The shelf is refilling. In 11 of 15 big metros more homes are for sale than in October 2019, yet fewer homes came to market in all 15.

11 of 15 metros list more homes than in October 2019, five of them more than half again; four list fewer.

Realtor.com counts the homes listed for sale in each metro every month, and the Federal Reserve Bank of St. Louis publishes the series. We read the active listing count for 15 large metros and compare October 2026 with October 2019, the last autumn before the pandemic. The same month in both years, so the season cancels out.

In October 2026 five metros list 53.1 % to 58.7 % more homes than in October 2019: Phoenix, Austin, Denver, Nashville and Seattle. Six are above by less than half, from +4.2 % in Atlanta to +39.0 % in Charlotte. Four are still short: Miami, Boston, New York and Chicago, at −10.0 % in Boston, −4.9 % in Miami, −38.0 % in New York and −58.0 % in Chicago.

The national picture hides the split. Across the whole country 1,183,520 homes were listed in October 2026 against 1,208,363 in October 2019, a difference of −2.1 %. The nation looks close to normal while the metros pull in opposite directions: a surplus in Phoenix, Austin, Denver, Nashville and Seattle is offset by a shortage in Miami, Boston, New York and Chicago. Among the 15 together the count is −1.0 % on October 2019.

The low point is worth remembering. In February 2022 the nation listed 346,515 homes, 28.7 % of the October 2019 count. Every one of the 15 metros has since refilled, and the ratio of October 2026 to its own lowest month since 2020 runs from 1.5× in New York to 9.7× in Seattle. Tap a bar below to see the metros in each band.

A count of listings is not a count of homes: a listing leaves the count when it goes pending, sells, is withdrawn or expires, and the series does not say which. It covers single-family and condo/townhome listings and not apartment buildings, and Realtor.com may restate its history. How to read the count

The bands are a way to count, not a forecast. A metro at +4.2 % and one at +39.0 % are in the same band here; the next part asks what put each there.

Fewer homes came to market in all 15 metros, and the median days on market is up in 10 of them.

A shelf can fill in two ways: more homes arrive, or the homes that arrived stay. The new listing count says the first did not happen. In October 2026 fewer homes came to market than in October 2019 in all 15 metros, by −19.2 % at the middle, from −3.0 % in Houston to −33.6 % in Las Vegas. Across the 15 together the drop is −19.4 %; for the nation it is −16.3 %.

The median days on market says the second did. It is higher than in October 2019 in 10 of the 15 metros: 57 to 84 days in Austin, 37 to 63 days in Nashville, 39 to 59 in Phoenix. It is lower in Houston, Miami, Boston, New York and Chicago: 69 to 53 days in New York and 54 to 37 days in Chicago. Four of those five also list fewer homes than in 2019; Houston lists +21.9 % more, and its new listings fell by the least of the 15. The longest wait in October 2026 is in Austin, 84 days, and the shortest in Chicago, 37 days.

Prices tell a second half of the story. The price-reduced count over the active count is the share of listings that cut their asking price. In October 2026 it runs from 21.9 % in New York to 46.3 % in Phoenix. Since October 2019 it rose in 7 of 15 and fell in 8. In the five metros where days on market grew most, Austin, Nashville, Phoenix, Denver and Tampa, it fell in 4 of the five. Where homes stay longer, sellers have not all cut sooner.

Switch the views to see it metro by metro: the count against 2019, against the metro’s own lowest month, new listings, days on market, the change in days, the price-cut share and the last 12 months.

What it means for an underwriter. An exit or a sell-down that assumes a buyer appears at the list price in the time a 2019 market took is a different assumption in Phoenix, where the median days on market is 59, than in Chicago, where it is 37. The review does not say prices fall; it says the clock runs differently, and that the clock should be a number in the model for the metro of the asset. How we read inventory and days

The last 12 months add direction. The count is higher than in October 2025 in 11 of 15 metros, by +29.8 % in Seattle, and lower in Tampa, Dallas, Miami and Chicago, down −9.4 % in Miami. The refilling is not over everywhere.

Every metro fell to a low in 2021 to 2024; eleven of the fifteen are back above October 2019, and the lines are still moving.

The chart shows each metro’s active listing count from July 2016 to October 2026 (124 months), with October 2019 as 100. The lines share a dip: every metro fell through 2020 and 2021 to a low between February 2021 and February 2024. Thirteen of the 15 hit it in 2021 or 2022; New York and Chicago did not reach theirs until January 2024 and February 2024.

After the low the lines part. At its low Austin stood at 18.6 and now stands at 157.0; Phoenix is at 158.7, and each of the 15 is plotted against its own October 2019. Chicago went from 27.6 at its low to 42.0, and New York from 40.9 to 62.0.

Pick a metro to bring its line forward against the grey of the other 14. Each pick also names the metro’s highest month in the 124. For 10 of the 15 it falls in 2025 or 2026; for Atlanta, Miami, Boston, New York and Chicago it came before 2020. The shape follows the count, and what moves it is our reading in the part before. How we read inventory and days

For a lender the date is the point. A loan sized on a sale or a lease-up in a metro that was at 30.1 % of its October 2019 count in 2021 carried one market, and the same loan today carries another with 20,646 homes for sale in Phoenix. The metro’s line, not the national count, is the one to read at the date of the loan.

What to do with this

If you underwrite, lend on or sell homes and apartment buildings: read the stock for sale and the wait in the metro of the asset, and read both at the date of the decision. A national count would hide both Phoenix and Chicago.

Titleman reads the offering memorandum, rent roll, operating statement and collateral of a property deal and returns the issues and the questions for the sponsor. The review can set a deal’s sale, lease-up or exit timing against the listings and days on market of its metro, so the check is a number in the file, not a feeling about the market.