Airbnb Occupancy Rates by City 2026: 12 Markets, 47% to 70%

New York posts the highest Airbnb occupancy of 12 US markets at 70%, Atlanta the lowest at 47%. The full ranked table, plus why occupancy rose in 10 of 12 markets while ADR fell in 10.

AirDNA forecasts US short-term rental occupancy will average 57.4% in 2026, marginally above the 57.0% pre-pandemic average. Across twelve large US markets as of July 2026 the actual spread runs from 47% in Atlanta to 70% in New York, which is a 23-point gap sitting inside that single national figure.

The more useful pattern in the 2026 data is the direction of travel. Occupancy rose year over year in 10 of those 12 markets while average daily rate fell in 10 of them. Operators are holding their nights by giving up rate.

Key takeaways

• AirDNA forecasts US short-term rental occupancy will average 57.4% in 2026, above the 57.0% pre-pandemic average, with demand and available listings both projected to grow 2.7% and RevPAR up 2.9%.
• New York posts the highest occupancy of the twelve markets at 70%, up 13.2% year over year, on a $250 ADR that fell 7.0% over the same period.
• Charleston, SC produces the highest RevPAR in the set at $260 from 63% occupancy and a $421 ADR, roughly 2.7 times the $95 RevPAR an Atlanta door generates.
• Active listings fell year over year in 11 of the 12 markets, led by Las Vegas at down 16.8% and New Orleans at down 12.5%. Atlanta was the only market that added supply, up 4.5%.
• RevPAR fell year over year in 8 of the 12 markets despite occupancy rising in 10, because ADR declines outpaced the occupancy gains.

Read the definition before you compare yourself to these numbers

AirDNA calculates occupancy as booked nights divided by available nights, not booked nights divided by calendar nights. A door you blocked for six weeks of renovation does not drag the market figure down, and it should not drag yours down either. Operators who measure against all 365 calendar days will always look 8 to 15 points worse than the market number they are comparing against, and will price into a problem that does not exist.

Second definition worth pinning: these are market medians across every listing type in the metro, from studio apartments to five-bedroom houses. A four-bedroom in a market with a 55% median is not competing against that median. It is competing against the four-bedroom subset, which in most leisure markets books fewer nights at a much higher rate.

MagicBNB's Listings table sorts every property by net revenue, occupancy, profit, and margin with health-colored occupancy pills at green above 80%, amber 60 to 80%, and red below 60%, so the comparison you make is property against property on identical available-night math before you ever open a market report.

Airbnb occupancy rates by city: the 2026 table

Every figure below comes from AirDNA's public market data pages, updated 5 August 2026, covering active listings as of July 2026 with year-over-year changes measured July 2025 to July 2026. Same five dimensions per city, same order: occupancy, ADR, RevPAR, active listings, and the year-over-year change in listing count.

• New York, NY: 70% occupancy (+13.2%), $250 ADR (-7.0%), $153 RevPAR (+3.5%), 28,348 active listings, supply down 2.8%.
• Denver, CO: 66% occupancy (+6.4%), $179 ADR (-11.5%), $111 RevPAR (-7.8%), 7,895 active listings, supply down 6.2%.
• Charleston, SC: 63% occupancy (+1.5%), $421 ADR (+9.8%), $260 RevPAR (+6.5%), 3,356 active listings, supply down 9.3%.
• San Diego, CA: 61% occupancy (+4.9%), $331 ADR (-7.8%), $192 RevPAR (-6.3%), 15,961 active listings, supply down 9.1%.
• Chicago, IL: 61% occupancy (+2.8%), $247 ADR (-0.3%), $136 RevPAR (-0.6%), 12,212 active listings, supply down 2.7%.
• Phoenix, AZ: 58% occupancy (+9.0%), $212 ADR (-6.4%), $115 RevPAR (+0.1%), 9,203 active listings, supply down 10.7%.
• Austin, TX: 55% occupancy (+8.9%), $264 ADR (-7.1%), $129 RevPAR (-4.9%), 17,165 active listings, supply down 11.1%.
• Tampa, FL: 55% occupancy (-4.0%), $183 ADR (-3.7%), $97 RevPAR (-10.6%), 8,989 active listings, supply down 1.3%.
• Nashville, TN: 54% occupancy (+1.2%), $355 ADR (-4.8%), $186 RevPAR (-7.7%), 13,820 active listings, supply down 2.6%.
• Las Vegas, NV: 49% occupancy (+1.3%), $271 ADR (-8.1%), $116 RevPAR (-12.6%), 21,697 active listings, supply down 16.8%.
• New Orleans, LA: 48% occupancy (-1.8%), $299 ADR (-4.6%), $140 RevPAR (-4.4%), 9,137 active listings, supply down 12.5%.
• Atlanta, GA: 47% occupancy (+2.3%), $214 ADR (+8.8%), $95 RevPAR (+7.1%), 17,064 active listings, supply up 4.5%.

Sort that list by occupancy and you get one ranking. Sort it by RevPAR and you get a different one. Nashville sits ninth on occupancy at 54% and third on RevPAR at $186, because a $355 ADR does more work than fourteen extra points of occupancy at Tampa prices.

Occupancy rose almost everywhere and RevPAR still fell

Ten of twelve markets gained occupancy year over year. Eight of twelve lost RevPAR. The mechanism is visible in the ADR column: rate fell in ten markets, and in Denver it fell 11.5%, which swamped a 6.4% occupancy gain and produced a 7.8% RevPAR decline.

