How Turnmetry identifies Bitcoin cycle tops and bottoms
A transparent framework for turning daily Bitcoin market data into early warnings, reversals and confirmations—without pretending uncertainty can be removed.
Last updated: September 18, 2026 · evaluation clarification; v3 rules unchanged
Why signals come in stages
An exact top is only knowable after the market has moved away from it. A useful cycle model therefore has to balance speed against certainty. Turnmetry keeps those jobs separate: candidates warn early, turns identify a reversal, and confirmations require stronger structural evidence.
Peak Candidate
An early warning when price reaches a new cycle high while timing, valuation or overheating evidence indicates elevated risk. It is deliberately allowed to be wrong.
Top Turn
A faster retrospective signal after price has retreated from a candidate high and short-term trend conditions have broken.
Top Confirmed
A slower signal requiring a decline large enough to qualify as structural rather than ordinary volatility.
Bottom Watch
A separate process combining valuation, drawdown, miner stress and trend recovery evidence to monitor capitulation and recovery.
Causal calculations and historical replay
For each replayed date, calculations use observations dated on or before that date in the available stored series. Provider history can later be revised, and reprocessing that history can change historical model states. Chronological processing therefore does not establish that the same inputs and signals were available on the original day.
A historical replay is not an immutable record of live assessments. Establishing a forward track record requires preserving the original daily inputs and outcomes separately from later recalculations.
The model uses freely and automatically obtainable market and network-derived data where practical. Missing or stale inputs can block a signal rather than silently inventing certainty.
How to interpret results
Signals are risk information, not instructions to buy or sell. A candidate can be followed by further upside. A confirmation can arrive materially below a top. Historical Bitcoin cycles are a small sample and market structure changes over time, so backtests cannot establish future performance.
The historical evaluation reports time matches and price distance separately, using the highest or lowest daily close inside fixed historical reference windows. Those references include local and cycle extremes chosen with knowledge of history. A time match alone does not establish forecast accuracy, and these results are not a clean out-of-sample test.
Turnmetry does not know your financial position, objectives or risk tolerance. It provides general research, not personal financial advice or automated trading.
Live beta and model changes
The beta exists to build a forward track record. Rules may improve as evidence accumulates, but changes should be versioned and should not rewrite what an earlier model actually signalled. Verified users can choose which signal transitions may trigger an email.
Create a beta account