Rolex price index: the pre-owned market, from sold results

This is the WristPrices Rolex price index. It is built from our data export of October 26, 2026 and updated as new sales arrive. The price figures on this page are drawn from our own sold-results database at build time, so they update when the data does; the outside figures quoted further down come from the sources named with them.

The short version: across the 402 published references with a twelve-month figure, the median change is +9.6%. Of the published references with a volatility label, 71 are Stable, 89 Moving and 58 Volatile. That is a description of recent sales, not a forecast.

This is not financial advice. Watches are consumer goods with costs to buy, insure, service and sell, and past price moves say nothing certain about future ones.

What the index is built from

An index is only as honest as its inputs, so here is exactly what goes in.

• Sold results only. Our database holds 108,034 sold results: 87,061 eBay sold listings and 20,973 auction results. Asking prices never count.
• Forum posts are kept apart. We also hold 44,070 forum posts marked sold, going back to 2006. Those figures are the listed price when the post was marked sold, not a confirmed sale price. They are left out of the ranges and the twelve-month figures; only the long-run chart further down uses them, for years before eBay and auction records are deep enough.
• Published references only. We track 793 catalog references. A reference enters the index only when it has enough recent sales for a published range, which 460 do today. Of those, 208 carry a high-confidence range.
• One configuration. Every price is normalised to a pre-owned watch with box and papers, sold privately. Watch-only sales are adjusted by a ratio we measured from our own sales, currently 13% below a full set, and unworn sales by 5% above.

The full rules, including the backtest that sets the confidence levels, are on our methodology page.

How the index is calculated

The index answers one question: if you held a typical basket of pre-owned Rolex references for the past year, how did their market prices move?

1. Per reference. For each published reference we compute a recency-weighted market price. Busy references weight recent sales more heavily, with a short half-life; thin ones use a longer half-life so a single sale cannot swing the figure. The half-lives were chosen by a walk-forward backtest, not by hand.
2. Year-over-year change. Each reference's change compares the median of its sold prices over the last twelve months with the median of the twelve months before. Both windows need at least three sales, or the reference has no figure. This is a plain median of sales, not the recency-weighted market price from step 1, so it is easy to reproduce.
3. Basket. The basket is every published reference that has a year-over-year figure. Within it, we report the median change, so one runaway reference cannot drag the whole index.
4. Breadth. Alongside the median we count how many references rose and how many fell. A median that rises on narrow breadth means something different from one that rises across the board.
5. Volatility. Each reference also gets a label from the spread of its last 90 days of sales: Stable, Moving or Volatile. The share of each label tells you how settled the market is.

Steps 3 to 5 give this edition's headline figures:

References that neither rose nor fell to one decimal place count as flat. Compare the rising and falling counts with the basket size before reading much into the median.

The long-run index

The twelve-month figures describe one year. The chart below chains yearly moves into a single line, with its first year set to 100. It applies the method behind our collection price-history pages to every published reference with a long enough record, so its early years lean on forum posts marked sold and later years on eBay and auction results. Read it as historical context: it feeds no price on this site.

This quarter's bellwethers

These are widely searched references, and together they cover the collections most people search for: Submariner, GMT-Master II, Cosmograph Daytona, Datejust, Explorer and Day-Date.

The table gives each reference's current range and confidence. The next table adds the two moves an index reader cares about: the change over twelve months and the volatility label.

What the bellwethers show

Read the change column against the basket median of +9.6%: a bellwether above it moved faster than the typical reference, one below it slower. The volatility column says how much weight a single sale deserves. A Stable or Moving label means recent sales cluster fairly tightly around the market price; a Volatile label means they are spread widely, so one result tells you less about where the next will land. The yellow gold Rolex Day-Date 40 228238, the only precious-metal reference in this set, is labelled Volatile in this edition.

Bellwethers are only nine references. Across the whole basket, 344 references rose and 56 fell. Thin references move more on fewer sales, which is why the index uses a median rather than an average.

Discontinued pieces move differently

Two discontinued references show why a single index number hides detail.

• The Pepsi GMT-Master II 126710BLRO: +18.1% over 12 months, labelled Stable. Our catalog records it as reported discontinued in 2026; Rolex did not confirm it, per the catalog's source. Bloomberg, reporting the Subdial index in March 2026 in a piece republished by SWI swissinfo.ch, described its secondary price rising amid speculation it had been discontinued.
• The previous steel Daytona, Rolex Cosmograph Daytona 116500LN: +8.2% over 12 months, labelled Moving. It was replaced in 2023.

When a reference leaves the catalog, its supply becomes fixed and its sales can behave differently from the current model. We keep both in the basket because people own and sell both.

How our index compares with others

Other trackers exist, and they measure different things.

• Morgan Stanley and WatchCharts publish a secondary-market report across many brands. Time+Tide's summary of the February 2026 edition said market-wide prices fell in 2023 and 2024 and recovered in 2025, and placed Rolex among the brands still trading above retail.
• The Bloomberg Subdial Watch Index tracks the 50 most-traded watches by value. Fortune's April 2024 coverage reported it far below its 2021 and early 2022 levels at that point.

Our index differs in three ways. It covers Rolex only. It is built from sold results we can show you, reference by reference, with the count of sales behind each one. And it publishes its confidence: each range carries a level whose promise was tested on sales the model had not seen.

What the index cannot tell you

• Your watch is not the basket. Condition, box and papers, service history and dial variant move a single watch more than the market moves in a quarter.
• A year-over-year figure compares two medians. Each window has its own mix of conditions and sets, so small moves are mostly noise.
• Thin references lag. A reference with few sales updates slowly, by design, so a fast market move may show up a quarter later.
• No causes. The index shows what prices did. It does not explain why. Where we mention context, it is from the cited reporting, not from our data.

How to read the next edition

For a single watch, skip the index and go straight to its reference page. If you are thinking about selling, our sell page starts from the same data.