The HotelBank Editorial Team tracked OTA-listed inventory at 72 business hotels (8,449 rooms) in Naha for stay dates from October 12 to 25, 2026 (14 days), from 45 days before each stay date through the day before. Of 784 property-dates in total, a sell-out was observed at least once by the day before on only 260 (33.2%), while the remaining 524 (66.8%) still had inventory through the day before. Market-wide estimated OCC (based on OTA-listed inventory) averaged 92.1% by the day before, so the picture is of a city that fills almost completely while only about one in three properties actually sells out. What is more, on 49.0% of the 206 property-dates that sold out two or more days before the stay, inventory had returned by the day before. This article looks at when sell-outs occur and how long they last by day of week, period and property size, then sets out the inventory and pricing moves that follow.
Scope: Naha business hotels, N=72 properties and 8,449 rooms (784 property-dates in total, based on listings on major booking sites). Booking curves cover business hotels in the same city, N=78–79 properties and 9,327–9,462 rooms. The price metric in this article is estimated settled ADR (the settled price level estimated from OTA and other sales data, tax-excluded equivalent), and occupancy is an estimate based on OTA-listed inventory. Definitions for both are at the end of the article. Data as of November 5, 2026.
• ADR (average daily rate) = an estimated settled rate (tax-excluded equivalent), calculated by applying a property-type adjustment factor to the lowest plan level each property publishes on OTAs (per-room rate for two guests sharing one room, tax included). Checked against property-level results disclosed by listed hotel REITs (91 properties, most recent three months), the median error is about 7%. These are estimates and differ from each property’s actual settled prices and accounting figures.
• ADR at the area level (prefecture, city, ward, town or village) is the median of the estimated settled rates of the properties covered (the level of a typical property in the area).
• References to listed prices mean the per-room rate for two guests sharing one room (tax included).
• OCC (occupancy rate) = rooms sold as a share of total rooms in the area (an estimate based on OTA sales inventory).
• Source: MetroEngines Research (continuously tracks public OTA prices at about 168,000 properties in Japan, of which about 27,000 are confirmed to be operating)
- — Sell-outs on 33.2% of property-dates: For Naha business hotels on stay dates from October 12 to 25, 2026 (N=72 properties, 8,449 rooms, 784 property-dates in total), a sell-out was observed by the day before on 260 property-dates. The other 66.8% still had inventory through the day before.
- — From 68.5% at 45 days out to 92.1% the day before: Estimated OCC for business hotels in the city (based on OTA-listed inventory, N=78–79 properties, simple average of 14 stay dates) built the most between 30 and 7 days out, at +11.4pt.
- — 53.2% for Friday and Saturday stays, 26.2% for Sunday to Thursday stays: The gap in estimated OCC the day before (93.9% and 91.4%) was just 2.5pt, yet the share of property-dates with an observed sell-out differed by about two times.
- — Inventory returned on 49.0% of early sell-outs: Of 206 property-dates that sold out two or more days before the stay, inventory had returned by the day before on 101. The rate was 35.2% in the normal week (stays of October 12–21) and 70.4% for stays of October 22–25, when flight cancellations piled up.
- — Estimated settled ADR up 13.0% YoY: The confirmed figure for October 2026 was ¥9,400 (N=105). September was also up 11.3%, so sell-out frequency is diverging between properties at a time when rates are running above last year.
The market-wide booking curve: from an average of 68.5% at 45 days out to 92.1% the day before
First, the market-wide booking pace. Taking a simple average of estimated OCC across the 14 stay dates for Naha business hotels (N=78–79 properties, 9,327–9,462 rooms, based on listings on major booking sites), the figure rose from 68.5% at 45 days out to 77.6% at 30 days out, 89.0% at 7 days out and 92.1% the day before. The total gain from 45 days out to the day before was +23.6pt, split into +9.1pt up to 30 days out, +11.4pt from 30 to 7 days out and +3.1pt in the final week. In other words, the bulk of the build came between one month and one week before the stay. Over the same 14 days, the average sell-out rate (the share of properties with no listed inventory visible on OTAs and other channels, estimated) also climbed in steps as the stay date neared: 6.5% at 45 days out, 9.7% at 30 days out, 17.3% at 7 days out and 23.5% the day before.
