Crypto prices swing sharply because little holds them in place: there are no profits or coupons to anchor value, supply does not respond to demand, borrowed money is force-sold when prices drop, and trading is spread across many thin venues that never close. Exchange failures, stablecoin breaks and hacks add sudden jumps. Bitcoin has calmed since its early years, but over the year to October 13, 2026 its volatility was still about 42% annualized, or one standard deviation of about 2.2% a day, with 14 days that moved 5% or more.

Key takeaways

  • Bitcoin and ether pay no cash flows, so prices rest on expectations; BIS data show crypto app users rising about two months after price increases, a loop that extends moves.
  • Bitcoin's code issues 3.125 BTC per block, about 450 a day or roughly 0.8% of supply a year, whatever the price.
  • Some platforms worldwide offer leverage of up to 150:1, including to retail users; on October 10, 2025, forced liquidations, with losses reported at USD 19 billion or more, deepened a half-hour price collapse.
  • Bitcoin's calendar-year volatility fell from 70–93% in 2017–2021 to 42–53% in 2023–2025; ether's was 47% in 2023 but 75% in 2025.
  • For assets that trade every day, one standard deviation of daily moves = annual volatility ÷ √365, so 70% a year means about 3.7% a day; most days move less, and a few move far more.

No cash flows to anchor a price

A share is a claim on profits and a bond on interest; bitcoin and ether are claims on neither. A Federal Reserve staff paper, The Financial Stability Implications of Digital Assets (2022), notes the argument that their fundamental value should therefore be zero, then gives three reasons people pay anyway: they expect others to pay more later, "as in a typical asset bubble"; they expect future cash flows; or they need the asset to transact on its ledger.

"Without an underlying claim on cash flows, assessing a crypto-asset's fundamental value is a challenge." (Federal Reserve FEDS 2022-058)

When a price rests on expectations of future adoption, a change in those expectations moves it with nothing to pull it back. A stock that falls 40% while its earnings hold steady starts to look cheap against those earnings; a coin that falls 40% offers no such yardstick.

Expectations also feed on price. The BIS Bulletin Crypto shocks and retail losses (February 2023) found that while bitcoin rose from $250 to $69,000 between August 2015 and November 2021, daily active crypto app users grew from about 100,000 to more than 30 million, with user growth lagging price rises by about two months. Rising prices draw in buyers, which lifts prices further; the loop also runs in reverse. Scheduled economic data matter less than for other assets: a New York Fed staff report, The Bitcoin–Macro Disconnect (2023), found bitcoin did not respond to monetary and macroeconomic news, unlike other US asset classes.

Supply that cannot respond

Bitcoin's issuance is written into its software. Bitcoin Core's block subsidy function starts at 50 BTC per block and halves the reward every 210,000 blocks. With its consensus parameters targeting one block every 10 minutes, total issuance converges to just under 21 million. Since block 840,000 was mined on April 20, 2024, the reward has been 3.125 BTC: at about 144 blocks a day, roughly 450 new coins a day and 164,000 a year, around 0.8% of the about 20.1 million issued by October 2026, computed from the schedule. The Bitcoin white paper describes new coins as the way to start circulation "since there is no central authority to issue them."

Fixed supply alone does not explain volatility, since a company's share count is also fixed on any given day. The difference shows up over months and years: a company can issue shares when its price is high or buy them back when it is low, and a commodity producer can open or close mines. Bitcoin's schedule ignores price, so every lasting change in demand is absorbed by the price and by holders' willingness to sell.

Ether follows different rules, but has no price-responsive issuer either. Per ethereum.org, new ether goes to stakers in amounts that vary with the total staked, and each transaction's base fee is burned, so net supply depends on staking and network activity rather than on a fixed schedule.

Leverage and liquidation cascades

Much crypto trading uses borrowed money through perpetual futures, contracts that track a coin's price but never expire. A funding rate, a regular payment between the two sides based on the gap between futures and spot prices, keeps them close. A BIS Financial Stability Institute paper, Cryptoasset service providers as financial intermediaries (April 2026), reports leverage at some platforms of up to 150 times, including for retail users. The paper covers platforms worldwide; the leverage open to a US customer depends on the venue and how it is regulated. Platforms usually do not promise to warn before a margin call; when margin falls below a set level, they close the position automatically.

