How the Clever Score Works
Methodology v1.0 — locked July 2, 2026. The factors, thresholds, and math below are frozen; any future change becomes a new, clearly-labeled version.
The score in one paragraph
Every coin is graded 0–100 from 30 factors across 5 pillars— tokenomics, on-chain health, technical momentum, capital flow, and liquidity & risk — each factor scored 1–10 against fixed, pre-registered thresholds. Pillars blend into the overall score using your chosen profile (Conservative, Balanced, Aggressive, or Elite custom weights). Scores refresh every ~4 hours across our full universe of 1,000 real cryptocurrencies — no stablecoins, wrapped tokens, staked derivatives, or tokenized funds.
Honesty rules
- No fake data. Every factor is backed by a named source (CoinGecko, DefiLlama, Santiment, CryptoQuant, Dune, GitHub) or computed transparently from price history.
- Not applicable ≠ missing.A store-of-value coin has no TVL; Bitcoin has no “strength vs Bitcoin.” Structurally inapplicable factors are excluded from the math entirely — never scored as failures.
- Thin data gets dampened.Coins with few measurable factors are pulled toward a neutral 50 by a confidence system, so a coin can’t post an extreme score off a handful of data points.
Pillar 1: Tokenomics & Value Accrual (balanced weight 20%)
Supply structure and whether holders are positioned to capture value — dilution risk, unlock overhangs, inflation, and where the market sits versus on-chain cost basis.
How much future dilution is priced in — fully-diluted value vs today's cap.
Whether large token unlocks loom that could flood supply.
How fast new tokens enter circulation and dilute holders.
Staking APR minus inflation — the yield that's actually real. Only applies to proof-of-stake assets.
Price versus the average on-chain cost basis — stretched or undervalued.
Share of supply underwater — trapped sellers waiting to exit into strength.
Pillar 2: On-Chain Health (balanced weight 20%)
Is the network actually being used and built on? Real usage, real fees, real developers — the fundamentals that survive market cycles.
Capital deposited into the protocol's contracts, 30-day trend. Applies to chains and DeFi.
Development events / commits over 30 days — is anyone still building?
Real fees paid to use the protocol — demand that can't be faked. Applies to fee-generating assets.
What the market pays per real daily user — valuation per participant.
30-day growth in active addresses — expanding or shrinking user base.
Market-cap rank change over 30 days — climbing or slipping vs the whole market.
Pillar 3: Quantitative & Technical Momentum (balanced weight 20%)
Price structure across timeframes — trend, volatility state, momentum, and volume confirmation, plus our proprietary risk-adjusted momentum measures.
Whether short, medium and long trends agree — confirmed uptrends score highest.
Current volatility vs its own history — coiled spring or already exploded.
Overbought / oversold positioning on the daily chart.
Volume-flow divergences that often front-run price reversals.
Proprietary: 90-day return divided by volatility — a crypto Sharpe ratio rewarding durable gains.
Proprietary: the coin's weekly risk-adjusted momentum ranked against every coin we score.
Pillar 4: Market & Capital Flow (balanced weight 20%)
Where capital is rotating — relative strength against Bitcoin and sector peers, volume-weighted money flow, and large-holder movement.
90-day performance versus Bitcoin — the benchmark every trader actually uses.
Proprietary: performance versus the coin's own sector average — leader or laggard among peers.
Volume on up-days vs down-days — is capital flowing in or being distributed?
Trend in $100k+ on-chain transfers — rising or falling large-holder activity.
Proprietary: the trend of the coin's own Clever Score — fundamentals improving or deteriorating.
Proprietary: average score trend across the coin's sector — sector rotation from our own universe.
Pillar 5: Liquidity & Risk (balanced weight 20%)
Can you actually trade it safely? Liquidity depth and authenticity, downside behavior, listing quality, and holder concentration.
Proprietary composite: venue breadth, spread tightness, and concentration.
Share of volume on tight-spread credible venues — filters wash trading.
Where price sits in its 90-day range — how it holds up under stress.
Listings on elite exchanges with strict standards (Binance, Coinbase, Kraken…) vs broad-listing venues.
How violently the coin moves relative to Bitcoin.
Share of supply in the largest wallets — extreme concentration flags dump risk.
Does it work? The backtest
After locking v1.0 we recomputed historical scores at 36 monthly dates (May 2023 – April 2026, 281 coins, 6,013 coin-months) using only data knowable at each date, then measured forward returns. High-scoring coins beat Bitcoin roughly 40–50% of the time versus ~31–33% for low-scoring coins — and the typical low-score coin lost double digits over 90 days.
| Score quintile at scoring date | Beat BTC over next 90d | Median 90d return |
|---|---|---|
| Q1 (lowest scores) | 33% | −6.0% |
| Q2 | 35% | −4.5% |
| Q3 | 36% | −1.4% |
| Q4 | 41% | 0.0% |
| Q5 (highest scores) | 40% | 0.0% |
Full disclosure: the backtest recomputes the 13 factors derivable from price/volume history (the remaining 17 use provider history that isn’t available retroactively and are treated as missing, exactly as the live engine treats them); the universe is today’s top coins, so results carry survivorship bias toward coins that lived; and results vary by regime — the 2024 memecoin bull was the noisiest period, the 2025–26 market the cleanest. Backtested performance is not a guarantee of future results.
What the Clever Score is not
The Clever Score is a research tool, not financial advice and not a buy/sell signal. It measures quality and momentum characteristics across public data; it cannot know your risk tolerance, time horizon, or the future. Crypto assets can lose most or all of their value. Do your own research.