Thanksgiving week hotel prices, 2026

Priced again October 27, 2026: the same Thanksgiving stay at 100 of the same hotels

On October 27, 2026, 29 days before check-in, we priced 100 of the 225 hotels again for the same Thanksgiving stay (Wednesday, November 25 to Sunday, November 29, 2026): the first 4 in each of the 25 cities that had a Thanksgiving going rate on October 12, 2026, with the same code and the same rules. The figures below compare the two days at the same hotels; the October 12 figures further down are unchanged. These are again one day's prices, and prices change daily.

1. 1. The Thanksgiving going rate on October 27 was, at the median hotel, the same as on October 12

   99 of the 100 hotels had a going rate for the Thanksgiving stay on both days. Hotel by hotel, the going rate on October 27 was, at the median hotel, the same as on October 12 (95% interval -3.5% to 0.0%), a median change of $0 a night. It was higher at 31 hotels, lower at 49 and the same at 19, a change within what chance alone could produce in a sample this size (exact sign test, p = 0.057). The median going rate at these hotels was $160 a night on October 12 and $155 on October 27, before taxes and fees. On the all-in going rate, where the sites reported taxes and fees on both days (99 hotels), the median change was 0.0%. The first day was 44 days before check-in, this one 29.

2. 2. On October 27, 93 of 99 hotels (94%) had a site 5% or more below the going rate

   For the same Thanksgiving stay, 93 of the 99 hotels with a price on October 27 (93.9%) had at least one booking site listing it 5% or more below the going rate. On October 12, at the same hotels, it was 87 of 100 (87.0%). Among the 99 hotels priced on both days, 8 had such a listing only on October 27 and 2 only on October 12: more on October 27, a difference within what chance alone could produce in a sample this size (exact McNemar test, p = 0.109). Where one existed on October 27, the median gap was 20.4%, about $29 a night before taxes and fees. Super.com had the lowest price most often (31 of 93, 33.3%), then Traveluro (16), then KAYAK.com (15).

3. 3. Same room name: a lower listing named the going rate's room at 0 of 93 hotels

   Of the 93 hotels with a listing 5% or more below the going rate on October 27, the lowest listing named a room at 2 (2.2%) and no room at 91, so its price alone does not say which room it buys. A lower listing whose room name matched the room at the going rate (the same room name) turned up at none of them. On October 12, at the same hotels: a room named at 5 of 87, the same room name at 4. The same name is not proof of the same room, rate or cancellation terms; read them on the booking site.

4. 4. Resort and destination fees: what a listed price may leave out

   The going rates and gaps here are before taxes and fees, and the all-in figures carry taxes and fees as each site reports them; a resort or destination fee may be extra, and not every site shows it. Google Hotels' own description of 7 of these 100 hotels mentions a resort, destination or similar property fee (3 in Las Vegas, 2 in Scottsdale and 2 in Honolulu). At all 7 of them, priced on October 27, a site listed the stay 5% or more below the going rate; a lower listing at such a hotel may be lower because it leaves the fee out, and the hotel then charges it at the property. Check the total, with any fee, on the booking page before you book.

Per-hotel data for October 27, 2026 (CC BY 4.0), each hotel's October 12 going rate beside it: thanksgiving-hotel-prices-2026-oct-27.csv.

Priced October 12, 2026: Thanksgiving week against two weeks later

1. 1. The going rate for Thanksgiving week was a median 8.1% lower than two weeks later

   213 hotels had a going rate (the rate most booking sites agreed on) for both stays. Hotel by hotel, the Thanksgiving rate was a median 8.1% lower than the rate for the same weekdays two weeks later (95% interval -12.9% to -4.1%), a median $14 a night less. It was higher at 70 hotels, lower at 128 and the same at 15 (exact sign test, p < 0.001). Across all hotels with a price, the median going rate was $158 a night for Thanksgiving and $167 two weeks later, before taxes and fees. On the all-in going rate, where the sites reported taxes and fees (213 hotels), the median was 7.0% lower. Thanksgiving week is compared here with early December, not with a typical week: a city's own events in either week move its figure.

2. 2. 186 of 220 hotels (85%) had a site 5% or more below the going rate for Thanksgiving

   For the Thanksgiving stay, 186 of the 220 hotels with a price (84.5%) had at least one booking site listing it 5% or more below the going rate. Two weeks later it was 187 of 216 (86.6%). Among the 213 hotels priced for both stays, 17 had such a listing only for Thanksgiving and 22 only two weeks later: fewer at Thanksgiving, a difference within what chance alone could produce in a sample this size (exact test, p = 0.52), so we do not read it as a holiday effect. For comparison, in the study's 17 snapshots up to October 10, 2026 (2-night stays, cities worldwide), the share ranged from 61% to 81%. Both stays here were above that range; they are 4-night stays in large US cities, and one snapshot cannot say which of those made the difference.