The starting point, however, differs widely by day of week. Friday and Saturday stays (4 dates) averaged 77.0% at 45 days out and 85.4% at 30 days out, while Sunday to Thursday stays (10 dates) stood at 65.1% and 74.5%, starting roughly 11–12pt lower at the same points. By the day before the gap had narrowed to 2.5pt, at 93.9% for Friday and Saturday stays and 91.4% for Sunday to Thursday stays, so weekday stays caught up over the final month and a half.
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
The chart overlays three stay dates of very different character. October 16 (Fri), the highest date in the period, gained +20.0pt from 78.6% at 45 days out to 98.6% the day before, and its sell-out rate rose from 12.8% to 56.4%. By contrast, October 13 (Tue) started low at 54.5% at 45 days out and gained +32.5pt to reach 87.0% the day before, yet its sell-out rate was still only 10.3% the day before. Even as occupancy levels converge, how widely properties sell out appears to differ greatly by day of week. The reasons estimated OCC and the sell-out rate move differently are set out in Estimated OCC vs Sell-Out Rate: Tokyo Business 77.7% vs Capsule 93.8%.
The curve for October 23 (Fri), meanwhile, behaved differently. Estimated OCC, which had climbed to 96.1% at 8 days out, fell to 90.9% at 3 days out and 86.6% at 2 days out, then recovered to 90.4% the day before. According to news reports, the large and strong Typhoon No. 30 approached Okinawa on the afternoon of October 23, and six airlines canceled a total of 137 flights on the 23rd (projected) and 69 on the 24th. October 22 (Thu) also fell from 96.1% at 7 days out to 89.9% the day before, and October 24 (Sat) from 92.7% at 7 days out to 86.6% at 3 days out, meaning inventory was widely observed coming back at the last minute in the week that overlapped the cancellations. This bears on how to read the “return” of inventory later in this article. That the share of properties seeing inventory return in the final two to three days reaches 3.1 times the normal level on typhoon-approach dates is also confirmed in Typhoon Stays: 48.3% of Okinawa Hotels Get Inventory Back, 3.1x Normal.
| Stay date | N (properties) | 45 days out | 30 days out | 7 days out | Day before | 45 days out to day before | Sell-out rate, day before |
|---|---|---|---|---|---|---|---|
| Oct 12 (Mon, holiday) | 78 | 61.3% | 68.2% | 79.2% | 86.3% | +25.0pt | 10.3% |
| Oct 13 (Tue) | 78 | 54.5% | 66.2% | 80.7% | 87.0% | +32.5pt | 10.3% |
| Oct 14 (Wed) | 78 | 56.1% | 68.1% | 82.6% | 90.0% | +33.9pt | 19.2% |
| Oct 15 (Thu) | 78 | 64.9% | 78.4% | 92.1% | 96.9% | +32.0pt | 30.8% |
| Oct 16 (Fri) | 78 | 78.6% | 87.2% | 95.2% | 98.6% | +20.0pt | 56.4% |
| Oct 17 (Sat) | 79 | 76.0% | 83.3% | 91.2% | 96.2% | +20.2pt | 36.7% |
| Oct 18 (Sun) | 79 | 66.6% | 75.2% | 84.4% | 90.2% | +23.6pt | 11.4% |
| Oct 19 (Mon) | 79 | 70.4% | 78.3% | 88.3% | 94.4% | +24.0pt | 19.0% |
| Oct 20 (Tue) | 79 | 67.4% | 76.3% | 89.7% | 95.9% | +28.5pt | 29.1% |
| Oct 21 (Wed) | 79 | 68.7% | 77.1% | 91.8% | 95.1% | +26.4pt | 24.1% |
| Oct 22 (Thu) | 79 | 72.7% | 81.7% | 96.1% | 89.9% | +17.2pt | 21.5% |
| Oct 23 (Fri) | 79 | 77.9% | 87.0% | 95.6% | 90.4% | +12.5pt | 24.1% |
| Oct 24 (Sat) | 79 | 75.5% | 84.2% | 92.7% | 90.5% | +15.0pt | 27.8% |
| Oct 25 (Sun) | 79 | 68.4% | 75.2% | 85.7% | 88.1% | +19.7pt | 8.9% |
| Simple average of 14 stay dates | 78–79 | 68.5% | 77.6% | 89.0% | 92.1% | +23.6pt | 23.5% |
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
