← Research · Replication · Macro and cross-asset

Does bitcoin follow gold with a lag?

We found no lead. Testing gold’s moves against bitcoin and ether at daily, weekly and monthly lags from August 2017 to September 2026, no lag predicted crypto returns once the number of lags searched is taken into account. The two do move together on the same day, but weakly: the daily correlation is about +0.12, so gold explains roughly 1.5% of bitcoin’s daily variance.

Did not replicateGold (London afternoon benchmark) vs BTC and ETH, daily, Aug 2017 – Sep 2026Last reviewed

The claim

Source we re-testTamuly, B. (2025). When gold price hits new highs, history shows ‘Bitcoin follows’ within 150 days — Analyst. Cointelegraph, 17 April 2025. cointelegraph.com/news/when-gold-price-hits-new-highs-history-shows-bitcoin-follows-within-150-days-analyst

The idea spread widely in 2025: when gold makes new highs, bitcoin follows a few months later. Cointelegraph summarised Joe Consorti of Theya as saying that bitcoin follows gold’s directional bias with a lag of 100 to 150 days, illustrated by charts that overlay the two price lines with one shifted by the lag. It is market commentary rather than a paper, but it is specific enough to test.

What we tested

Gold at the London afternoon benchmark, with BTC and ETH sampled at the same moment, daily from August 2017 to September 2026. We tested gold leading crypto at lags of 1–20 days, 1–26 weeks and 1–12 months; a rule that holds bitcoin while gold’s trend is up; the shifted-overlay chart itself; a long-run equilibrium between the two; and whether chart levels on the BTC/gold ratio hold better than on BTC/USD.

Result

None of the lead-lag combinations survives correction for the number tried, and the reverse direction — bitcoin leading gold — fails too. The best version of the trend rule had a Sharpe ratio of 0.81, against 0.66 ± 0.23 for random rules with the same exposure: it is simply being long bitcoin. The overlay chart is the most persuasive and the least informative. The best lag, 240 days, correlates the price levels at 0.830, versus 0.828 with no shift at all, and pairs of unrelated random walks searched the same way typically reach 0.84.

Gold versus BTC and ETH, August 2017 – September 2026

Claim testedResult
Gold leads BTC or ETH (daily, weekly, monthly lags)No lag survives the search
Bitcoin “catches up” with gold’s trendSharpe 0.81 vs 0.66 ± 0.23 for random rules
Shifted overlay of price levelsBest lag 0.830 vs 0.828 unshifted
Long-run equilibrium (cointegration)Not found
Support and resistance on the BTC/gold ratio48.4% of touches bounce vs 48.9% at random levels
Same-day co-movementCorrelation +0.12 daily, +0.16 to +0.18 weekly

What it means

Gold and bitcoin share a small same-day component, partly through the dollar, but nothing that gives advance notice. The positive sign also means gold has not hedged bitcoin. A lag picked by eye on a price chart is a classic trap: two trending series look alike at almost any shift.

What would change the verdict

The same-day link is not stable — its one-year correlation has ranged from −0.10 to +0.41, and 2026 sits near the top at +0.39. A lead would have to be fixed in advance and hold on data after this date. Shared drivers such as real yields and the dollar are a more promising place to look than gold itself.

This note re-tests a claim made by the cited authors on our own crypto market data; the summary of their method is ours. Tested · published · last reviewed . Research, not investment advice.