, the label alone tells you very little; the number is more informative.

Why do different sites show different fear and greed values? Because they are different indices with the same name. On 29 October 2026 Alternative.me read 51 and CoinMarketCap read 46. On 25 September 2026 CoinMarketCap read 73 while CFGI read 55, a gap of 18 points. Note too that providers set their band boundaries differently, so the same number can carry a different label depending on where you read it. They use different inputs, different universes and different update frequencies.

Is the crypto fear and greed index only about bitcoin? Alternative.me states that its index covers bitcoin only. One of its live components is bitcoin dominance, and its volatility and momentum inputs are bitcoin's. If you hold a diversified basket, the index is not measuring your exposure.

Is the fear and greed index reliable for trading? It is a description of current crowding rather than a forecast. Its extremes, below 20 and above 80, mark moments when several measures align, which is genuinely rare and worth noting. Readings in the middle carry little information, and the index cannot anticipate news.