The Divergence Index.
Where mainstream and crypto-native media disagree.
An experimental metric that compares the sentiment of mainstream financial media with crypto-native publications, expressed as a gap in percentage points.
Example values to show how the reading works. The index is based on a limited number of articles per group, so single readings can move a lot.
Two media ecosystems. One gap that matters.
The Divergence Index is a metric developed by HafidWatch that measures the sentiment gap between two distinct categories of financial media: mainstream outlets (Bloomberg, Reuters, Financial Times, Wall Street Journal, Forbes, CNBC, NYT) and crypto-native publications (CoinDesk, Cointelegraph, The Block, Decrypt, Bankless, Blockworks, CryptoSlate, Bitcoin Magazine, Protos).
Every day, these two groups cover the same market. Most of the time, they broadly agree — when Bitcoin rallies on ETF inflows, both sides report it as bullish. The Divergence Index reads near zero. Nothing unusual is happening.
But periodically, a gap opens. Crypto-native media turns aggressively bullish while mainstream financial outlets remain skeptical — or vice versa. The Divergence Index captures that gap, expressed in percentage points, and updates as new articles are processed.
“Think of it like two experienced weather forecasters standing on opposite sides of a mountain. Both are competent. Both have good instruments. But they are measuring different atmospheric conditions. When they sharply disagree, it is worth asking why.”
Why does this divergence exist? Because these two groups serve different audiences with different incentive structures. Bloomberg and Reuters write for institutional portfolio managers and macro strategists — their filters emphasize regulatory risk and systemic stability. CoinDesk and The Block write for crypto-native traders and builders — their filters emphasize adoption signals, protocol upgrades, and on-chain momentum.
Neither group is wrong. They are looking at the same market through different lenses. The Divergence Index quantifies how far apart those two readings are, so the disagreement can be tracked over time. Whether that gap says anything about future price moves is an open question, and we are testing it.
How the Divergence Index is calculated.
The calculation follows a four-stage pipeline that runs continuously across the sources we monitor.
Source Classification
Every source is permanently classified as either “mainstream financial” or “crypto-native.” Bloomberg, Reuters, FT, WSJ, Forbes, CNBC, and NYT are mainstream. CoinDesk, Cointelegraph, The Block, Decrypt, Bankless, Blockworks, CryptoSlate, Bitcoin Magazine, and Protos are crypto-native. Classification is fixed at the source level — it never shifts based on individual article tone. Sources without a clear editorial classification are excluded from the gap calculation to preserve signal integrity.
Article-Level Sentiment Scoring
An AI model classifies each article as bullish, bearish, neutral or mixed. For aggregation, these labels are mapped to a 0-to-1 scale: bullish = 1, neutral or mixed = 0.5, bearish = 0. The classification is automated and can be wrong on individual articles.
Group Aggregation
Sentiment scores are aggregated separately for each media group over a rolling 24-hour window — every qualifying article counts equally within its group. We show the number of articles (n) behind each group so you can judge how much weight a reading deserves. Tier- and recency-weighting is on the roadmap, not yet part of the calculation.
Gap Calculation
The Divergence Index is the absolute difference between the two group scores, expressed in percentage points. If mainstream sentiment sits at 0.48 and crypto-native sentiment sits at 0.86, the index reads 38pp, with the direction labeled “Crypto-native more bullish.”
How to read the Divergence Index.
Small gaps are normal: mainstream and crypto media naturally differ in tone on any given day. The ranges below are illustrative. They have not been validated against price data.
Since 9 June 2026, weekly readings based on at least 5 mainstream articles have ranged from about 0 to 25pp, and we have not observed readings above 40pp. With a small sample, a single article in one group can move the index by roughly 10pp.
Aligned
Both media groups are telling roughly the same story.
Moderate gap
One group leans in a direction the other has not adopted. Worth watching, and often within normal variation.
Large gap
Uncommon in our data so far. Check the article counts (n) first: a large gap on a small sample is weak evidence.
Why we track it, and what we don’t claim.
Mainstream financial media and crypto-native media serve different audiences and work on different editorial cycles. Mainstream outlets typically have longer review processes and more conservative framing. Crypto-native outlets are closer to builders and trading desks and often move faster on narratives.
Our hypothesis is that a persistent gap between the two may reflect a narrative one side has not yet adopted. We are testing whether the gap has any relationship with later price moves. We have not established one, so the Divergence Index should be read as a description of how two groups of media differ, not as a forecast.
The Divergence Index does not predict price, does not tell you what to buy, and is not a guarantee of anything. It is an input for your own research.
