Metric methodology

Per metric, the API serves: current value with normalized score and signal (/v1/metrics/{id}), full history with provenance and value_normalized (/v1/metrics/{id}/timeseries), the precomputed variant grid — change variants (absval/abschg/relchg) × time variants (now/past1h..past30d and, where forecasting passed its persistence gate, next24h/next7d/next30d) — at /v1/metrics/{id}/variants, and every computation method's result side by side at /v1/metrics/{id}/methods. Metric IDs are assigned sequentially and carry no meaning; the Type column is the classification: Proprietary = computed by Black Belt Labs and not (easily) available elsewhere — its methodology is a trade secret; Classic = a metric whose substance is available elsewhere, served here with full history, variants and provenance. Rows are anchor-linkable by metric id (e.g. #M50031).

All metrics are computed by the Black Belt Labs engine from raw ingested market data. Quantum-computed metrics run on Qiskit simulators with hardware-ready circuits. Historical values carry a provenance marker: live (computed at the time), backfill_recomputed (same formula re-run on historical data), backfill_approximate (reconstructed from approximate sources), or backfill_model_retro (AI metrics only: model retrodiction under a blind, no-dates protocol — hindsight leakage minimized by construction but not fully excludable), or forecast (a model-projected future value, never an observation).

Forecasts. Selected metrics expose next24h/next7d/next30d variants produced by statistical time-series models retrained weekly. A forecast is published for a metric and horizon only if it beat a naive no-change baseline in out-of-sample backtesting; where it did not, the variant is deliberately absent rather than fabricated. Slow-publication metrics (e.g. attested reserve breakdowns) use the current value as the stated forecast. Realized forecast accuracy is tracked continuously and errors are retained. Forecasts are statistical projections, not predictions of actual outcomes, and are not investment advice.

Computation methods. Per our data doctrine, the same metric may be computed with multiple independent methods — progr (classical/programmatic), learn (AI/machine learning), and quantum (quantum circuits on simulators, hardware-ready). Each computation is stored and labeled separately; the catalog lists every method used per metric and /v1/metrics//methods returns all current computations side by side. Agreement across methods strengthens a signal; divergence is itself information.

Live freshness: status page · pricing · API docs

AI

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M10121	Regime Transition Probability (7d)	Proprietary	progr	Classical (SQL)	hourly	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M50019	AI Market Narrative	Proprietary	learn	AI-Enhanced (Gemini)	1h	pro	public	Proprietary methodology — Black Belt Labs trade secret.
M50020	AI Anomaly Detection	Proprietary	learn, progr	AI-Enhanced (Gemini)	30min	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.

Analytics

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M50016	Regime Detection	Proprietary	learn, progr	AI-Enhanced (Gemini)	30min	pro	public	Proprietary methodology — Black Belt Labs trade secret.

QC

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M50021	Portfolio Optimization	Proprietary	progr, quantum	Quantum Computing (Qiskit) — simulator, hardware-ready	1h	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M50022	Risk Simulation (VaR)	Proprietary	progr, quantum	Quantum Computing (Qiskit) — simulator, hardware-ready	1h	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M50050	QC Portfolio Optimization	Proprietary	progr, quantum	Quantum Computing (QAOA) — simulator, hardware-ready		enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M50051	QC Monte Carlo Risk	Proprietary	progr, quantum	Quantum Computing (QAE) — simulator, hardware-ready		enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M50052	QC Correlation Solver	Proprietary	progr, quantum	Quantum Computing (Eigensolver) — simulator, hardware-ready		enterprise	public	Proprietary methodology — Black Belt Labs trade secret.

defi

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M10033	DeFi Rehypothecation Ratio	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M50025	DeFi TVL Momentum	Classic	progr	Classical (SQL)		basic	public	Total DeFi TVL in billions USD across tracked protocols; signal from average 24h TVL change.
M50026	Protocol Security Score	Classic	progr	Classical (SQL)		pro	public	Percentage of tracked DeFi protocols with positive 24h TVL change.
M50027	Smart Contract Risk Index	Classic	progr	Classical (SQL)		pro	public	Absolute average 7-day TVL change across tracked DeFi protocols.