When sell-outs happen: 66.8% of the 784 property-dates kept inventory through the day before
Next, the view shifts from the market as a whole to individual properties. “Sell-out” here means the state (estimated) in which a property’s listed inventory could not be found on OTAs and other channels on a given observation day from 45 days before the stay date through the day before; the stay date itself is not counted. The sample is properties with listed inventory confirmed for at least 30% of their rooms on each stay date (48–65 properties depending on the date). Across the whole period that is 72 properties and 8,449 rooms, and counting one property on one stay date as one “property-date” gives 784 in total.
In aggregate, property-dates with no sell-out observed through the day before were the clear majority at 524 (66.8%). Splitting the 260 property-dates with an observed sell-out (33.2%) by when the first one occurred gives 42 (5.4%) at 31–45 days out, 58 (7.4%) at 15–30 days out, 61 (7.8%) at 8–14 days out, 45 (5.7%) at 2–7 days out and 54 (6.9%) the day before, with no concentration in any one period. Note that sell-outs starting 8 or more days out (161 property-dates, 20.5%) outnumbered those starting within 7 days (99 property-dates, 12.6%).
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
| Category (45 days before stay date to day before) | Property-dates | Share | Of which also sold out the day before | Of which had inventory the day before |
|---|---|---|---|---|
| Listed inventory through the day before (no sell-out observed) | 524 | 66.8% | — | — |
| First sell-out 31–45 days out | 42 | 5.4% | 21 (50.0%) | 21 (50.0%) |
| First sell-out 15–30 days out | 58 | 7.4% | 26 (44.8%) | 32 (55.2%) |
| First sell-out 8–14 days out | 61 | 7.8% | 29 (47.5%) | 32 (52.5%) |
| First sell-out 2–7 days out | 45 | 5.7% | 29 (64.4%) | 16 (35.6%) |
| First sell-out the day before | 54 | 6.9% | 54 (100.0%) | — |
| Total | 784 | 100.0% | 159 | 101 |
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
The key point here is that a sell-out is not necessarily the final state. Only 159 property-dates (20.3% of the 784) were still sold out the day before, and on 101 of the 260 property-dates with an observed sell-out, inventory had returned by the day before. The median number of days on which a sell-out was observed was 5, and on 72 property-dates (27.7%) it was observed on just one day. Among the 45 property-dates whose first sell-out came 2–7 days out, 64.4% remained sold out the day before, against 47.5% for 8–14 days out and 44.8% for 15–30 days out, suggesting that inventory was more likely to move along the way the earlier the sell-out occurred. How widespread post-sell-out restocks are and how the returned rooms are priced is explored using nationwide Obon measurements in 73.9% of Sold-Out Hotel Dates See Rooms Return: Obon 2026 Measured.
Day of week, period and property size: sell-outs observed on 53.2% of Friday and Saturday stays, 26.2% of Sunday to Thursday stays
Sell-out frequency divided first and most sharply by day of week. Of the 201 property-dates for Friday and Saturday stays, a sell-out was observed on 107 (53.2%), and 30.8% were still sold out the day before. Of the 583 property-dates for Sunday to Thursday stays, the figures were 153 (26.2%) and 16.6%, both roughly half the Friday and Saturday level. As noted above, estimated OCC the day before is close at 93.9% for Friday and Saturday stays and 91.4% for Sunday to Thursday stays, yet the spread of sell-outs differs by a factor of two.
