What is CryptoRefract? It is an independent, transparent reference that decomposes a crypto asset's price into a waterfall — book value, residual income, and the premium priced on top — and reports the coverage (the measured anchor): how much of the price is anchored to value we can measure today. It reads each protocol in the language of finance (production cost as book value, base-fee burn as a buyback, a discount rate to capitalize cash flow) so the question "is this price grounded or hopeful?" can be examined with normal financial tools.
Does it tell me a fair price, or whether to buy or sell? No. CryptoRefract does not publish fair values, price targets, or buy/sell signals, and nothing on it is investment advice. It describes how much of a price is currently anchored to measurable value and by what mechanism value reaches the token. Any decision is yours to make.
Is this investment advice? No. Everything here is informational and descriptive. We deliberately avoid words like "overvalued," "undervalued," "cheap," "expensive," or "bubble." See the Disclaimer for the full statement.
Where does the data come from? The decomposition is built only from data we compute ourselves ("anchor") — on-chain base-fee burn via public RPC, and self-read supply where a chain exposes it (Bitcoin via mempool.space, Solana via the getSupply RPC). Price comes from an exchange public ticker (Kraken) and on-chain DEX trades (DexScreener); macro from US Treasury, World Bank, and the NY Fed; sentiment from Alternative.me (credited). Price and macro are context with a source + as-of badge — they never silently feed the ratios. Every source is keyless and commercial-use-clean; node-only metrics are shown as N/A, never estimated.
Why only some protocols, and not every coin? We cover a curated set of fee-generating protocols — the ones where "revenue" and "holder accrual" are meaningful enough to decompose. For assets with no real fees or no accrual mechanism, the model has little to say, so including them would add noise, not signal. The scope grows deliberately as we extend coverage.
What does the premium being "narrative-based" mean? When we solve the price for the growth that a perpetual cash-flow model would need to justify it, and that implied growth exceeds the discount rate, no such model can justify the price — so the premium leans on finite-horizon, narrative-based expectation. We show that as a qualitative label, an observation, never a warning or a recommendation.
Why do some assets show N/A? Either the metric doesn't apply under the asset's lens (e.g. a monetary asset has no holder cash flow to divide), or it needs infrastructure we don't run (UTXO cost-basis metrics, total gas revenue, some chains' supply). N/A is never a zero, and never a fabricated number — the blank is honest.
How often is it updated? Anchor figures refresh daily; every figure carries an as-of date and an annualization basis. Stale data is flagged rather than shown as current.
Is it free? Yes. CryptoRefract is free and ad-supported — the reference is openly readable, with no paywall.
How is the decomposition calculated? We follow Ohlson's identity, conservatively: price = book value + residual income + premium. Book value is the realized net-asset stock (for Bitcoin, cumulative production cost); residual income is the holder cash flow (base-fee burn) capitalized with no growth at a discount rate; the premium is whatever the price exceeds those two by. We call (book value + residual income) the measured anchor, and the premium the future expectations priced on top; coverage = measured anchor ÷ price — the same number on the asset page and the screener. See the Methodology for detail.
Can the figures be wrong or change? Yes — protocols change their mechanisms (a fee switch turns on, a buyback pauses), and data sources have limits. We record the verified state and its date, re-check before relying on it, and show confidence levels openly. When we are not sure, we say so rather than guess.