Why Are There Differences in Greed and Fear Indicators?

Many people may have wondered why, even though they are both called greed and fear indicators, the values from Binance, CoinMarketCap, and Coinglass differ.

Today, Coinglass’s greed and fear index dropped to 9, indicating extreme fear, while Binance’s index is still at 25, indicating fear but not yet extreme fear.

Why do these indicators, with the same name, have different values?

The reason lies in their underlying calculation methods, which differ.

According to publicly available information,

Coinglass’s fear and greed index calculation rules:

Currently, the index is only calculated for Bitcoin, updated daily, covering six core dimensions:

1. Volatility (25%): Compares Bitcoin’s current volatility and maximum drawdown with the average over the past 30 and 90 days. Abnormally high volatility usually indicates market fear.

2. Market momentum and trading volume (25%): Compares current trading volume with the historical average of market momentum. When the market is rising and buying volume continues to grow, it often indicates excessive greed.

3. Social media (15%): Analyzes Bitcoin-related topics and interactions on Twitter. If activity is abnormally high, it suggests increased public interest and a market sentiment leaning toward greed. (Reddit data analysis is still in development.)

4. Market dominance (10%): Reflects the market cap share of a specific coin within the entire market. An increasing Bitcoin dominance usually indicates conservative investor behavior, showing fear; a decreasing dominance suggests investors are more willing to take risks, showing greed.

5. Search trends (10%): Tracks Bitcoin-related search keywords via Google Trends. For example, a surge in searches for “bitcoin price manipulation” often indicates rising market panic.

The remaining 15% comes from surveys, but this method is currently paused.

In contrast, Binance and CoinMarketCap’s calculation methods are not publicly disclosed, but the rough assumptions are as follows:

• Volatility (25%): Implied volatility of BTC and ETH (BVIV/EVIV), not historical volatility.

• Market momentum (20%): Total market trading volume plus capital flow.

• Derivatives (20%): Options open interest ratios, perpetual contract funding rates, open interest.

• Stablecoins / dominance (15%): Bitcoin dominance plus changes in stablecoin supply.

• Social / search (10%): Overall social media activity across all coins plus search trends.

• On-chain data (10%): Exchange balances, stablecoin market cap share.

The fear and greed index was launched in 2018. At that time, the calculation was based solely on observable dimensions, with a model focused only on Bitcoin, covering assets centered on Bitcoin. Data sources included price, volume, social media, and search data, with a relatively low update frequency—usually daily.

Binance and CoinMarketCap’s calculation methods have since incorporated more complex models, including multiple cryptocurrencies, derivatives, and institutional data, covering more assets like BTC, ETH, mainstream coins, and derivatives. They also include data from the entire market, options, perpetual contracts, and on-chain metrics, with update frequencies increased to hourly, leading to a typical difference of 5-15 points higher than Coinglass.

In summary, both approaches are valid. Coinglass’s method is more conservative, stable, and traditional.

Binance and CoinMarketCap adopt a more aggressive and comprehensive approach, keeping pace with the times. Both methods complement each other and may be more efficient together.