Returns and drawdowns explained

Dates and daily closes

Start and end dates are inclusive UTC calendar days. Each day stands for that day's completed close. The current, unfinished day is not used.

A price is required on every day in the window. If a day is missing, the line breaks at the gap and the affected metrics show as unavailable. A missing day is never treated as zero and never filled with the previous price. An asset that was not yet listed is a coverage gap, not a price of zero.

Because only closes are used, a metric can miss moves that happened and reversed within a day.

Price data

The prices come from Coinbase Exchange and Kraken, using the market quoted in US dollars (USD) for each asset. The table below names the source and market of every asset.

• Daily close. The close for a day is the last trade price of that UTC day on the named exchange, in the USD market.
• Completed days only. The current, unfinished UTC day is never used.
• Refreshed daily. The data is refreshed once a day, so the latest close is normally the previous UTC day. Daily closes through 3 Nov 2026 (UTC), source: Coinbase Exchange and Kraken; retrieved 4 Nov 2026.
• Gaps are shown, not filled. If an exchange has no close for a day, the day stays empty. It is not carried forward or estimated, and metrics that need it show as unavailable.
• Suspicious moves are flagged. The build checks the daily series for unusual jumps and flags them.
• Assets left out. An asset is left out of the Lab when a suitable daily history is not available. The reasons are listed below.

The figures on the site currently come from Dataset 2026-11-03.120e407e and method version 1. The dataset label changes each time the daily prices are refreshed; the method version changes only when a formula changes.

Base 100

Let P(t) be the price on day t, P(0) the price on the first day and P(T) the price on the last day.

I(t) = 100 × P(t) / P(0)

Every asset starts at exactly 100 on the first day, so the lines can be compared on one scale. The chart shows the level, not the price.

Period return

R = P(T) / P(0) − 1

It is shown as a signed percentage. It compares the last day with the first day and says nothing about what happened in between.

Drawdown

The high-water mark is the highest price seen from the start of the window up to day t:

H(t) = max(P(s)), for s from the start through t

DD(t) = P(t) / H(t) − 1

Drawdown is always zero or below. It measures how far the price is below its earlier peak inside the selected window. It is not an all-time drawdown, and a different window can give a different answer.

Worst drawdown in this period

min(DD(t))

This is the lowest drawdown value in the window. It is based on daily closes and can miss intraday extremes.

Annualized volatility

The daily log return is r(t) = ln(P(t) / P(t−1)).

Annualized volatility is the sample standard deviation of the daily log returns multiplied by the square root of 365.

At least 30 daily returns are needed, which means 31 prices. With fewer, the table shows volatility as unavailable and gives the reason. The figure describes how much prices varied in the period observed. It is not a probability of future loss.

Worked example

Four daily closes of an imaginary asset: 100, 150, 75 and 120. These are made-up numbers, not market data.

• Period return: 120 / 100 − 1 = +20%.
• Worst drawdown: the lowest value in the drawdown column, −50%, on day 3.
• Volatility: unavailable, because three daily returns are fewer than the 30 required.

Price return and total return

These figures are price return. They are not total return. They leave out staking rewards, airdrops, tax, inflation, trading or other fees, and anything specific to an individual holder.

Survivorship bias

The tool covers a small, curated set of assets that are still listed. Assets that failed or disappeared are not in it. Results for this set should not be read as a picture of the whole market.

What the tool does not calculate

There is no annualized growth rate, Sharpe ratio, correlation, index or historical ranking. Historical figures are for information only and are not financial, investment or legal advice. Past performance does not predict future results.