Source-labelled market dataSource, observation time and freshness are shown with every quoteData statusAll pages
GoldObserve

HISTORICAL

Historical Gold Prices

Explore source-labelled historical gold prices by currency and timeframe, with return, volatility and drawdown calculations.

HISTORICAL GOLD DATA

Gold price history with the assumptions left visible

Select a currency and period to inspect the returned daily series. GoldObserve calculates the start value, latest value, period high and low, total return, annualized return, annualized volatility and maximum drawdown from those observations. It does not splice in invented prices when the source is unavailable.

HISTORICAL PERFORMANCE ANALYZER

Measure return, volatility and drawdown

Historical data is temporarily unavailable.No return or risk metric is estimated without a valid series.

CHOOSING A TIMEFRAME

What each historical view can—and cannot—tell you

1 monthRecent direction

Useful for reviewing a current move, but easily dominated by one event or currency swing.

3–6 monthsIntermediate trend

Shows whether a move persisted through several market cycles without becoming “long term.”

Year to dateCalendar context

Easy to communicate, but the January starting point has no special economic meaning.

1 yearFull-year comparison

Includes seasonal and policy cycles, while still depending heavily on the selected endpoints.

5 yearsCycle-level view

Better for studying compounding and drawdowns, but still too short for claims about all monetary regimes.

Since 1971Modern free-price era

Requires a licensed or appropriately sourced long-run benchmark series beyond this tool’s current five-year feed.

INTERPRETATION

Why historical gold returns change with the viewer

Currency can change the answer

Gold is commonly discussed in US dollars per troy ounce, but a non-US holder experiences both the gold move and the exchange-rate translation. A flat USD gold price can still produce a gain or loss in another currency. Compare currencies separately instead of treating one chart as universal.

Price return is not investor return

The series measures the change in a gold reference price. It does not deduct a coin or bar premium, custody, insurance, tax, fund expense, trading fee, financing cost or dealer spread. Those costs can be material, especially over short holding periods.

Endpoints can flatter or punish a result

A return is always conditional on a start date and end date. Review the full path, maximum drawdown and several horizons before concluding that one selected percentage describes the asset.

DOWNLOAD & REPRODUCE

A reproducible historical-data workflow

01Fix the basis

Select currency, troy-ounce unit and date range before exporting or comparing.

02Record the source

Keep the provider and observation timestamp beside the values.

03Check gaps

Count observations and inspect missing dates before calculating risk statistics.

04State exclusions

Separate benchmark return from the costs of the product actually owned.

RELATED RESEARCH

Continue the analysis

METHODOLOGY & PRIMARY SOURCES

How this page is built

The interactive series uses the GoldObserve history endpoint backed by the reviewed local NBU daily accounting baseline. Non-USD daily views use ECB reference rates delivered by Frankfurter with providers=ECB and matched to each UTC date, with a prior published rate carried for no more than seven days when weekends or holidays have no new FX observation. For benchmark context, the LBMA explains its daily auction prices, while the World Gold Council data hub describes longer-run gold price, return and volatility datasets. Those external datasets are not silently substituted into this tool.

HISTORICAL GOLD FAQ

Frequently asked questions

How far back does this chart go?

The current interactive feed supports up to five years. Longer historical claims require a separately sourced and licensed series.

Are the displayed points intraday prices?

No. The history tool uses returned daily observations. The latest live quote can differ from the latest historical close.

Does annualized return mean the same return happened every year?

No. It is the constant compounded rate that connects the selected start and end values. The actual path can be much more volatile.

Does maximum drawdown predict the next loss?

No. It reports the deepest peak-to-trough decline inside the selected sample. Future losses can be smaller or larger.

Can I compare two currencies directly?

Yes, but run each currency separately and keep the same dates. The difference reflects currency translation as well as gold.