How crypto volatility is calculated
Volatility measures how widely an asset's returns scatter around their own average. It is calculated as the standard deviation of periodic returns — usually daily — and then scaled up to an annual figure so that assets measured over different windows can be compared. A high number means the price has been moving a lot, in both directions.
Crypto markets trade every day of the year, so daily volatility is annualised by multiplying by the square root of 365 rather than the square root of 252 used for stock markets. That factor is roughly 19.1, which is why a daily standard deviation of even 3% translates into an annualised figure most equity investors would find alarming.
The scaling works because variance adds over independent periods while standard deviation grows with the square root of time. It is an approximation, and a known one: it assumes returns are independent and identically distributed, which crypto returns are not, since volatility clusters and extreme days arrive together. It remains the standard convention because it makes assets, periods and markets comparable on a single scale.
Worked example
Five consecutive daily returns of +3%, -2%, +5%, -4% and +1%. Five observations is far too few for a real estimate — thirty or ninety days is more usual — but it keeps every step visible.
| Input | Value |
|---|
| Daily returns | +3, -2, +5, -4, +1 (%) |
| Mean daily return | 0.60% |
| Daily standard deviation | 3.65% |
| Annualising factor | sqrt(365) = 19.10 |
| Annualised volatility | 69.67% |
| An asset at 1% daily instead | 19.10% annualised |
| An asset at 5% daily instead | 95.52% annualised |
The last two rows are the point of the exercise. Daily moves that feel similar in the moment — 1% versus 5% — are a five-fold difference in annualised risk. Volatility also says nothing about direction: an asset that rose steadily and one that collapsed steadily can post identical volatility figures.
How to use this calculator
- Choose a window and stick to it — 30 days reacts quickly, 90 days is steadier, and mixing the two makes comparisons meaningless.
- Enter closing prices at a consistent time of day; crypto has no official close, so an inconsistent snapshot time adds noise that is not real volatility.
- Let the tool convert prices to percentage returns rather than using raw price differences, which bias the result toward higher-priced assets.
- Read the annualised figure when comparing assets, and the daily figure when sizing a position against a stop.
- Recalculate after any major market event — a volatility figure from a calm quarter tells you very little about a turbulent one.
Frequently asked questions
Does high volatility mean an asset is a bad investment?
No. Volatility measures dispersion, not loss. An asset that tripled in a straight line would show high volatility. What high volatility does mean is that outcomes over any short window are wide and unpredictable, so position sizing and time horizon matter far more than they would for a stable asset.
Why multiply by the square root of 365 and not 365?
Because variance, not standard deviation, scales linearly with time. Annual variance is roughly 365 times daily variance, and standard deviation is the square root of variance — so the daily figure is multiplied by the square root of 365. Multiplying by 365 would overstate annual volatility roughly nineteen-fold.
Should I use 365 or 252 trading days?
Use 365 for crypto, which trades continuously including weekends and holidays. Use 252 for equities, which do not. Applying the equity convention to crypto understates the annualised figure by about 17%. The only rule that matters is applying the same convention to both sides of any comparison.
What does volatility fail to capture?
Quite a lot. It treats upside and downside moves identically, assumes returns are roughly symmetric, and badly understates tail risk — crypto produces extreme single-day moves far more often than a normal distribution predicts. It also ignores liquidity, counterparty and protocol risk entirely.
Is realised volatility the same as implied volatility?
No. Realised volatility, which this calculates, looks backward at prices that already happened. Implied volatility is derived from options prices and reflects what the market currently expects. They frequently disagree, and the gap between them is itself something traders watch.
How much history should I use?
Enough to be statistically meaningful, but not so much that it averages away the current regime. Thirty daily observations is a common floor. Fewer than that and a single outlier day dominates the result; several years and you blend calm and chaotic periods into a number describing neither.
This calculator is an educational tool. Results are estimates based on the inputs you provide and do not constitute financial advice. Verify figures with your bank, broker or a qualified advisor before acting on them.