Ready to access the Divergence Index?
Available via the Intelligence Report (weekly PDF) or the Sentiment API (endpoint).
Divergence Index vs. alternative sentiment tools.
Each sentiment product measures something different. The Divergence Index looks at the gap between two groups of professional media, which is not what the tools below measure.
| Feature | Divergence Index (HafidWatch) | Fear & Greed (Alternative.me) | Galaxy Score (LunarCrush) | Social Volume (Santiment) |
|---|---|---|---|---|
| What it measures | Gap between mainstream & crypto-native media | Composite market emotion | Social media engagement | Social mention volume |
| Data sources | Bloomberg, Reuters, FT, WSJ, Forbes, CNBC, NYT, CoinDesk, Cointelegraph, The Block, Decrypt, Bankless, Blockworks, CryptoSlate, Bitcoin Magazine, Protos | Market data, Google Trends, social, surveys | Twitter/X, Reddit, YouTube, TikTok | Twitter/X, Reddit, Telegram |
| Update frequency | As new articles are processed | Daily | Hourly | Real-time |
| API available | Yes — /v1/divergence/current | Yes (free) | Yes (paid) | Yes (paid) |
| Price | From $49/mo | Free | Paid tiers | Paid tiers |
← Swipe the table to see all columns →
The Fear & Greed Index is useful as a broad daily market sentiment indicator but doesn’t differentiate between media types. LunarCrush excels at measuring retail attention and viral momentum — it does not analyze professional financial journalism. Santiment tracks social buzz velocity but measures crowd behavior, not editorial divergence. Each has legitimate use cases. The Divergence Index is a different measurement, not a replacement for them.
Three ways to use the Divergence Index.
The Divergence Index is available through three HafidWatch products, each designed for a different use case.
Intelligence Report
Weekly PDF delivered every Thursday. Includes the Divergence Index reading with editorial interpretation, narrative context, and week-over-week comparison. Best for traders who want analysis alongside the data.
See details →Sentiment API
Two dedicated divergence endpoints: /current for the latest gap and /history for weekly history — plus 14 more endpoints covering sentiment, tickers, signals, market data, and webhooks. Built for developers and quant traders who need programmatic access.
Premium Alerts
The Narrative Divergence alert fires automatically to Telegram when the index crosses a configured threshold. Includes the gap value, direction, market data, and AI-generated context.
See alert types →Common questions about the Divergence Index.
How often is the Divergence Index updated?
The index updates as new articles enter the HafidWatch pipeline, which monitors its sources continuously throughout the day. In practice, the value shifts several times a day during active news cycles and may remain stable for longer stretches during quiet periods.
What sources does the Divergence Index include?
The mainstream group includes Bloomberg, Reuters, Financial Times, Wall Street Journal, Forbes, CNBC, and NYT. The crypto-native group includes CoinDesk, Cointelegraph, The Block, Decrypt, Bankless, Blockworks, CryptoSlate, Bitcoin Magazine, and Protos. Sources are classified permanently by editorial type. Sources without a clear editorial classification — such as community platforms and aggregators — are excluded from the gap calculation to keep the signal clean.
Is the Divergence Index a trading signal?
No. The Divergence Index is an analytical measurement of media sentiment divergence. It identifies when two media ecosystems are telling different stories about the same market. We have not established that it precedes price movements. It does not generate buy or sell recommendations and does not constitute financial advice. See our Financial Disclaimer.
How is it different from the Fear & Greed Index?
The Fear & Greed Index combines market volatility, trading volume, social media sentiment, Google Trends, and surveys into a single composite score representing overall market emotion. The Divergence Index measures the gap between mainstream financial media sentiment and crypto-native media sentiment. A market can be at “Extreme Greed” with a low Divergence Index (both media groups agree) or with a high Divergence Index (they sharply disagree about why). The two measure different things.
Can I access historical Divergence Index data?
Yes, on every plan — depth scales with your plan. Developer gets 7 days of history, Starter 30 days, Pro 90 days, and Enterprise the full 365 days. The Intelligence Report includes week-over-week comparisons but not full historical datasets.
Does the Divergence Index work for altcoins or only Bitcoin?
Today, the Divergence Index reading is market-wide: it covers crypto-related coverage rather than isolating a single asset. Ticker-level divergence is on the roadmap but not available yet. For per-ticker sentiment today, see /v1/tickers/{ticker}/sentiment, which is live now.
Access the Divergence Index.
Start with a free trial of the Sentiment API or subscribe to the Intelligence Report.