The “return” of inventory takes on a clearer character when split by period. Among property-dates whose first sell-out came two or more days before the stay, the share with inventory back by the day before was 35.2% (44 of 125 property-dates) for stays of October 12–21, rising to 70.4% (57 of 81 property-dates) for stays of October 22–25, which overlapped the wave of flight cancellations. The overall 49.0% figure is pushed up by that week, but even in the normal week inventory came back in one of every three cases.
By property size, the share with an observed sell-out was low at 24.7% for properties with 100–199 rooms, against 37.0% for those under 100 rooms and 38.8% for those with 200 rooms or more. For properties with 200 rooms or more, only 20.1% were still sold out the day before, while the rate of inventory returning after a sell-out two or more days out was high at 70.3%. Larger properties, in other words, saw inventory move more readily after selling out.
| Category | Property-dates | Sell-out observed | Also sold out the day before | Share with inventory returned (first sell-out two or more days out) |
|---|---|---|---|---|
| All (14 stay dates) | 784 | 33.2% | 20.3% | 49.0% (101/206) |
| Friday and Saturday stays (4 dates) | 201 | 53.2% | 30.8% | 47.9% (45/94) |
| Sunday to Thursday stays (10 dates) | 583 | 26.2% | 16.6% | 50.0% (56/112) |
| Stays of October 12–21 | 569 | 29.2% | 21.4% | 35.2% (44/125) |
| Stays of October 22–25 | 215 | 43.7% | 17.2% | 70.4% (57/81) |
| Under 100 rooms (35 properties, 1,878 rooms) | 378 | 37.0% | 23.5% | 44.7% (51/114) |
| 100–199 rooms (24 properties, 3,325 rooms) | 267 | 24.7% | 15.7% | 43.6% (24/55) |
| 200 rooms or more (13 properties, 3,246 rooms) | 139 | 38.8% | 20.1% | 70.3% (26/37) |
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
Converting to frequency per property makes the differences in type even easier to see. Of the 58 properties with seven or more covered dates (6,669 rooms, 737 property-dates in total), 9 (15.5%) never had a sell-out observed, 17 (29.3%) each had one observed on under 25% and on 25% to under 50% of covered dates, and 15 (25.9%) had one observed on 50% or more of dates. Even among business hotels in the same city, the range runs from properties that almost never sell out to those that do so on more than half of dates.
| Frequency of observed sell-outs per property | Properties | Share |
|---|---|---|
| No sell-out observed | 9 | 15.5% |
| Observed on under 25% of covered dates | 17 | 29.3% |
| Observed on 25% to under 50% | 17 | 29.3% |
| Observed on 50% or more | 15 | 25.9% |
| Total | 58 | 100.0% |
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
Price levels: October estimated settled ADR at ¥9,400, up 13.0% YoY
Inventory movement is also worth checking from the rate side. Estimated settled ADR for Naha business hotels had a confirmed figure of ¥9,400 (N=105) for October 2026, up 13.0% from ¥8,300 (N=106) a year earlier. September, at ¥8,900 (N=105), also beat the prior-year ¥8,000 (N=105) by 11.3%, so both autumn months landed a little over 10% above last year. Among confirmed months in 2026, February was the highest at ¥13,600 (N=105).
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
Looking ahead, November stands at ¥10,300 (N=104) and December at ¥10,200 (N=104), both estimates based on current sales. These figures can change with future sales, and a simple comparison with the prior-year confirmed figure (¥8,900 for November, N=105) should wait until the month closes. That sell-out frequency differs so widely between properties while rates run above last year gives operators a reason to review how they combine inventory release and pricing.
