Crypto Mindshare Tracker: What Watchlist Share Measures

What crypto mindshare measures on Sharpe

"Mindshare" gets used loosely in crypto. On Sharpe it means one specific thing: each narrative's share of a CoinGecko watchlist score across every narrative we track. The inputs are CoinGecko watchlist users and CoinGecko up-vote sentiment. No social-media post, mention, or follower count enters the score.

That makes it a narrow, checkable number. It tells you which narratives CoinGecko users are watching, relative to each other, normalized so AI Agents (a small basket of tokens) and Layer 1 (a large basket) can be compared on the same axis. It does not tell you where price goes next, and we have not published evidence that it leads price.

Where watchlist share sits right now

The leaderboard below pulls live mindshare data from the Sharpe tracker: 30 curated narratives, refreshed every 12 hours. The 7d and 24h columns are changes in share, in percentage points: a narrative at 6.0% that was at 5.5% a week ago shows +0.5 pp, not +9%.

A few things to notice when you read this view:

• Sentiment is separate from mindshare share. It is CoinGecko's up-vote percentage, weighted by market cap across the narrative's tokens that report it, and it moves a token's score by at most 10% of its watchlist users.
• Tokens column shows how many coins are in each narrative basket. The score uses at most each narrative's top 40 coins by market cap. Narratives with few tokens swing more, because a single coin drags the whole basket.
• Share is relative. A narrative's share can rise because its own watchlists grew or because other narratives' watchlists shrank. It says nothing about whether total interest in crypto went up.

The Sharpe mindshare formula

Sharpe computes a token-level watchlist score first, then rolls it up into narratives. The formula mirrors the production pipeline:

Token watchlist score = (watchlist users × 0.70) + (watchlist users × ((sentiment up % − 50) / 100) × 0.20)

Where:

• Watchlist users are CoinGecko's watchlist_portfolio_users: the number of CoinGecko users who added the token to a watchlist or portfolio.
• Sentiment is CoinGecko's bullish vote percentage, centered around 50. Neutral sentiment contributes 0; 100% bullish adds 10% of watchlist weight; 0% bullish subtracts 10%. A token with no vote sample gets no sentiment term at all, rather than a neutral one.

A token that belongs to several narratives is split evenly across them, so overlapping categories are not counted twice. Mindshare percentage is each narrative's split score divided by the total score across all tracked tokens, so the column always sums to 100%. 8% mindshare on AI Agents means 8% of the tracked watchlist score sits in that narrative right now.

Market data (price, market cap, 24h change) is shown beside the score for context, but it is not weighted into the core mindshare percentage. Telegram reach and GitHub developer metrics were removed after CoinGecko stopped publishing community and developer data in August 2026.

Why we publish the formula

Kaito and most other attention products gate their methodology behind enterprise pricing. We took the opposite view: publishing the methodology lets traders check exactly what the number is, including what it leaves out, and gives researchers and journalists a method they can reference.

The full methodology, including the signal weights and data source for each field, is described above and reflected in the live tracker at /mindshare. We version the update notes.

Reading changes in share

A change in mindshare share has a few possible causes, and the tracker cannot tell them apart on its own:

Watchlist growth. CoinGecko users added the narrative's tokens to watchlists faster than tokens elsewhere.

Basket changes. Each narrative's top 40 coins are re-picked by market cap at every sync. A large token entering or leaving the basket moves the share even if no user changed a watchlist.

Sentiment swings. A shift in up-vote percentage moves each token's score by up to 10% of its watchlist users.

Other narratives moving. Because the shares sum to 100%, one narrative's gain is another's loss.

A move that persists across several 12-hour syncs is more meaningful than a single step.

How to use mindshare data

Mindshare is context, not a trade signal on its own:

1. Compare share with size. Put a narrative's watchlist share next to its aggregate market cap on the narrative tracker. A narrative that CoinGecko users watch far more, or less, than its size suggests is worth a closer look, not a position by itself.

2. Read sentiment alongside share. High share with low sentiment means many watchers voting bearish; low share with high sentiment means a small, bullish audience. Neither pattern is a forecast.

3. Treat the narrative as a basket. Mindshare is computed at the narrative level. The Sharpe narrative tracker shows each basket's coin composition, market data, and derivatives positioning, so you can check a watchlist-share move against liquidity, funding, and open interest.

Common mistakes when reading the tracker

Treating 24h Δ as a trade signal. Twelve-hour snapshots are noisy. News events can spike watchlists or sentiment without any change in positioning. Prefer the 7d Δ, and check whether a move persists over several syncs.

Calling it social-media attention. The score counts CoinGecko watchlists and votes only. It does not see posts, mentions, or followers on X, Telegram, or anywhere else.

Mixing memecoin reflexivity with regular narratives. The memecoin tracker is a separate product with its own methodology and refresh cadence. Memecoin mindshare doesn't translate cleanly to narrative mindshare and vice versa.

Treating price context as a score input. Market cap and price change are displayed beside the score, but the core mindshare percentage comes from watchlists and sentiment. Price fields are not baked into the score.

Mindshare vs. on-chain TVL vs. price action

Mindshare is one of three separate readings worth tracking together:

• Mindshare = CoinGecko watchlist share. Twelve-hour cadence.
• TVL = capital deposited in protocols. Daily cadence. See DefiLlama or our ecosystems tracker.
• Price = what the market pays. Real-time cadence.

They measure different things, and they do not have to move together. When they diverge, the divergence is a question to investigate, not an answer about what happens next.

What's next

If you want a deeper view of any narrative on this leaderboard, click through to the per-narrative page (e.g. /mindshare/ai-agents). Each shows the historical mindshare time series, the underlying token basket, the tokens with the largest watchlist scores, and the price context beside the score.

The methodology page at /mindshare explains the exact weights and source caveats. The data is also exposed through /api/mindshare/data at the free 30 req/min tier. See the API reference for details.

Mindshare is free and the methodology is published. That's deliberate: we'd rather earn citations than gate the formula.