Live prices are currently unavailable — the exchange feed could not be reached and no recent cached reading is held.

Bitcoin Data GuideBitcoin Reference

Returns & Performance

How the shape of returns changed across cycles

The magnitude of Bitcoin's advances has fallen and the duration of them has lengthened. This page measures both together, and shows the dispersion of annual returns within each cycle, so the change in profile can be seen rather than asserted.

2011-2025Source: Coinbase Exchange daily candles; Bitstamp and CoinDesk historical series for 2011-2014Advances are bounded by closing-price lows and highs; annual figures are open-to-close.

The profile of each cycle

Each row combines two datasets. The advance, duration and gain-per-day columns come from the cycle-return series, which measures each low-to-high move. The best and worst year columns come from the yearly return series, restricted to the calendar years the cycle spans. Read together, they describe not just how far each cycle travelled but how it travelled — quickly or slowly, in a few violent years or across several moderate ones.

Each Bitcoin cycle with its advance, duration, gain per day, the years it spans and the best and worst calendar year within it.
CycleAdvanceDurationGain per dayYears spannedBest yearWorst year
No records are available for this dataset.

2011-2025Source: Coinbase Exchange daily candles; Bitstamp and CoinDesk historical series for 2011-2014

Gain per day. The gain-per-day column divides the total advance by the number of days it took, which is a crude measure of how fast the move was. It is not an annualised rate and it is not compounded; it is simply the average daily percentage change over the advance. It is included because it makes the trade-off between size and speed visible in a single figure, and it should be read as a comparison between rows rather than as a rate in its own right.

What changed, and what did not

The clearest change is in magnitude. The earliest advances multiplied the price many times over; the later ones multiplied it by a much smaller factor. This is the arithmetic of a growing market, and it is the single most reliable pattern in the table. A larger asset requires more capital to move the same percentage, and the capital available has not grown as fast as the asset has. The consequence is that the spectacular percentage gains of the early years are unlikely to recur at today's scale.

The second change is in duration. The early advances were compressed into short bursts; the later ones stretched across a year or more. The gain-per-day column makes this explicit: the earliest cycles gained far more per day than the later ones, even though the later ones took longer. A slower, smaller advance is what a maturing market produces, and the table shows the transition happening across its rows.

What has not changed is the dispersion within each cycle. Every cycle in the table contains at least one year of very large gains, and the gap between the best and worst year within a cycle is wide in every case. Loss years are not universal: the 2011, 2013 and 2017 cycles each contain a year that finished lower, but the 2021 and 2024-2025 cycles do not. The profile has changed in scale and speed; it has not changed in character. A reader who expects the next cycle to be smoother than the last is expecting something the record does not show.

The sample is five cycles, which is not enough to establish a trend with confidence. The table describes what has happened across a small number of events in a market whose structure changed materially between them. The cycle comparison page places the same cycles side by side on duration, gain and drawdown, and the low-to-high returns page publishes the advances with their dates.

Related reading