Bitcoin Priced in Gold: the Hard-Money Cross Rate

Almost every Bitcoin chart is priced in dollars, which means every one of them is measuring two things at once: what Bitcoin is worth, and what a dollar is worth. Pricing Bitcoin in gold removes the second variable. Both sides are then assets whose supply cannot be expanded by decree, and the ratio shows nothing but relative demand for the two.

The record

One bitcoin bought 0.37 ounces of gold in September 2014, and 0.15 ounces at the January 2015 low. Since then, from the chart:

2017-12-16 — 15.5 ounces
2018-12-15 — 2.6 ounces
2021-11-08 — 37.1 ounces
2022-11-09 — 9.3 ounces
2025-10-06 — 31.5 ounces
2026-06-30 — 14.6 ounces

The durable pattern is on the low side: each cycle's trough in the ratio has been well above the last — 2.6, then 9.3, then 14.6 ounces. Stated against gold rather than against a currency whose own supply is a policy choice, that is the strongest version of the hard-money claim available from this data.

The finding most Bitcoin charts hide

Bitcoin's all-time high measured in gold is 40.1 ounces, set on 2024-12-17. That is not either of the recent cycle peaks. It came ten months before Bitcoin's October 2025 dollar high, at a moment when the dollar price was well below its eventual peak.

The reason is that gold has been running hard. On our indexed comparison, with both assets set to 100 in September 2014, gold reached 149 by November 2021 and 324 by October 2025 — it more than doubled in those four years, having gained only 49% over the preceding seven. Bitcoin rose over that same window too, but not as fast in proportion.

So a chart showing Bitcoin making a new dollar high in October 2025 and a chart showing Bitcoin's gold price ten months past its own peak are both correct. They differ only in what they are measured in, which is precisely the point.

Why this is the better test

Comparisons against the money supply or government debt are rhetorically powerful and methodologically soft. They set an asset price against a policy aggregate, index both to 100, and invite you to read causation into two lines that both slope upward. We publish them, with the caveats on the page.

The gold ratio has no such problem. It is one market price divided by another, with no indexing, no offset and no choice of start date driving the result. If the hard-money argument only showed up against a depreciating unit, it would be an argument about the unit. Against gold, it has to stand on its own.

It also supplies a genuine control. Gold is the asset that has actually done the job for several thousand years. "Bitcoin went up a lot in dollars" is a much weaker claim than "Bitcoin gained against the thing that has historically held value", and only the second one is visible here.

What the ratio would have to reach

The ratio also makes one popular argument checkable. Gold's above-ground stock is conventionally estimated at around 215,000 tonnes, which is roughly 6.9 billion troy ounces. Bitcoin's supply is capped at 21 million coins, of which close to 20 million exist today.

For Bitcoin's total value to match gold's, one bitcoin would need to buy roughly 345 ounces — about 6.9 billion ounces divided by 20 million coins. At the December 2024 record of 40.1 ounces, Bitcoin's market value was therefore around 12% of gold's. At the June 2026 reading of 14.6 ounces, closer to 4%.

That is the arithmetic behind every "Bitcoin could still be worth many times more" claim, stated in a form you can check rather than as a price target. It says nothing whatever about whether the ratio should converge — only what convergence would mean.

What the ratio cannot tell you

Neither side is a fixed measuring stick, so a move in the ratio does not identify which asset moved. The fall from 40.1 ounces in December 2024 to 14.6 in June 2026 was driven substantially by gold's own rally, not only by Bitcoin's decline — which you can only see by looking at the two series separately.

The two also trade on different calendars. Gold does not trade at weekends, so the ratio holds Friday's gold price across to Monday and any weekend move in Bitcoin shows up unmatched. Over a long chart this is noise; over a few days it is not.

And relative performance is not risk-adjusted performance. Bitcoin's realized volatility has fallen a long way — from a mean of 110% in 2013 to 41% in 2025 — and still runs several times gold's. A ratio that has gone from 0.37 to double digits says nothing about how it felt to hold through a 76.7% drawdown along the way.

Last updated 2026-08-10.