Loading history…
25-signal consensus model · P(bottom) computed daily · backfill from 2017-01-01

Bitcoin Cycle Bottom History (2017–2026)

This chart shows the daily Bitcoin cycle bottom probability score backfilled from January 2017 through today. Each data point represents the composite reading of 25 on-chain and macro signals on that date — the same signals tracked on the live dashboard. The score peaks near confirmed cycle lows and drops to near-zero during bull market tops.

How to Read This Chart

Score spike → 60–100%
Signals clustering near historical bottom conditions. The chart shows these clearly at the 2018, 2020, and 2022 lows.
Score plateau → 40–60%
Mixed on-chain readings — typical of mid-cycle consolidations and bear market bounces where accumulation is reasonable.
Score trough → 0–40%
Euphoric bull market conditions where on-chain data warns against new capital deployment at current prices.

Every Major Bitcoin Cycle Bottom — What the Signals Said

December 2018BTC ~$3,200

MVRV Z-Score: −0.40 · Puell Multiple: 0.31 · Funding Rates: deeply negative · Exchange reserves surging

The end of the 2017–2018 bear market. Bitcoin fell 84% from its ATH of ~$20k. Every major on-chain metric reached extreme undervaluation territory simultaneously.

March 2020BTC ~$4,900

MVRV Z-Score: −0.24 · Hash Ribbon: buy signal · Realized Price: near market price · Fear & Greed: 8 (Extreme Fear)

The COVID-19 crash. Bitcoin dropped 50% in 48 hours on March 12–13. Despite the speed of the drop, on-chain metrics quickly confirmed capitulation — and the recovery was equally fast.

November 2022BTC ~$15,500

MVRV Z-Score: −0.23 · Puell Multiple: 0.44 · NUPL: −0.17 (Capitulation) · STH-SOPR: 0.91

The FTX collapse triggered the final capitulation of the 2021–2022 bear market. Long-term holders were distributing at a loss for the first time since 2018, while exchange reserves peaked — a textbook bottom fingerprint.

Key Signals That Fire at Every Bottom

Across all three confirmed cycle bottoms, a consistent set of on-chain signals reached extreme readings simultaneously. No single metric called every bottom alone — the edge comes from convergence.

Historical scores are computed using the same model weights applied to historical on-chain data. Not financial advice. Past performance does not guarantee future results.