Data sources behind Turnmetry model v3
Turnmetry favours sources that can be collected freely and automatically. Raw provider fields, locally derived indicators and approximations are kept conceptually separate.
Last reviewed: July 12, 2026
Primary market and network data
The daily model pipeline uses Bitcoin data from the Coin Metrics Community API, including USD price, market capitalisation and the current MVRV ratio where available. These fields form the basis for both direct observations and reproducible local calculations.
Provider availability, definitions and access terms can change. “Free” describes the current access route, not a guarantee that a source will remain free indefinitely. Turnmetry can replace a provider when necessary, but a source change must be tested and documented because seemingly equivalent fields can differ.
Price and market value
Daily Bitcoin price and market-cap data provide cycle location, drawdown and the base series for long-term moving averages.
MVRV ratio
MVRV compares market value with realised value. It supports valuation context and the local derivation of NUPL and MVRV-Z.
Price ratios
Mayer Multiple, Pi Cycle ratio, the two-year moving-average ratio and the 200-week average are calculated from the daily price series.
Puell-style miner metric
Turnmetry estimates daily miner revenue from price and the scheduled Bitcoin issuance, then compares it with its 365-day average.
How derived indicators are calculated
Drawdown is measured from the highest daily close observed up to that date. NUPL is derived as 1 − 1/MVRV. MVRV-Z uses the difference between market and realised value divided by an expanding standard deviation of market capitalisation, with a minimum history requirement.
The Mayer Multiple divides price by its 200-day average. The Pi Cycle ratio compares the 111-day average with twice the 350-day average. The two-year ratio compares price with its 730-day average. These are correlated price transformations, so model v3 does not count them as fully independent evidence.
The Puell-style series is an approximation based on scheduled issuance and price. It is not a direct feed of actual miner revenue, fees or miner operating costs.
Freshness and missing data
The model evaluates closed daily observations. Required inputs are aligned to the same date, and the pipeline can refuse to create a new snapshot when mature model data is incomplete or stale. This is preferable to silently carrying an old value forward and presenting it as current.
Optional market data may be collected for research without participating in the active v3 decision path. For example, perpetual-futures funding is not required by the current core gates. A future change to that role belongs in the model changelog.
From inputs to signals
Data quality is only one part of the system. The methodology explains causal snapshots and staged confirmation. See the top indicator guide and bottom indicator guide for how the active model interprets these families.
View current cycle status