For revenue managers running business hotels in Naha: implications and an action plan
1. Sell-outs are the minority, and how the last few percent are sold makes the difference. Even with estimated OCC averaging 92.1% the day before, 66.8% of the 784 property-dates kept inventory through the day before. Having your own inventory left until the day before is the market norm in itself, and there is room to be creative about the terms on which the remainder sells. The final week, when the sell-out rate rises from an average of 17.3% at 7 days out to 23.5% the day before, is worth treating as the window in which nearby options are dwindling.
2. Design Friday and Saturday stays and Sunday to Thursday stays as separate products. Friday and Saturday stays saw sell-outs on 53.2% of property-dates, with estimated OCC already high at 77.0% at 45 days out. Sunday to Thursday stays stood at just 26.2%, with a long build from 65.1% to 91.4% the day before. Rather than measuring your booking pace with a single yardstick, separating “dates that fill early” from “dates that move until the last minute” by day of week makes it easier to structure sales allotments and rate tiers.
3. An early sell-out is not the final state. Even in the normal week, inventory had returned by the day before on 35.2% of property-dates that sold out two or more days out. Operating on the assumption that inventory will come back through cancellations and the like, and deciding in advance the terms on which returned rooms go on sale, can be said to be effective. At properties with 200 rooms or more the return rate was especially high at 70.3%, offering more chances to resell.
4. In weeks of weather risk, returns double. For stays of October 22–25, which overlapped the wave of flight cancellations, the share with inventory returned rose to 70.4%, and estimated OCC for October 23 fell from 96.1% at 8 days out to 86.6% at 2 days out. Naha is a location where demand tends to depend on flight operations, so it is worth preparing in normal times for how to resell inventory that returns at the last minute before the day before.
Action plan by timing
| Timing | Action | Decision trigger (figures in this article) | Aim |
|---|---|---|---|
| 45 days out | Compare your booking progress for the stay date against the market booking curve | Market estimated OCC averages 68.5% at 45 days out and 77.0% for Friday and Saturday stays. Applies if your Friday and Saturday stays are already close to selling out at this point | Sort dates to sell out early from dates to hold back by day of week, and look for room to rethink how allotments are released |
| 30 days out | Review sales plans separately for Friday and Saturday stays and Sunday to Thursday stays | Market levels at 30 days out are 85.4% for Friday and Saturday stays and 74.5% for Sunday to Thursday stays. Applies if your own progress is flatter than this gap | Consider whether differences in sell-through pace by day of week can be reflected in rates and plan mix |
| 30–8 days out | Log stay dates showing as sold out, and decide in advance the resale terms for when inventory returns | Even in the normal week (stays of October 12–21), inventory had returned by the day before on 35.2% of property-dates that sold out two or more days out | Reselling returned inventory on pre-set terms, rather than dumping it in a hurry, can be said to be effective |
| 7 days out to day before | Revisit the sales terms for final inventory in line with demand up to the day before | The market sell-out rate rises from an average of 17.3% at 7 days out to 23.5% the day before. Applies if you still have inventory at 7 days out | There is room to capture last-minute demand as sell-outs increase nearby |
| When weather risk arises | Confirm in advance the resale procedure and communication lines for inventory returned through cancellations | For the October 23 stay, estimated OCC fell from 96.1% at 8 days out to 86.6% at 2 days out, then recovered to 90.4% the day before | Be ready to rebuild bookings even in the short sales window after cancellations |
| Following-month review | Check your own share of sold-out dates and monthly rate against the market yardsticks | Observed sell-out shares are 33.2% overall, 53.2% for Friday and Saturday stays and 26.2% for Sunday to Thursday stays; October estimated settled ADR is ¥9,400 (N=105, confirmed) | Build the sales plan for the same period next time from both sell-out frequency and rates |
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
Every one of these actions starts with keeping records of your own booking progress and sell-outs by stay date. Market figures are a snapshot at the time of collection and do not promise results, but they are useful as a yardstick for putting your own performance in context.
