Bitcoin Cycle Extremes
How cheap or expensive Bitcoin is against its own history, graded from four gauges that lined up at past cycle turns. Green bars up are cheap, red bars down are expensive, and ±1 is as far as the gauges have ever gone. Shaded columns mark the days at least two checks called an outright extreme — what past bottoms and tops were made of. History, not advice.
Four gauges that lined up at past cycle turns are each scored from −1 (as cheap as that check has ever called it) through 0 (neutral) to +1 (as expensive as it has ever called it), then averaged and flipped so that green up is cheap and red down is expensive. Beside it the original strict count is kept: cheap yes-or-no checks passed minus expensive ones, −4 to +4, and the shaded columns light when two or more agree.
Positive scores mean cheaper conditions; negative scores mean more expensive conditions against these gauges. A score near +1 or −1 means the gauges are near their cheap or expensive thresholds. For example, −0.4 is an average expensive-side reading of 40% of the scale, not a 40% price forecast. Orange and green stripes mark agreement among strict checks, not confirmed future tops or bottoms.
In words. Mayer: price ÷ its 200-day average, neutral at 1, extremes at 0.8 and 2.4. Puell: today's miner income ÷ its 365-day average, neutral at 1, extremes at 0.5 and 4. Price ÷ its 200-week average, extremes at 1× and 3×, neutral in the middle. Trend percentile: where today's price ranks against its own long-term trend, extremes at 5 and 95, neutral at 50. Each reading is placed between its extremes (in log space for the three ratios, linear for the percentile), clamped to ±1, the four are averaged, and the sign is flipped so cheap is positive. The count is cheap checks passed minus expensive checks passed; a column lights at ±2 or beyond.
For each date, the long-term trend and its percentile use only closes up to that date. Adding later prices does not recalibrate earlier readings; corrections to source data can. Thresholds were chosen using past cycles, so this is not an independently validated trading backtest.
For each date, the long-term trend and its percentile use only closes up to that date. Adding later prices does not recalibrate earlier readings; corrections to source data can. Thresholds were chosen using past cycles, so this is not an independently validated trading backtest.
Our own full BTC price history (btc_price_daily: blockchain.info daily closes, 2010-08 to today, 5800+ points). No external API at request time. + fees for Puell.
Own compute · full BTC price history 2010+ (no external API)