Bitcoin's 3-sigma days outnumber 2018 despite calmer volatility
CoinDesk analysis finds Bitcoin logged 10 three-sigma trading days in 2026, more than the eight in 2018, even as annualised volatility fell to about 46% from 84%.
Why it matters: The analysis suggests standard risk models may understate Bitcoin's tail risk, which could lead investors to hold larger positions than the data warrants.
Key points
- Bitcoin recorded 10 three-sigma days in 2026, against eight in all of 2018, when it lost 73% of its value.
- Annualised volatility is about 46% this year versus 84% in 2018, and three-sigma moves averaged roughly 7%, down from about 10%.
- Since 2024 Bitcoin has logged 26 three-sigma days, compared with eight for Nvidia, 16 for the S&P 500 and 12 for gold.
Bitcoin's 3-sigma days are more frequent in 2026 than during the 2018 bear market, even though the cryptocurrency's overall volatility has fallen sharply, according to a CoinDesk analysis.
CoinDesk measured each day's move against Bitcoin's 30-day realised volatility, counting any move of at least three times that amount, up or down, as a three-sigma day. In a normal bell-shaped distribution, 99.7% of moves fall within three sigma, so such days are rare and are used to flag outsized swings.
The persistence of extreme moves challenges investors who size positions using volatility-based risk models. Value-at-risk estimates a portfolio's likely loss on a bad day, and some versions lean heavily on recent price swings, so a calm stretch can make Bitcoin look safer than it is. Luuk Strijers, chief executive of Deribit, told CoinDesk that standard VaR does not properly assess full tail risk, and that the industry has been moving towards expected shortfall.
Market participants attribute the jolts to macro shocks and crowded derivatives positioning. Paradigm's Nicolas Quatravaux said traders short volatility in quiet ranges, so a single headline can trigger an outsized day, while Two Prime's Alexander S. Blume pointed to crowded call overwriting.
Live updates on NewsFeeds:BitcoinDesk editor: Keshav Aggarwal (accountable for this page; this summary was not edited by hand)
How this was produced: an AI model wrote this summary of CoinDesk’s report, which we credit and link above. Read our Editorial and AI policy. Spotted a mistake? See Corrections.
Not investment advice. Read the risk disclaimer.