Adverse move that wipes out the deposit ≈ 1 ÷ leverage
leverage = position size ÷ trader's own deposit (margin)

Illustrative example. A trader posts $1,000 and opens a $10,000 bitcoin position, 10 times leverage. A 10% fall costs $1,000, the whole deposit. At 20 times the deposit is gone after a 5% fall, at 100 times after 1%, at 150 times after about 0.67%, and platforms liquidate before it reaches zero. In the year to October 13, 2026, bitcoin's daily reading in FRED fell 5% or more on 4 days, each enough on its own to wipe out a 20-times long position, and prices can swing further within a day than between daily readings.

Each forced sale pushes the price down and triggers the next tier of liquidations. The same chain works upward: closing a liquidated short position means buying, which can push the price higher. A BIS working paper, Crypto carry, found the futures-spot gap can reach 60% a year, driven by trend-chasing smaller investors seeking leveraged upside and scarce capital on the other side, and concludes that this, with the high leverage involved, may help explain why severe crashes are frequent in crypto.

October 10, 2025

Box 4 of the BIS paper describes crypto prices falling sharply over half an hour that Friday evening (Central European Summer Time), amplified by automated liquidations of derivatives. Direct losses from the liquidations were reported at USD 19 billion, a lower bound because exchanges cap how many liquidations they report per second. One large exchange had an outage under "high service load" that kept customers from closing positions. A New York Fed Liberty Street Economics post (June 2026) ties the start to news of a potential additional 100% tariff on Chinese goods, after which liquidity "vanished from key exchanges." In FRED's Coinbase bitcoin series the reading for the sell-off day fell 7.2% (FRED dates it October 9; see the note under the table below), and ether's fell 12.2%.

Fragmented markets that never close

The same coin trades on many exchanges worldwide, each with its own order book. Makarov and Schoar's Trading and arbitrage in cryptocurrency markets (Journal of Financial Economics, 2020) documented large, recurring price gaps between exchanges, much larger across countries than within them, which they linked to capital controls that limit arbitrage, and found that a common component of buying and selling pressure explained 80% of bitcoin's returns. Pressure spreads across venues, but arbitrage capital cannot always move freely enough to close the gaps, so each venue's order book is thinner than the market as a whole.

Crypto also trades around the clock. In US stocks, market-wide circuit breakers halt trading for 15 minutes when the S&P 500 falls 7% or 13% before 3:25 p.m., and for the rest of the day at 20%. No comparable rule pauses all crypto venues at once, so a shock on a weekend night meets whatever liquidity is there.

Concentrated holdings

A blockchain analysis by Makarov and Schoar, NBER Working Paper 29396 (2021), estimated that at the end of 2020 the top 1,000 individual investors controlled about 3 million bitcoins and the top 10,000 more than 5 million, about a quarter of all bitcoins outstanding, probably an understatement. Exchanges and other intermediaries held another 5.5 million. With so much in few hands, a single large holder's decision to buy or sell can be big relative to the coins on offer at any moment.

Event shocks and stock-market links

Single events add jumps. The table shows changes in FRED's Coinbase series around five episodes, with FRED's S&P 500 series over the same dates. Each window also captures whatever else happened those days.

Episode Window (daily readings) Bitcoin Ether S&P 500
Pandemic sell-off Mar 11–12, 2020 −37.4% −43.1% −9.5%
TerraUSD stablecoin depeg May 6–12, 2022 −19.6% −27.2% −4.7%
FTX collapse Nov 7–9, 2022 −23.0% −30.1% −1.5%
Spot bitcoin ETP approval Jan 10–22, 2024 −15.1% −10.3% +1.4%
Tariff news and liquidations Sell-off day, Oct 10, 2025* −7.2% −12.2% −2.7%

*Around this date FRED labels its Coinbase readings one day earlier than Coinbase's own daily candles; the figures cover the sell-off day either way.