Las Vegas is the sharpest version. Supply contracted 16.8%, the steepest in the set, occupancy still edged up only 1.3%, ADR dropped 8.1%, and RevPAR fell 12.6%. A market losing one listing in six while rate keeps falling is a demand problem, not a supply problem, and no pricing tool fixes it.

The two exceptions ran the opposite way. Atlanta added 4.5% supply and still posted the second-best RevPAR gain at 7.1%, because ADR rose 8.8%. Charleston lost 9.3% of its listings and pushed ADR up 9.8% for a 6.5% RevPAR gain on the highest rate in the table at $421.

MagicBNB's YoY comparison puts a period-corrected delta pill on every KPI across every view, so a property showing more booked nights than last year but less revenue per available night shows both movements side by side instead of one flattering number in isolation.

If RevPAR and RevPAN are still interchangeable in your head, the distinction and why it matters for multi-property portfolios is covered in our RevPAN explainer.

What the national 57.4% forecast actually tells you

The 2026 forecast describes a flat market with moving parts underneath. AirDNA's July 2026 midyear outlook puts demand growth and listing growth both at 2.7%, which cancels out, and RevPAR growth at 2.9% driven by rate rather than nights. Nightly rate growth accelerated from 0.7% year over year in January to about 3% by spring.

Two numbers in that release deserve more attention than the headline. International inbound demand ran 12% below the previous spring, with Canada down 32% from 2024 levels. If your portfolio sits in a border market, a gateway city, or anywhere that leans on Canadian or Western European guests, the national occupancy forecast is describing a different market than yours.

The other is where growth actually showed up. San Francisco led RevPAR growth at 12.1%, followed by Anaheim at 11.0% and Philadelphia at 10.1%, each in a market where supply tightened. Supply contraction plus stable demand is the only combination in the 2026 data that reliably produces rate power.

Occupancy is a decision input, not a scoreboard

Run the RevPAR column out over a year and the acquisition question answers itself. A Nashville door at $186 RevPAR generates $67,890 of revenue per available night across 365 nights. A Tampa door at $97 generates $35,405. Same operator, same effort, less than half the top line.

Now apply the trend. Nashville RevPAR fell 7.7% year over year, so that same door was on pace for roughly $73,550 twelve months ago. Tampa fell 10.6%, from roughly $39,600. Both markets are shrinking per-door revenue, and an operator holding four Tampa doors lost about $16,800 of annualized top line across the portfolio without a single guest complaining or a single listing dropping in the rankings.

That is the case for measuring RevPAR monthly rather than occupancy monthly. Occupancy can rise for eight straight months while the revenue that pays your mortgage falls, and the calendar will look healthy the entire time.

MagicBNB's Portfolio Overview carries occupancy, ADR, RevPAN, and net payout in one KPI strip with a net payout sparkline and delta against the prior period, on MTD, last 30, last 90, and YTD presets, so the rate-versus-nights tradeoff is one screen rather than three exports.

For benchmarks by market type rather than by named city, beach against ski against urban, see our Airbnb occupancy rate benchmarks guide.

Occupancy can rise for eight straight months while the revenue that pays your mortgage falls.

FAQ

What is the average Airbnb occupancy rate in 2026?

AirDNA forecasts US short-term rental occupancy will average 57.4% in 2026, slightly above the 57.0% pre-pandemic average, with demand and available listings each projected to grow 2.7%. Individual large markets in July 2026 ranged from 47% in Atlanta to 70% in New York.

Which US city has the highest Airbnb occupancy rate?

New York records the highest occupancy of the twelve large markets in this table at 70% as of July 2026, up 13.2% year over year, on a $250 average daily rate. Denver follows at 66% and Charleston at 63%.

Is a 55% occupancy rate good for an Airbnb?

55% is roughly at market median in Austin and Tampa as of July 2026 and well below median in New York at 70%, Denver at 66%, and Charleston at 63%. It depends entirely on the market and on whether you measure against available nights or calendar nights, because the two methods differ by 8 to 15 points for most operators.

Why is my occupancy up but my revenue down?

Because ADR fell faster than occupancy rose, which happened across most of the US in the year to July 2026. ADR declined in 10 of the 12 markets in this table while occupancy rose in 10, and the net effect was RevPAR falling in 8 of them.

Where can I get free Airbnb occupancy data for my city?

AirDNA publishes a free public market page for every US city showing occupancy, ADR, RevPAR, active listing count, and year-over-year changes, at airdna.co/vacation-rental-data. Inside Airbnb publishes raw listing-level scrapes under a Creative Commons BY 4.0 licence for anyone comfortable computing their own distributions.

Benchmark your own doors against these numbers

Pull your last 90 days of booked nights and available nights per property, compute occupancy and RevPAR for each, and put them next to the city row above. Any door more than eight points under its market median on occupancy has a pricing or listing problem worth a week of attention. Any door at or above median occupancy with RevPAR falling year over year is being underpriced, which is the more expensive mistake and the harder one to see. Run both numbers against your portfolio at magicbnb.io.

About MagicBNB

MagicBNB is a portfolio intelligence platform for short-term rental operators running multiple doors. The Property Health Grid puts a margin-derived health dot, weekly occupancy, and MTD revenue on every property card so a failing door surfaces before it costs a month of bookings. Property Detail runs month-by-month year-over-year comparison per property with best-month and worst-month highlights and an expense breakdown by category. Channel mix reporting splits revenue across Airbnb, Vrbo, Booking.com, and direct on every view, including year-over-year, so a market-wide rate decline can be separated from a channel-specific one. See it at magicbnb.io.