3. 3. Where a lower listing existed, the typical gap was about $28 a night

   At the 186 Thanksgiving hotels with a listing 5% or more below the going rate, the median gap was 19.0%, about $28 a night before taxes and fees, or about $112 over the 4 nights. Two weeks later: 18.1%, about $29 a night, at 187 hotels.

4. 4. Super.com had the lowest price most often for Thanksgiving

   Among the 186 Thanksgiving hotels with a listing 5% or more below the going rate, Super.com had the lowest price most often (99 of 186, 53.2%), then Vio.com (15), then KAYAK.com and Traveluro (13 each). Two weeks later, among 187: Super.com (85), then KAYAK.com (20), then Vio.com (19). A site that is lowest at one hotel is often not lowest at the next, so the hotel and dates you want are what to check.

5. 5. At 176 of 186, the lowest listing for Thanksgiving named no room

   Of the 186 Thanksgiving hotels with a listing 5% or more below the going rate, the lowest listing named a room at 10 (5.4%) and no room at 176, so its price alone does not say which room it buys. On the study's original room check, a lower listing whose room name matched the room at the going rate turned up at 7 of the 186. Two weeks later: a room named at 15 of 187, a match at 11.

The same hotels for Christmas and New Year's Eve: Christmas and New Year's Eve hotel prices, 2026.

City by city

The median going rate a night for each stay, before taxes and fees; the median change of the going rate, hotel by hotel, for the hotels priced for both stays (how many in brackets); and how many hotels with a price had a site 5% or more below the going rate. Nine hotels a city: a city's figures are a small sample, so read them as a pointer, not a city average. The median was higher for Thanksgiving in 6 of 25 cities, most in San Diego (+38.8%), Detroit (+24.1%), Philadelphia (+4.6%), and lower in 17, most in San Francisco (-47.2%), Las Vegas (-39.7%), Tampa (-39.1%); the same in 2. Read with care: New Orleans (3 of 9 hotels had a going rate two weeks later; the others had no price on Google Hotels for those dates).

What this shows, and what it does not

It shows what booking sites listed on Google Hotels on one day, October 12, 2026, about six weeks before Thanksgiving, for 225 hotels: 9 in each of 25 US cities. It is not a forecast and not every hotel: prices for these stays will move before November, a sample this size has room for chance (each figure above says how much where it matters), the hotels are US only, and Google Hotels does not list every site. The going rate and the gaps are nightly rates averaged over the 4 nights, before taxes and fees as most sites list them; the hotel pages and the checker show prices all-in where the sites report taxes and fees. Thanksgiving week is compared with the same weekdays two weeks later, which is early December, not a typical week of the year.

The October 27 figures are a second look at 100 of the 225 hotels (4 a city, by a rule fixed before any search: the first in each city with a Thanksgiving going rate on October 12), for the same stay. Two days 15 days apart say how these hotels moved between them, not where prices go next, and not what every hotel did. A resort or destination fee may be extra on any of these prices.

Thanksgiving hotel prices: common questions

Method and data

The same method as the hotel price dispersion study, with the same scoring code as its weekly snapshots. The 25 cities anchor 25 of the largest US hotel markets, each named as the study names its city there (Anaheim for the Anaheim and Santa Ana area, Honolulu for Oahu, Tampa for Tampa and St. Petersburg, Scottsdale for the Phoenix area, where the study has no Phoenix city), and were fixed before any search. In each city we took the first 9 hotels, in Google's order, of that city's newest study snapshot that had a going rate there; that list too was fixed before any search. Each hotel was priced by its Google property token, once for November 25 to 29, 2026 and once for December 9 to 13, 2026, both on October 12, 2026 (UTC), for 2 adults, as a traveler in the US, in US dollars: 450 searches in all, one a hotel a stay. For each hotel and stay we kept every booking site Google Hotels listed. The going rate is the rate the largest number of listings agree on, with ties broken low; a lower listing counts when it is 5% or more below it; the room check asks whether the lowest listing names a room, and whether a lower listing's room name matches the room at the going rate. A hotel with no price for a stay is left out of that stay's figures and kept in the CSV with its price columns empty.

On October 27, 2026 (UTC) we priced 100 of the same hotels again for November 25 to 29, 2026, the first 4 in each city, in Google's order, that had a Thanksgiving going rate on October 12, 2026: 100 searches, one a hotel, with the same scoring code. The change is each hotel's going rate on the second day over its going rate on the first, at the hotels with a going rate on both days; its 95% interval is distribution-free and its test the exact sign test. The property fee count reads the hotel description Google Hotels returned with each price.

Per-hotel data, both stays (CC BY 4.0): thanksgiving-hotel-prices-2026.csv. Same columns as the study's CSVs, plus the all-in going rate where the sites reported it. Each hotel's measured prices from the study, all-in: hotel prices by city.