Bitcoin's volatility has plunged, but extreme price swings are more frequent than in 2018

A three-sigma trading day is one when Bitcoin’s price moves at least three standard deviations from its recent pattern. CoinDesk measures this by comparing the day’s price change with Bitcoin’s 30-day realized volatility. The result can be positive or negative, so it flags unusually large gains and losses.

For example, if Bitcoin’s recent daily volatility is 2%, a move of roughly 6% or more would qualify. The calculation uses three times the recent typical movement, rather than a fixed dollar or percentage threshold. That means the threshold changes as market conditions change.

Three-sigma moves are considered rare in a normal bell-shaped distribution. About 95% of moves fall within two sigma, and 99.7% fall within three. Bitcoin recorded 10 such days in 2026, showing that calmer average trading has not eliminated sudden repricing.

Bitcoin had 10 three-sigma trading days in 2026, according to CoinDesk’s analysis. That exceeds the eight recorded during all of 2018, when Bitcoin lost 73% of its value. The comparison shows that extreme moves can remain frequent even when overall volatility declines.

The gap also appears against other volatile assets. Since 2024, Bitcoin’s volatility has been roughly similar to Nvidia’s, at about 47%. Yet Bitcoin recorded 26 three-sigma days during that period, compared with eight for Nvidia. The S&P 500 recorded 16, while gold recorded 12.

These figures suggest Bitcoin still experiences more outsized moves relative to its recent behavior. Deeper liquidity, institutional participation and stronger risk management may have made ordinary days calmer. They have not removed sudden shocks linked to macro events, leverage and market positioning.

Overall volatility measures the typical size of Bitcoin’s price changes across a period. A three-sigma count measures how often daily moves exceed a threshold based on recent volatility. These measures answer different questions, so they can move in opposite directions.

In 2018, Bitcoin’s annualized volatility was about 84%, and its three-sigma moves averaged roughly 10%. In 2026, annualized volatility fell to about 46%, while three-sigma moves averaged roughly 7%. The individual shocks became smaller, but the recent-volatility benchmark became smaller too. More days could therefore exceed three times that benchmark.

This pattern means Bitcoin is calmer on average but still prone to sudden repricing. The market’s deeper liquidity, ETFs and institutional participation have helped reduce ordinary fluctuations. Macro shocks, leverage and positioning can still produce outsized moves compared with the immediately preceding quiet period.

Value-at-risk, or VaR, estimates how much a portfolio could lose on a bad day. Many VaR models use recent volatility as a central input. When Bitcoin has a prolonged calm period, that input declines, and the model may calculate a smaller potential loss.

For example, Bitcoin’s 30-, 90- and 180-day volatility measures have fallen. A model using those lower figures could suggest that investors increase their Bitcoin exposure. But Bitcoin still recorded 10 three-sigma days in 2026, showing that recent calm did not prevent unusually large moves.

VaR also sets a loss threshold without showing how severe losses may become beyond it. That blind spot matters for Bitcoin because rare shocks can be large. If portfolio targets ignore tail risk, a quieter market can lead to broader allocations and make sudden jumps more damaging.

Tail risk describes the chance of unusually large losses that sit outside an asset’s normal trading pattern. It matters because an investment can appear stable most of the time while still suffering severe damage during rare shocks. Bitcoin’s recurring three-sigma days illustrate this problem.

VaR estimates a loss threshold for a bad day, such as a level that losses are not expected to exceed under the model. But it does not explain the size of losses after that threshold is crossed. Expected Shortfall goes further by calculating how damaging losses are among the worst days.

This difference can change portfolio decisions. Falling Bitcoin volatility may reduce a VaR estimate, but repeated extreme moves warn that the tail remains important. Deribit CEO Luuk Strijers said the industry has been moving toward Expected Shortfall and similar measures because they account for that risk more fully.

Leverage lets traders control larger positions with less capital. That can magnify gains and losses. When a macroeconomic shock moves Bitcoin sharply, leveraged traders may need to reduce positions, creating additional buying or selling pressure. The result can turn an initial price change into a faster repricing.

Options can add another feedback loop. In a quiet market, some investors sell options to collect premiums because expected price swings appear small. If a sudden shock changes prices and volatility, option sellers may need to hedge their exposure by trading Bitcoin or related instruments. Crowded positions can make those hedges move in the same direction.

The article identifies macro shocks and crowded derivatives trades as potential amplifiers. It also notes that deeper liquidity, stronger risk management and institutional participation have helped absorb shocks. Even so, quiet conditions do not guarantee that leverage and positioning will remain harmless.

Volatility describes the size of an asset’s price fluctuations over time. Standard deviation measures how widely observations, such as daily returns, spread around their average. In finance, realized volatility is commonly calculated from those past price changes and often expressed as an annualized percentage.

For Bitcoin, CoinDesk compared each daily move with 30-day realized volatility. A larger standard deviation means daily returns have varied more widely, so a three-standard-deviation move requires a larger price change. When recent volatility falls, that comparison threshold falls as well.

Investors use these measures to estimate how much risk an asset may add to a portfolio. They can inform position sizes, risk limits and models such as VaR. But they mainly describe typical or recent behavior. Bitcoin’s repeated three-sigma days show why investors also need to consider rare moves and tail risk.