Slice Tottori’s post-Obon Saturdays at fixed checkpoints — 45, 30 and 14 days before the stay date, plus the latest reading — and the question of when a date fills splits sharply by property type. For Saturday, August 22, which has already passed its stay date, business hotels moved from 83.0% at T-45 to 93.8% the day before (+10.8pt), while ryokan went from 68.1% to 90.5% (+22.4pt). It is not just the size of the pickup that differs but which stretch it lands in, and there the two are opposites. Business hotels concentrated +6.5pt into the two weeks from T-30 to T-14 and then moved only +1.4pt after T-14. Ryokan did the reverse, adding +10.1pt after T-14 with the largest single shift inside the final 48 hours. For the two Saturdays in early September (September 5 and September 12), business-hotel estimated OCC at T-45 was 18.7pt apart — far too wide for the two dates to be managed as one “September Saturday”.
Scope: accommodation properties in Tottori, N=190 (187–190 depending on the stay date; of which business hotels 45–47, ryokan 71–73, resort hotels 10). Occupancy in this article is an estimate based on OTA-listed inventory (estimated OCC); price levels are out of scope. Definitions are at the end of the article. Data as of August 27, 2026.
• OCC (occupancy) = sold rooms as a share of total rooms in the area, estimated from inventory offered on OTAs. This article labels it “estimated OCC” and calculates it as 100 − 100 × OTA-listed remaining rooms ÷ total rooms. It is defined differently from actual room occupancy and reads higher.
• Sold-out property rate (estimated) = the share of properties for which no listed inventory can be confirmed on OTAs for that stay date (an estimate).
• T-45 / T-30 / T-14 = the observation checkpoints at 45, 30 and 14 days before the stay date.
• This article is built solely on the occupancy (estimated OCC) axis and does not cover any price-level metric.
• Source: MetroEngines Research (continuous tracking of OTA-published rates and inventory for roughly 168,000 properties in Japan)
- — Business hotels settle at T-14. On August 22, a date that has now passed, they had reached 92.4% by 14 days out, and the pickup from there to the day before was just +1.4pt (N=47).
- — Ryokan add roughly +10pt in the final two weeks. August 22 gained +10.1pt from T-14 to the day before, and August 29 gained +10.0pt over the same stretch. The same shape repeats across both dates (N=71–72).
- — Two Saturdays in the same month, 18.7pt apart at T-45. For business hotels, T-45 was 85.8% on September 5 and 67.1% on September 12. A uniform month-level inventory rule rounds that difference away.
- — The absorption rate on remaining inventory gives you a landing range. Applying the measured quartiles from the 61 stay dates already completed puts all properties on September 12 at 86.3–88.8% and September 5 at 92.1–93.6% (see the landing-scenario table below).
- — Demand is leaning toward the five-day holiday that starts September 19. With 24 days remaining, estimated OCC across all properties stands at 97.5% and the sold-out property rate (estimated) for business hotels has reached 95.7%.
Post-Obon into early September: what the Saturdays look like at T-45, T-30 and the latest reading
Start with the overall picture. The table below lines up estimated OCC (based on OTA-listed inventory) for four Saturdays in Tottori at 45, 30 and 14 days before the stay date and at the latest reading. Alongside it sits the share of properties with no confirmable listed inventory at the latest reading, referred to below as the “sold-out property rate (estimated)”. August 22 has already passed its stay date, so its curve is observable all the way from T-45 to the day before. August 29 is a snapshot with 3 days remaining, September 5 with 10 days, and September 12 with 17 days.
| Stay date | Property type | N | T-45 | T-30 | T-14 | Latest | Sold-out property rate (est.) |
|---|---|---|---|---|---|---|---|
| 8/22 (Sat) Passed | Business hotel | 47 | 83.0% | 85.9% | 92.4% | 93.8% day before | 51.1% |
| Ryokan | 71 | 68.1% | 74.8% | 80.4% | 90.5% day before | 57.7% | |
| All properties | 187 | 80.7% | 84.4% | 89.9% | 93.3% day before | 64.2% | |
| 8/29 (Sat) 3 days out | Business hotel | 47 | 69.7% | 73.4% | 78.2% | 84.9% | 23.4% |
| Ryokan | 72 | 59.6% | 62.7% | 64.5% | 74.5% | 25.0% | |
| All properties | 190 | 69.1% | 72.7% | 76.2% | 83.0% | 30.0% | |
| 9/5 (Sat) 10 days out | Business hotel | 45 | 85.8% | 89.0% | 93.5% | 94.3% | 64.4% |
| Ryokan | 73 | 66.7% | 65.3% | 73.3% | 75.7% | 28.8% | |
| All properties | 189 | 80.3% | 82.2% | 86.8% | 88.2% | 39.7% | |
| 9/12 (Sat) 17 days out | Business hotel | 46 | 67.1% | 73.7% | — | 78.0% | 19.6% |
| Ryokan | 73 | 63.4% | 64.9% | — | 73.8% | 17.8% | |
| All properties | 190 | 67.9% | 72.3% | — | 77.7% | 21.6% |
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
What stands out is the gap between September 5 and September 12. Both are Saturdays in the first half of September, yet at T-45 business hotels stood at 85.8% on September 5 against 67.1% on September 12 — a 18.7pt spread. September 5 had already reached a sold-out property rate (estimated) of 42.2% at T-45, rising to 64.4% by the latest reading. September 12, by contrast, sat at 10.9% at T-45 and is still only 19.6% now. Treating early September as a single “shoulder period” rounds that difference away. For where September 5’s depth actually comes from, SHINE UP FES 2026: Tottori’s Demand Peak Sat 100km From the Venue breaks the prefecture down to the city level.