Failures, depegs and hacks

The SEC's complaint against FTX's founder alleges years of diverted customer funds and says that, facing billions in withdrawal demands, FTX and affiliates filed for bankruptcy on November 11, 2022. In its Terraform Labs charges, the SEC says UST "depegged from the U.S. dollar" in May 2022 and fell, with its sister tokens, to close to zero. The BIS bulletin puts crypto value lost after the Terra collapse at over $450 billion, and another $200 billion after FTX. Chainalysis counted more than $3.4 billion stolen from January to early December 2025, about $1.5 billion of it in one February attack on the Bybit exchange.

News cuts both ways

The SEC approved spot bitcoin exchange-traded products on January 10, 2024, with the Chair's statement calling bitcoin "primarily a speculative, volatile asset." The widely expected approval was followed by a 15.1% fall in 12 days, then a rise that left bitcoin's March 13, 2024 reading about 57% above approval day. The headline did not say which way the price would go, or when.

Moving with stocks since 2020

An IMF blog, Crypto Prices Move More in Sync With Stocks, Posing New Risks (January 2022), reported the correlation of bitcoin and S&P 500 returns rising from 0.01 in 2017–19 to 0.36 in 2020–21, and the Fed staff paper found crypto "highly susceptible to changes in risk sentiment in traditional financial markets." Weekly FRED returns (Wednesday to Wednesday) show the link persisting: bitcoin's correlation with the S&P 500 was 0.11 in 2017–19, 0.43 in 2020–21, 0.51 in 2022–23 and 0.40 from 2024 to mid-October 2026; ether's was 0.18, 0.50, 0.62 and 0.48.

Has volatility come down?

The table gives annualized volatility by calendar year (the standard deviation of daily log returns from FRED's Coinbase prices, times √365 because crypto trades daily) and counts days with a move of 5% or more either way.

Year Bitcoin volatility Bitcoin days ±5% or more Ether volatility Ether days ±5% or more
2016 50% 25 110% (from May 18) 48 of 225
2017 93% 92 132% 115
2018 84% 74 111% 112
2019 70% 48 83% 66
2020 79% 46 104% 85
2021 81% 76 107% 116
2022 64% 37 87% 71
2023 44% 23 47% 20
2024 53% 27 64% 44
2025 42% 13 75% 62
2026 (to Oct 13) 43% 13 of 286 58% 22 of 286

Bitcoin has clearly calmed: 70–93% a year in 2017–2021, 42–53% since 2023, and days with 5% moves down from 74–92 a year in 2017–2018 to 13 in 2025. Over rolling 365-day windows it hit a low of 40% at the end of May 2026 and stood at 42% on October 13, 2026. Ether has not followed a steady path: after 47% in 2023 it rose to 64% in 2024 and 75% in 2025, and was 58% over the 365 days to October 13, 2026.

Averages hide the extremes. Measured against the standard deviation of the previous 365 days, bitcoin had 59 days since the start of 2017 that moved more than three times the typical amount, about six a year, where a normal bell curve would give about one. Ether had 48, about five a year.

Frequently asked questions

Does a halving push bitcoin's price up?

The code sets issuance, not price. The April 2024 halving cut new supply from about 900 to about 450 bitcoins a day, and any price effect must come through demand, which the schedule does not control. Only four halvings have happened, at blocks 210,000, 420,000, 630,000 and 840,000, too few to separate their effect from everything else going on.

Do spot bitcoin ETFs stop trading when bitcoin crashes?

Not because bitcoin falls. Their shares trade on stock exchanges, so they pause only under stock-market rules: market-wide circuit breakers apply, as do single-security limit up-limit down price bands from 9:30 a.m. to 4:00 p.m. Eastern. Bitcoin itself keeps trading on crypto venues during halts, overnight and on weekends, so fund shares can open well away from the previous close.

Can a stablecoin be volatile too?

Stablecoins are designed to hold a fixed value, usually $1, but the design can fail. TerraUSD fell to close to zero in May 2022, according to the SEC. On October 10, 2025, the synthetic stablecoin USDe briefly traded at $0.65 on one large exchange, according to both the BIS paper and the New York Fed post.