Summary: three yardsticks for measuring sell-through in Naha
To sum up, sell-through at Naha business hotels in the second half of October can be captured with three yardsticks. The first is frequency: sell-outs were observed before the stay date on 33.2% of the 784 property-dates, at 53.2% for Friday and Saturday stays and 26.2% for Sunday to Thursday stays. The second is timing: first sell-outs were spread from 31–45 days out through the day before, with those starting 8 or more days out at 20.5% and those starting within 7 days at 12.6%. The third is persistence: among sell-outs two or more days out, the share with inventory back by the day before was 35.2% in the normal week and reached 70.4% in the week of flight cancellations.
Even in an environment where market estimated OCC built from 68.5% at 45 days out to 92.1% the day before and October estimated settled ADR was confirmed up 13.0% YoY, the way sell-outs occurred varied widely by property. Applying these three yardsticks to when, on which days of the week and for how long your own property sells out should offer clues for considering how to release inventory and when to revise rates.
Frequently asked questions
Q. How many business hotels in Naha sell out by the day before the stay date?
A. Of 784 property-dates for stays of October 12–25, 2026 (N=72 properties, 8,449 rooms), a sell-out was observed at least once between 45 days out and the day before on 260 (33.2%). Only 159 property-dates (20.3%) were still sold out the day before, and two-thirds of property-dates kept inventory through the day before.
Q. Why do so few properties sell out when estimated OCC is above 90%?
A. Estimated OCC measures how far listed inventory has been absorbed across the whole market, while a sell-out is an observation of zero inventory at an individual property. Estimated OCC the day before averaged 92.1% across the 14 stay dates, but the remaining inventory was spread thinly across many properties, and only some properties reached zero. The difference between the two is also covered in the related article cited in the body text.
Q. How much inventory returns after a sell-out is observed?
A. Of 206 property-dates whose first sell-out came two or more days before the stay, listed inventory had returned by the day before on 101 (49.0%). The rate was 35.2% for stays of October 12–21, the normal week, and 70.4% for stays of October 22–25, which overlapped the wave of flight cancellations as the typhoon approached. It was high at 70.3% for properties with 200 rooms or more.
Q. How do sell-outs differ between Friday and Saturday stays and weekday stays?
A. Of 201 property-dates for Friday and Saturday stays, a sell-out was observed on 53.2%, and 30.8% were still sold out the day before. For the 583 property-dates for Sunday to Thursday stays the figures were 26.2% and 16.6%, roughly half the Friday and Saturday level. Weekday stays started from a lower estimated OCC (based on OTA-listed inventory) at 45 days out and built up until the last minute.
Q. Is a “sell-out” in this article the same as actually being fully booked?
A. No. A sell-out here means the state (estimated) in which a property’s listed inventory could not be found on OTAs and other channels on a given observation day. Rooms sold through channels other than OTAs and temporary suspensions of listings can also be observed as zero inventory. Observations on the stay date itself are not counted.
About the data
- Estimated OCC (based on OTA-listed inventory): OTA-listed-inventory occupancy = 100 − 100 × OTA-listed rooms remaining ÷ total rooms. The sample is a fixed cohort (the same set of properties for each stay date) of properties listing at least 30% of their total rooms on OTAs. It is an estimate based on the absorption of inventory sold on OTAs and is defined differently from actual room occupancy (it reads high because unlisted rooms are counted as absorbed). Coverage is Naha business hotels, N=78–79 properties and 9,327–9,462 rooms (based on listings on major booking sites), for stay dates October 12–25, 2026.
- Booking curve: based on observations from 45 days before the stay date through the most recent observation.
- Sell-out and sell-out rate: a sell-out (estimated) is the state in which listed inventory could not be found on OTAs and other channels on an observation day from 45 days before the stay date through the day before. The sell-out rate is the share of properties with no listed inventory visible on OTAs and other channels (estimated). The stay date itself is not counted. “Inventory returned” means listed inventory was confirmed at the last observation before the stay date (the day before; in some cases two days before) after a sell-out had been observed.