derivatives

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M10028	Variance Risk Premium	Classic	progr	Classical (SQL)	daily	basic	public	Deribit DVOL implied volatility minus BBL-computed 30-day realized BTC volatility (annualized, from hourly prices), in volatility points. Positive = options priced rich vs delivered volatility; negative = cheap. The gate for judging whether protective options are worth their premium.
M10037	Funding Extremity Percentile	Classic	progr	Classical (SQL)	daily	basic	public	Percentile rank of the current 3-day average BTC perp funding rate within its trailing 2-year distribution. Values near 100 = extremely crowded longs, near 0 = extremely crowded shorts; extremes historically precede mean-reversion (H10054). Computed from Binance BTCUSDT funding history since 2019.
M10111	Open Positions (binance)	Classic	progr	Classical (SQL)	1h	enterprise	public	Total BTC+ETH perp open interest on Binance, billions USD (OI x mark price).
M10112	Open Positions (okx)	Classic	progr	Classical (SQL)	1h	enterprise	public	Total BTC+ETH perp open interest on OKX, billions USD.
M10113	Open Positions (bybit)	Classic	progr	Classical (SQL)	1h	enterprise	public	Total BTC+ETH perp open interest on Bybit, billions USD.
M50006	Counterparty Exposure Index	Proprietary	progr	Classical (SQL)	15min	basic	public	Proprietary methodology — Black Belt Labs trade secret.
M50014	Funding Rate Sentiment	Classic	progr	Classical (SQL)	5min	pro	public	Average annualized perpetual funding rate across Binance, OKX and Bybit BTC/ETH perps (8h rate × 3 × 365).
M50015	Open Interest Ratio	Classic	progr	Classical (SQL)	5min	pro	public	BTC share of combined BTC+ETH perpetual open interest in USD (Bybit).
M50041	Options Put/Call Ratio	Classic	progr	Classical (SQL)		pro	public	Total put open interest divided by total call open interest on Deribit BTC+ETH options.
M50042	Perpetual Premium Index	Classic	progr	Classical (SQL)		pro	public	Average perp basis: (mark - index) / index in percent across tracked perps.

exchange

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M10024	Shadow Settlement Ratio	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M50005	Exchange Reserve Risk	Classic	progr	Classical (SQL)	15min	basic	public	Proxy for exchange reserve drawdown risk from BBL-tracked ETH exchange hot wallets (5 addresses: Binance, Coinbase, Kraken). Cumulative net inflow forms a reserve-level proxy; the metric is its drawdown from the trailing 365-day peak, expressed in days of typical (180-day average) gross flow. Higher = reserves draining faster from their peak; normalized 50 = about 6 days of typical flow withdrawn. Declared proxy: ETH hot wallets only (BTC cold wallets trade too rarely for a flow-normalized level); does not cover full exchange reserves.
M50044	Exchange Liquidity Depth	Classic	progr	Classical (SQL)		pro	public	Average combined bid+ask depth within ±1% of mid (USD) across tracked order books.

gaming

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M50034	Gaming Token Index	Classic	progr	Classical (SQL)		pro	public

macro

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M10026	Carry Exodus Pressure	Classic	progr	Classical (SQL)	daily	basic	public	3-month US T-bill yield minus annualized 30-day average BTC perp funding, in percentage points. When the risk-free real dollar out-yields the crypto-native dollar, the structural reason to keep capital inside the stablecoin system erodes - measures when the yield engine of H10022 loses fuel.
M10030	Cross-Asset Decoupling Shock	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M10038	Liquidity Tide Index	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M10041	MSTR Relative Bid	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M10042	Weekend Drift Differential	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.

market

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M50023	Altcoin Season Index	Classic	progr	Classical (SQL)		basic	public	Percentage of top-50 non-stablecoin altcoins outperforming BTC over 30 days.
M50024	Market Breadth Indicator	Classic	progr	Classical (SQL)		basic	public	Percentage of top-100 coins with positive 24h price change.

nft

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M50033	NFT Market Health Index	Classic	progr	Classical (SQL)		pro	public