Is volatility the same thing as risk of loss?

Volatility measures how widely returns spread around their average, up and down alike; it does not show how far a price can fall from a peak or for how long. The Fed staff paper noted that by mid-2022 bitcoin was about 70% below its November 2021 peak, after similar declines following peaks in 2017, 2013 and 2011.

Does ether always swing more than bitcoin?

In every calendar year from 2016 through 2026 to date, ether's volatility in the FRED data was higher than bitcoin's, but the gap varies. In 2023 the two were close, at 47% and 44%; in 2025 ether was at 75% against bitcoin's 42%.

Reading volatility as a holder

σ_daily = σ_annual ÷ √365

σ_annual = annualized volatility (standard deviation of returns over a year)
σ_daily  = one standard deviation of daily moves
365      = trading days a year for markets open daily (stocks use about 252)

Illustrative example. At 70% annual volatility, σ_daily = 70% ÷ 19.10 ≈ 3.7%, or about $366 on a $10,000 holding; at bitcoin's recent 42%, about 2.2%, or $220. Volatility also drags on compound returns: a 50% gain followed by a 50% loss turns $10,000 into $7,500.

One σ is a yardstick, not the usual day. In a bell curve about one day in three moves more than one σ; over the 365 days to October 13, 2026, bitcoin did so on 83 days (23%), its median day moved about 1.1%, and 14 days moved 5% or more. That is the fat-tailed pattern: more quiet days and more extreme days than a bell curve predicts.

  • Divide a current trailing volatility figure by √365 to get one standard deviation of daily moves.
  • Multiply by the holding's size to see a one-σ day in dollars, expect most days to be smaller, and allow for several days a year at three times that.
  • For any position with borrowed money, compare 1 ÷ leverage with one σ and with the largest daily falls of the past year.
  • Expect weekend and overnight moves, when stock markets and stock-listed funds are closed.

Sources

Written by

AdvancedFinanceNet Editorial Team

Published Jul 10, 2026

Crypto

Stablecoin Reserves: Assets, Liabilities and Disclosure

A stablecoin issuer is a balance sheet: tokens are debts, reserves are assets, and the gap is the cushion. This guide reads reserve reports that way. It sets out what the GENIUS Act requires (permitted reserve assets, at least 1:1 backing, monthly examined reports, and audited financial statements for the largest issuers that are not already SEC filers), and explains why a point-in-time attestation is not an audit. It then rebuilds Circle's August 2026 USDC report and Tether's June 2026 report from the issuers' own figures, including how each one measures its token liabilities. Worked examples cover the coverage ratio, a frozen bank deposit, market-price moves and a hypothetical redemption run, and a checklist covers reading any reserve report.

Read the full guide
Crypto

How NFTs Record Ownership on a Chain

An NFT is a ledger entry linking a token ID to an owner address, not the artwork itself. This guide shows what the chain records, using a small simulated ledger of three tokens and one transfer. It then covers where the image and metadata usually live, why owning the token does not by itself give you the copyright, why creator royalties are voluntary unless a collection's contract enforces them, and how approvals and key custody decide who can actually move a token. A hypothetical sale with a 5% creator royalty and a 2.5% marketplace fee shows the arithmetic, and the closing checklist lists what to confirm before relying on an NFT record.

Read the full guide
Crypto

The Real Meaning of Decentralization

Three Bitcoin mining pools produced 62.9% of recent blocks, one measure of how decentralized it really is. A network can have thousands of nodes and still depend on a handful of pools, one client codebase, hosted RPC services or an issuer that can freeze tokens. This guide splits decentralization into layers, shows how to compute a Nakamoto coefficient and which threshold to use, and applies it to dated figures for Bitcoin mining pools, Bitcoin nodes and Ethereum staking, where two datasets that group entities differently give different answers. It also covers how the SEC's 2026 interpretation and the EU's MiCA regulation use the word, and ends with a checklist for testing any "decentralized" claim.

Read the full guide

Educational content only. Nothing on this site is personal financial, investment, tax or legal advice.