How the numbers are calculated.

All four indices are rebased so that the January 2014 value equals 100 for each series. This allows direct visual comparison of relative performance from a common starting point, regardless of the absolute level of each index.

Raw values are stored in Sanity CMS exactly as published by the source. The rebasing calculation runs in code at render time. If the base date is ever changed, no historical data migration is required.

CAGR is calculated from January 2014 to the most recent data point (currently May 2026, representing 12.33 years).

All figures are price return only. A total-return FTSE 100 investor (dividends reinvested) would have earned materially more than the 3.4% price-return figure shown. Fine wine pays no dividends, so all fine wine returns are price return by definition.

Pearson correlation coefficient between the annual returns of the Liv-ex 1000 and the FTSE 100. Annual returns are calculated from January-to-January data points (12 observations, 2014 to 2026).

The result is approximately 0.12, which falls within the range of 0.10--0.20 cited in academic literature on wine and equity markets. The 95% confidence interval with 12 annual data points is wide (approximately ±0.30), so the figure should be interpreted as evidence of low correlation rather than a precise estimate. Monthly data would narrow the confidence interval.

Correlation calculated from annual returns Jan 2014 to May 2026. Monthly data would refine this figure marginally.

Volatility is the standard deviation of annual (January-to-January) price returns. Because these are already annual returns, no additional annualisation factor is applied.

Gold volatility (18.3%) is heavily influenced by the 54.6% annual return in the 2025-2026 period. The Liv-ex 1000 (9.7%) is notably lower than FTSE 100 (11.4%) over this window, a reflection of the index's smoother cycle even through the 2022-2026 correction. These figures are based on 12 annual observations; more frequent data would improve precision.

Maximum drawdown is the largest peak-to-trough decline observed in the data, measured on January data points. Because we use annual (January) snapshots rather than daily prices, intra-year moves are not captured. The FTSE 100 COVID trough in March 2020 (approximately −34% from the February 2020 peak) does not appear in our annual series.

Benchmark data is updated monthly, on the first working day of each new month. New data points are added to Sanity CMS manually from the authoritative sources listed above. The chart updates within 60 seconds of publication.

The stat cards (correlation, track record, and tax treatment) are reviewed quarterly. If a methodology change is made, it will be noted here with a date.