- Estimated settled ADR: the settled price level (tax-excluded equivalent) estimated from OTA and other sales data (lowest plan level × property-type factor, an ensemble across multiple channels). Past months are confirmed figures; current and future months are estimates based on current sales. The median error against published operating results is 6.6%. Coverage is Naha business hotels, N=103–109 properties (varies by month).
- Sample for the sell-out distribution: Naha business hotels with listed inventory confirmed for at least 30% of their rooms on each stay date. Across the whole period, N=72 properties, 8,449 rooms and 784 property-dates in total (48–65 properties per stay date). By size: under 100 rooms, 35 properties and 1,878 rooms; 100–199 rooms, 24 properties and 3,325 rooms; 200 rooms or more, 13 properties and 3,246 rooms. Frequency by property was calculated for the 58 properties and 6,669 rooms with seven or more covered dates.
- Data as of November 5, 2026. Sales and inventory change daily, so the figures in this article are a snapshot at the time of collection.
References and sources
■ Data sources
OTA public inventory and price data continuously collected by MetroEngines Research. The sell-out analysis covers Naha business hotels with listed inventory confirmed for at least 30% of their rooms on each stay date (across the whole period, N=72 properties, 8,449 rooms, 784 property-dates in total); booking curves use a fixed cohort of business hotels in the same city (N=78–79 properties, 9,327–9,462 rooms); estimated settled ADR uses monthly figures for business hotels in the same city (N=103–109 properties, January 2025 to December 2026). Typhoon No. 30 and flight cancellation figures are from news reports (below). Data as of November 5, 2026.
■ Calculation assumptions
Stay dates covered are the 14 days from October 12 to 25, 2026. A sell-out is the state in which listed inventory could not be found on an observation day from 45 days before the stay date through the day before, excluding the stay date itself. “Inventory returned” means listed inventory was confirmed at the last observation after a sell-out was observed (the day before; in some cases two days before). Estimated OCC and the sell-out rate are simple averages of the 14 stay dates, and YoY figures are calculated only for months confirmed in both years. No projections of future outcomes or scenarios are included.
■ Limitations and caveats
All figures are estimates based on OTA-listed inventory and are defined differently from actual room occupancy or being fully booked. Estimated OCC reads high because unlisted rooms are counted as absorbed. Coverage is limited to two weeks of stay dates, and the last four days overlap flight cancellations caused by the approaching typhoon, so the results may not carry over directly to other seasons or periods. Estimated settled ADR for November and December 2026 is based on current sales and may change.
- Aviation Wire, “Typhoon No. 30: 69 flights canceled on the 24th, affecting 8,800 passengers across six airlines including JAL and ANA” (October 24, 2026; in Japanese)
- KAB Kumamoto Asahi Broadcasting, “Typhoon No. 30 brings violent winds to Okinawa: injuries, fallen trees and power outages, with a string of JAL and ANA cancellations” (October 24, 2026; in Japanese)
Related reading
- Typhoon Stays: 48.3% of Okinawa Hotels Get Inventory Back, 3.1x Normal
- 73.9% of Sold-Out Hotel Dates See Rooms Return: Obon 2026 Measured
- Estimated OCC vs Sell-Out Rate: Tokyo Business 77.7% vs Capsule 93.8%
- Naha Tug-of-War Weekend 2026: Okinawa Business Hotels 97.7% to 84.8%
- Okinawa Occupancy 2026: 61.8% Official vs 89.3% OTA, Why a 27.5pt Gap
- Okinawa 3 Hotel Segments: 2.29x ADR Gap, Oct Long-Weekend Curves
- Okinawa Oct 3-Day Weekend: Final Night Resort 82.8%, Business 75.9%
- Read Booking Curves at 3 Points: T-45 Under 80% Adds +14.3pt