onchain

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M10025	Urgency Asymmetry (currently stale — data source pending)	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M10029	Miner Margin Squeeze	Classic	progr	Classical (SQL)	daily	basic	public	Miner economics proxy: BTC price divided by (network difficulty x natural-gas price), indexed to its 2-year average (value 100 = average margin). Normalized score is the inverse percentile - high score = miners squeezed hardest in 2 years, historically a capitulation/bottom antenna. Energy proxied by NG futures; difficulty from the public Bitcoin dataset.
M10032	Mining Pool Concentration	Classic	progr	Classical (SQL)	daily	basic	public	Herfindahl index of Bitcoin mining-pool block shares over trailing 30 days, from coinbase-tag identification of mined blocks (public blockchain data). Reported alongside the identified share; unidentified blocks are excluded from the index and disclosed. High concentration = fewer independent block producers than the decentralization narrative implies.
M10036	Bridge Concentration Index	Classic	progr	Classical (SQL)	daily	basic	public	Herfindahl index of 30-day flow shares across BBL's tracked L2 bridge contracts (Base, Optimism, zkSync, Polygon), normalized 0 (even) to 100 (single bridge). High concentration = more value exposed to a single bridge's failure - the H10053 blast-radius gauge. Declared fixed four-bridge set.
M10039	Sanctioned Wallet Activity	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M10040	Halving Cycle Phase	Classic	progr	Classical (SQL)	daily	basic	public	Position within the current Bitcoin halving cycle: days elapsed since the last halving as a percentage of the historical inter-halving interval. A pure calendar function (halvings 2012-11-28, 2016-07-09, 2020-05-11, 2024-04-19) - the narrative clock that H10060 says is tradeable despite being fully public.
M10123	Exchange Cold Wallet BTC Net Flow (7d)	Proprietary	progr	Classical (SQL)	hourly	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M50010	Gas Price Index	Classic	progr	Classical (SQL)	1h	basic	public	AI NLP analysis of regulatory news
M50011	Network Congestion Score	Classic	progr	Classical (SQL)	1h	basic	public	Latest Ethereum block gas utilization: gas_used / gas_limit, in percent.
M50012	Mempool Pressure Index	Classic	progr	Classical (SQL)	15min	pro	public	Pending transaction count in the Ethereum mempool.
M50013	Transaction Volume Index	Classic	progr	Classical (SQL)	30min	basic	public	Transaction count of the latest sampled Ethereum block.
M50032	Smart Contract Activity	Classic	progr	Classical (SQL)		pro	public	Contract calls as a percentage of transactions in the latest sampled block.
M50037	Whale Accumulation Index	Classic	progr	Classical (SQL)		pro	public	Sum of tracked non-exchange whale wallet balances (ETH).
M50038	Exchange Reserve Flow	Classic	progr	Classical (SQL)		pro	public	Sum of tracked exchange wallet balances (ETH); lower is bullish.
M50039	Bridge Flow Monitor	Classic	progr	Classical (SQL)		pro	public
M50040	Smart Contract Activity Index	Classic	progr	Classical (SQL)		pro	public
M50043	Exchange Inflow/Outflow Ratio	Classic	progr	Classical (SQL)		pro	public

regulatory

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M10022	Announcement Inversion Score	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M10035	Regulatory Tone Whiplash	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M50028	Regulatory Sentiment Index	Proprietary	progr	Classical (SQL)		pro	public	Proprietary methodology — Black Belt Labs trade secret.
M50029	Compliance Risk Score	Classic	progr	Classical (SQL)		pro	public	Average GDELT negative-tone score of crypto-regulation news, scaled ×10 (0-100).

sentiment

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M50017	Sentiment	Classic	progr	Classical (SQL)	15min	basic	public	AI NLP sentiment from news and social
M50030	Social Sentiment Score	Classic	progr	Classical (SQL)		basic	public
M50031	Fear & Greed Index	Classic	progr	Classical (SQL)		basic	public	Passthrough of the Alternative.me crypto Fear & Greed index (0-100).

social

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M10023	Degenerate-Canary Divergence (currently stale — data source pending)	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M10027	Herding Reflexivity Index	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.

stablecoin

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M50001	Stablecoin Health Index	Proprietary	progr	Classical (SQL)	5min	basic	public	Proprietary methodology — Black Belt Labs trade secret.
M50002	Tether Printer Activity	Classic	progr	Classical (SQL)	5min	basic	public	USDT total supply, in billions.
M50003	Stablecoin Dominance Ratio	Classic	progr	Classical (SQL)	10min	basic	public	USDT market cap as a percentage of total tracked stablecoin market cap.
M50004	Stablecoin Flow Indicator	Proprietary	progr	Classical (SQL)	15min	basic	public	Proprietary methodology — Black Belt Labs trade secret.
M50035	Tether Treasury Flow	Classic	progr	Classical (SQL)		pro	public	USDT market cap, in billions USD.
M50036	Stablecoin Velocity Index	Classic	progr	Classical (SQL)		pro	public	USDT circulating supply as a percentage of total supply.

