Q3 2026 Sneaker Market Report

Completed sales on StockX and GOAT, recorded as one average sale price per sneaker per day across all sizes. The lowest new-condition asks on StockX and GOAT are used only to screen out used and damaged pairs (see below). eBay is not used in the Market Index.

This report covers 9,964 sneakers: every sneaker with at least 3 days of recorded sales between the weeks just before the quarter and its close. Retail prices come from release data where available.

A sneaker released later than 45 days before the quarter began is counted here but left out of the index and the movers, because its early prices are a pre-release or launch premium, not a comparable starting price. Where the release date is unknown, it is left out if its first recorded sale came later than 15 days before the quarter began.

For each sneaker we take its price at the end of the quarter and divide it by its price at the start. The index is the weighted average of those ratios, multiplied by 100. Each sneaker is weighted by the square root of its average daily sales times its liquidity score, so heavily traded sneakers count for more.

100 means prices ended the quarter where they started, on average. Above 100 means they rose; below 100 means they fell. So the index minus 100 is the quarter's price move: an index of 97 would mean prices fell 3% over the quarter. Each quarter is measured from its own start, so every change in this report is that quarter's own price move, not a comparison of one quarter's index with another's. Brand indices apply the same calculation to each brand's sneakers.

A start or end price is never a single sale. The end price is the median daily price over the 15 days up to the sneaker's last sale; the start price is the median within 15 days either side of the date one quarter (92 days) earlier.

A sneaker is counted as unchanged (a ratio of 1) when either window is too thin to trust: fewer than 5 sales or 3 trading days, a median implausibly far from the sneaker's own typical price, or no sale in the last 14 days of data. Where asks exist, individual size sales more than 20% below the typical lowest new-condition ask are removed from both windows, since they are usually used or damaged pairs.

Where the retail price is known, a sneaker is not listed as a mover if its start or end price is below half of retail. Prices that far under retail usually reflect used pairs or a mix of cheap sizes rather than the market for new pairs. Such a sneaker still counts toward the index.

Among sneakers with a known retail price, the share whose end price is above retail, and the share at or below it.

This is not the same figure as the 59.5% from our August 2026 study, which looked at 3,538 sneakers released between August 2024 and August 2026 and compared each one's last sale with its retail price. This report covers every tracked sneaker with a known retail price, whatever its release date, and uses its price at the end of the quarter.

Movers are split into grails (end price of $500 and up) and sneakers under $500, then ranked by percentage change. The report shows the top 5 gainers and losers in each tier.

To be listed, a sneaker must pass every start and end price check above, have a confirmed product name, and have new-condition asks at both ends of at least 75% of its start and end prices, so a move built on used or mismatched sales is not published. Kids' sizes (GS, PS and TD) are left out of the movers and the data download, because their prices move on thin volume and don't reflect the adult market.