A Wider View of Bitcoin’s Volatility

~7 min read. Part One measured what institutional money did to Bitcoin's volatility. Several readers asked the obvious next question.

The objection came back more than once, and it is the right one.

Part one found that Bitcoin's volatility fell by about a fifth between 2022 and today, that the spot ETF collapsed its cascade risk, and that sudden drops returned once the options market built out around it.

All of which might be true and still mean nothing. Volatility fell across most of the world over those years. Maybe Bitcoin simply got quieter because everything did.

There is only one way to find out. Run the identical measurements, over the identical windows, on assets that have nothing to do with crypto.

The control group

Same three periods as part one. Same seven measures. Three other assets: gold, the Nasdaq 100, and the S&P 500.

Where things stood in the current period, November 2024 to September 2026:

Source: Bloomberg Finance L.P.; Pandemonium calculations. Periods: Jan 2022 – Jan 2024, Jan – Oct 2024, Nov 2024 – Sep 2026.

What the rows mean

• Overall volatility — annualised realised volatility over the period. How much the asset moved, in the conventional sense.
• Change vs the ETF year — the change in that same annualised volatility between the middle period and the current one. Bitcoin's fell 10%; gold's rose 66%.
• Share of movement from drops — how much of the total movement arrived in sudden discontinuities rather than ordinary trading.
• Extreme-day reading — this is excess kurtosis. Zero is the benchmark: a textbook well-behaved market, where the worst days are only moderately worse than the typical ones. The higher the number, the further the outliers sit above the everyday. Crucially it is scale-free, so it says nothing about how big the moves are. Bitcoin at 6.11 and gold at 1.44 in the pre-ETF years means Bitcoin's worst days towered over its own ordinary days far more than gold's did over gold's — even though gold's ordinary days were a quarter the size in absolute terms.
• Vol of vol — how much the volatility reading itself moves around. High means calm stretches punctuated by chaotic ones.

Three things fall out of this. Two of them I did not expect.

1. Bitcoin is the only one that got calmer

Every other asset in the table became more volatile relative to its own history in the current period. Gold by two thirds. The S&P by nearly a third. The Nasdaq by a fifth.

Bitcoin went the other way, and it is the only one that did.

So the answer to the objection is no. Bitcoin did not get quieter because markets in general got quieter. Markets in general got noisier. Bitcoin moved against them.

2. But the drops came back everywhere, and Bitcoin got off lightest

Part one explained Bitcoin's returning gap risk by pointing at its own options market: dealers hedging their positions around expiry dates.

That explanation now looks too simple for the job. Gold, the Nasdaq and the S&P all became jumpier over exactly the same period — and none of them trade on a crypto derivatives exchange.

Extreme-day reading, across all three periods:

Source: Bloomberg Finance L.P.; Pandemonium calculations.

Read across the middle column and every asset is placid. Read across the right-hand one and every asset has deteriorated.

The size of the deterioration is the point. Bitcoin's extreme-day reading roughly quintupled. The Nasdaq's went up nearly thirteenfold and the S&P's twelvefold.

So, the drops did not come back because of something happening inside crypto. They came back because they came back everywhere.

Part one's options explanation may still hold at the margin. But it cannot be the main cause of a change that hit gold, the Nasdaq and the S&P at the same time, and hit them harder.

3. Bitcoin now has the best-behaved distribution of the four

This is the one I did not see coming.

On every measure of shape — how often extremes arrive, how much of the movement comes in discontinuous jumps — Bitcoin is now the tamest asset in the table, judged against each asset's own volatility trajectory rather than against any absolute scale.

Its extreme-day reading of 3.68 is less than a quarter of the S&P 500's. Its share of movement arriving as drops is the lowest of the four, comfortably below both equity indices.

Set that against where it started. Compare the two extreme-day readings directly and the reversal is stark:

Before the ETF, Bitcoin's worst days stood nearly five times further from its ordinary days than the S&P's did from its own. Today the S&P's tails are roughly four times more pronounced than Bitcoin's. This is a comparison part one never made, because part one had nothing to compare against.

That is a genuine structural change, measured against a control, and it is a stronger result than anything in part one.

What emphatically has not changed

Now look at the top row again.

Bitcoin still moves 45% a year. The S&P moves 16%.

And that ratio has barely shifted through any of this. Bitcoin was 2.9 times as volatile as the S&P before the ETF existed. It is 2.8 times as volatile today.

Every structural improvement of the past four years — the regulated wrapper, the unlevered buyers, the custody, the options complex, the sharp fall in cascade risk within crypto itself — has left the amount Bitcoin moves essentially untouched. (Fall, not disappearance: October 2025 saw $19 billion of forced liquidations, so the mechanism is still live, just less dominant.)

The shape of Bitcoin's return distribution converged on conventional markets. The size of its daily moves did not.

That is a strange pair of facts to hold together. A market can become better structured, better hedged, less prone to disorder, and statistically better behaved than the index it is supposedly graduating towards, while still moving three times as much.

A footnote on gold, because it matters

Bitcoin used to be four times as volatile as gold. It is now less than twice.

That sounds like convergence, and it is — but almost none of it came from Bitcoin settling down. Gold's own volatility rose by two thirds in the current period, and its vol of vol nearly tripled, from 3.4% to 9.7%.

The gap closed because gold got wilder, not because Bitcoin got calmer relative to it.

Anyone reaching for the digital-gold comparison should sit with that for a moment. The two assets are converging, and the direction of travel is not the flattering one.

Where this leaves part one

Part one's conclusion was that Bitcoin's ordinary days got calmer while nothing arrived to catch the dangerous ones. The control group sharpens that rather than overturning it, but it moves the emphasis.

The shape of Bitcoin's volatility has normalised further than almost anyone claims — further, on the tails, than the S&P 500's. Better market structure is the obvious explanation, and the timing fits, but this data shows the change rather than its cause. Some of it may simply be that the shocks of the past two years landed on equities and gold rather than on crypto.

The amount — the absolute level of volatility, still 2.8 times the S&P's — has not moved at all. Four years of better market structure have made no impression on it whatsoever.

Which leaves a question the volatility data cannot answer. If plumbing has made Bitcoin behave more normally without making it move less, something else is holding the amplitude up — something no amount of custody, hedging or regulated access reaches.

That something is the subject of part two: who actually owns this asset, and whether anyone is ever obliged to buy it.

Part two: why nobody is obliged to own Bitcoin, and what that does to the money.