tether

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M10004	USDT Peg Deviation	Classic	progr	Classical (SQL)	1h	enterprise	public	USDT market price deviation from 1.000 in percent (internal: CoinGecko price).
M10005	USDT Bitfinex Premium	Proprietary	progr	Classical (SQL)	1h	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M10008	Tether Gross Printing 24h	Classic	progr	Classical (SQL)	1h	enterprise	public	Change in USDT total supply over 24h, billions.
M10009	Tether Net Printing 24h	Classic	progr	Classical (SQL)	1h	enterprise	public	Change in USDT circulating supply over 24h, billions.
M10011	Tether Published Treasury	Classic	progr	Classical (SQL)	1h	enterprise	public	Tokens authorized but held in Tether treasury (unissued), billions, from Tether published transparency data.
M10016	Inverse-Perp Dissonance	Proprietary	progr	Classical (SQL)	hourly	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M10019	Dollar-Refuge Demand Index	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M10020	Peg Suppression Index	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M10021	Print-to-Pump Differential	Proprietary	progr	Classical (SQL)	daily	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M10101	Tether Reserves: Cash	Classic	progr	Classical (SQL)	1h	enterprise	public	Tether reserves per asset class, billions USD, from Tether published attestation figures.
M10102	Tether Reserves: Tbills	Classic	progr	Classical (SQL)	1h	enterprise	public	Tether reserves per asset class, billions USD, from Tether published attestation figures.
M10103	Tether Reserves: Metals	Classic	progr	Classical (SQL)	1h	enterprise	public	Tether reserves per asset class, billions USD, from Tether published attestation figures.
M10104	Tether Reserves: Bitcoin	Classic	progr	Classical (SQL)	1h	enterprise	public	Tether reserves per asset class, billions USD, from Tether published attestation figures.
M10105	Tether Reserves: Bonds	Classic	progr	Classical (SQL)	1h	enterprise	public	Tether reserves per asset class, billions USD, from Tether published attestation figures.
M10106	Tether Reserves: Equities	Classic	progr	Classical (SQL)	1h	enterprise	public	Tether reserves per asset class, billions USD, from Tether published attestation figures.
M10107	Tether Reserves: Loans	Classic	progr	Classical (SQL)	1h	enterprise	public	Tether reserves per asset class, billions USD, from Tether published attestation figures.
M10108	Tether Reserves: Other	Classic	progr	Classical (SQL)	1h	enterprise	public	Tether reserves per asset class, billions USD, from Tether published attestation figures.
M10109	Tether Reserves: Total	Classic	progr	Classical (SQL)	1h	enterprise	public	Tether reserves per asset class, billions USD, from Tether published attestation figures.
M10122	USDT De-Peg Probability (30d)	Proprietary	progr	Classical (SQL)	hourly	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M10124	Tether Treasury USDT Net Flow (7d)	Proprietary	progr	Classical (SQL)	hourly	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M50053	Stablecoin Chatter Stress	Proprietary	progr	Classical (SQL)	hourly	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.

volatility

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M50018	Implied Volatility Index	Classic	progr	Classical (SQL)	15min	basic	public	Deribit BTC DVOL (annualized implied vol) divided by sqrt(365): daily implied volatility in percent.

volume

ID	Metric	Type	Methods	Computation	Cadence	Tier	Personalization	Methodology
M10018	Volume/Depth Honesty Ratio	Proprietary	progr	Classical (SQL)	hourly	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M10031	Cap-to-Depth Illusion Ratio	Proprietary	progr	Classical (SQL)	hourly	enterprise	public	Proprietary methodology — Black Belt Labs trade secret.
M50007	Volume Authenticity Score	Proprietary	progr	Classical (SQL)	30min	basic	public	Proprietary methodology — Black Belt Labs trade secret.
M50008	Liquidity Fragmentation Index	Classic	progr	Classical (SQL)	15min	basic	public	Cross-venue liquidity fragmentation from BBL orderbook snapshots (Binance, Coinbase, Kraken; BTC and ETH books). Each venue's share of total quoted depth within 2% of mid (bids+asks, latest snapshot per book) feeds a Herfindahl index, normalized so 0 = all depth on one venue and 100 = depth spread evenly across the three tracked venues; averaged over BTC and ETH. Venue set is declared and fixed at three.
M50009	Market Depth Ratio	Classic	progr	Classical (SQL)	15min	pro	public	Average bid-to-ask depth ratio within ±1% of mid across tracked exchange order books (30h window).