How does crypto volatility compare with other asset classes?

This lesson does one thing: it puts crypto volatility on the same axis as everything else you have studied, using arithmetic you can reproduce. No adjectives, no comparisons to lottery tickets, no defence of the asset class either. Just the number and how to compute it without embarrassing yourself.

The formula, and the trap inside it

Annualised volatility = standard deviation of daily returns × √(periods per year).

The trap is the day count. Equities trade about 252 days a year; crypto trades 365. Using the equity convention on crypto understates the result by a factor of √(365/252) = 1.20 — a 20% error produced entirely by a constant, with no bug in the code and no warning from any library.

The comparison, worked

Rounded, illustrative daily standard deviations, each annualised with its own correct day count:

• Bitcoin, daily σ 3.0% → 3.0 × √365 = 3.0 × 19.10 = ≈ 57%
• A large-cap equity index, daily σ 1.0% → 1.0 × √252 = 1.0 × 15.87 = ≈ 16%
• Gold, daily σ 0.85% → ≈ 13%
• EUR/USD, daily σ 0.45% → ≈ 7%
• A high-beta single stock, daily σ 2.5% → ≈ 40%

So the honest headline is: major crypto assets have historically realised volatility several times that of a broad equity index, and comparable to or above that of a volatile individual stock. Smaller tokens have typically realised more than the majors, not less.

Two qualifications, because averages hide the interesting part. Crypto volatility is itself highly variable — bitcoin's trailing annualised volatility has spent long stretches in a roughly 40–100% band and has been both well above and somewhat below it in different eras. And it has trended lower over its history as the market has grown, which means a figure computed on 2014 data describes a different market from the one trading today.

Tails and drawdowns

Standard deviation assumes the distribution is tidy. It is not. Daily moves of ±10% are a recurring feature of crypto history rather than a generational event, whereas a ±10% day in a major equity index marks a crisis by name.

Drawdowns tell the same story more bluntly. Bitcoin has recorded several peak-to-trough declines exceeding 70% — in 2011, in 2013–15, in 2018 and in 2022 — each followed by long recovery periods. That is a factual property of the historical record, stated without any implication about what comes next.

The comparison hygiene

Three things you must control before any cross-asset number means anything:

1. Calendar alignment. Drop crypto weekends when pairing against equities, or the Saturday–Sunday move is silently attributed to Monday and every correlation you compute is wrong.
2. The daily cut-off. Unit 1, Lesson 1. Different cut-offs give different daily returns and therefore different σ.
3. The day count. 365 for crypto, 252 for equities — applied per asset, not per project.

Volatility is a description of realised behaviour. It is not a forecast, and high volatility is neither a reason to do something nor a reason to avoid it. This course teaches measurement, not action.

In the data

The day count is not an assumption you make; it is a property of the data. Both series over the same year:

Bitcoin has a daily bar for every calendar day, about 365 a year; a US equity fund has one per trading day, about 252. Annualise both with the same constant and you have manufactured the 20% error described above, out of two datasets that are each perfectly correct.

Try it now

1. Before computing anything, count bars on the two charts above: roughly 30 a month on the first, roughly 21 on the second. That is 365 against 252 a year, and it is the reason two correctly computed volatilities can still be incomparable.
2. Measure the same calendar month on each and write the two percentage moves down. A year of daily log returns is too many rows to work by hand, so the standard deviations are below, computed on 28 September 2026 from daily adjusted closes, 26 September 2025 to the latest bar (367 returns for bitcoin, 250 for the fund). Annualise each with its own day count, √365 and √252, and check that bitcoin's figure is several times the fund's.

3. Now deliberately annualise the bitcoin figure with √252 and write down the error you introduced, in percentage points and as a share of the right answer. Then the calendar question, on the same data. Keeping bitcoin's own day-by-day returns and discarding the weekend ones before pairing them with the fund gave a correlation of 0.47. Dropping bitcoin's weekend bars first, so that its Monday return runs from Friday's close exactly as the fund's does, gave 0.49. Write down what each version does with the Saturday and Sunday moves, and which one you would state in a comparison table. Three small decisions, three different numbers, and none of them is visible in the headline.