An Expanding-Window Percentile Confluence Model of Bitcoin Market Cycles

This paper documents the SatoshiMacro Model (SMM), a daily score from 0 to 100 that summarises where Bitcoin sits within its historical cycle range. The score combines 48 heterogeneous signals in six tiers: cycle timing and market psychology, valuation, miner economics, sentiment and derivatives positioning, rotation and institutional flows, and macro conditions. Of these, 44 are normalised by their expanding-window percentile rank, so each reading uses only that signal's own history up to the same day, and four enter through fixed mappings. Tier means are combined with fixed weights, renormalised when tiers are missing, and passed through a piecewise-linear calibration. On daily BTC/AUD data from 1 January 2013 to 5 October 2026 (5,026 days), the calibrated score places all seven reference cycle inflections in their target zones, compared with four of seven before calibration. That result is in-sample by construction: the calibration knots were set on the same seven dates, and the top zone is occupied on 19.9 per cent of all days. Forward returns by zone are not monotone. The median 365-day BTC/AUD return is 84.6 per cent from the Accumulation zone, -24.8 per cent from Distribution and 22.7 per cent from Cycle Top, and the rank correlation between the score and 365-day forward returns is -0.14. The series is released under CC BY 4.0. Working paper documenting the construction, data and in-sample diagnostics of the SatoshiMacro Model. The daily model series is archived separately at doi:10.5281/zenodo.23131351. Live model and methodology: satoshimacro.com/tools/crypto/satoshimacro-model/