Ryokan also show a quirk that business hotels do not. On September 5, ryokan were at 66.7% at T-45 but 65.3% at T-30 — -1.4pt, flat or worse across that stretch. Adding listed inventory moves the denominator, which can make occupancy dip temporarily, but either way the assumption that a curve “builds steadily through to T-30” does not transfer to Tottori’s ryokan as-is. For how to separate a temporary dip in the curve into demand softening versus inventory being added or withheld, Booking Curves Across 47 Prefectures: +4.1pt Median Pickup from T-45 works through the distinction in detail.
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
Plotted as curves, the difference in character across the four dates becomes clearer still. August 22 and September 5 are “high-starting” curves already in the 80s at T-45, flattening out somewhere inside the last 14 days. August 29 and September 12 start in the high 60s and keep climbing at a steady gradient right through to the latest reading. The former are dates where the use of inventory is effectively decided early; the latter are dates where demand is still winnable close in.
What the completed August 22 shows about property type — business settles at T-14, ryokan add +10.1pt in the final two weeks
August 22 is the only date observed continuously from T-45 through to the day before, so it is the one date where the full allocation of pickup by property type can be checked. The chart below overlays business hotels, ryokan and resort hotels.
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
Business hotels (N=47) were already high at 83.0% at T-45, built to 85.9% at T-30 and 92.4% at T-14, and then moved only +1.4pt across the last 14 days to the day before. They had essentially reached their landing level by T-14. Ryokan (N=71) set off from 68.1% at T-45 and were still at 80.4% at T-14 — and then added +10.1pt to reach 90.5% the day before. Most of that came in the final 48 hours: 83.9% with two days left, then 90.5% the day before. Resort hotels (N=10) traced 80.7% → 85.8% → 89.9% → 92.8%, a front-loaded shape closer to business hotels.
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
Breaking the pickup down by stretch makes the contrast sharper still. The table below shows how many points estimated OCC moved across three stretches — T-45→T-30, T-30→T-14 and T-14→latest — for August 22 and August 29.
| Stay date / property type | N | T-45→T-30 | T-30→T-14 | T-14→latest | Total |
|---|---|---|---|---|---|
| 8/22 Business hotel | 47 | +2.9 | +6.5 | +1.4 | +10.8 |
| 8/22 Ryokan | 71 | +6.7 | +5.6 | +10.1 | +22.4 |
| 8/22 Resort hotel | 10 | +5.1 | +4.1 | +2.9 | +12.1 |
| 8/22 All properties | 187 | +3.7 | +5.5 | +3.4 | +12.6 |
| 8/29 Business hotel | 47 | +3.7 | +4.8 | +6.7 | +15.2 |
| 8/29 Ryokan | 72 | +3.1 | +1.8 | +10.0 | +14.9 |
Source: MetroEngines Research; compiled by the HotelBank Editorial Team. “Latest” for 8/29 is the reading with 3 days remaining.
On August 29 as well, ryokan moved +10.0pt from T-14 to the latest reading — almost exactly the +10.1pt seen on August 22. Two dates are not enough to settle the question, but the view that Tottori’s ryokan have “little headroom left in the final two weeks” does not hold for either of them. It is closer to the reality to read the headroom as concentrated in the final two weeks. The dates with small late pickup are, if anything, the high-occupancy days at business hotels: August 22 (T-14→day before, +1.4pt) and September 5 (T-14→latest, +0.8pt) both reached the 90s by T-14 and barely moved over the remaining fortnight. This relationship — the higher the earlier checkpoint, the less headroom close in — is not specific to Tottori; Tochigi Booking Curves by Type: Resort 81.1% at T-45, Only +2.6pt Left finds the same pattern across four property types.
The early-September trough and the five-day holiday beyond it — checking the day-of-week shape too
Two things help explain why early-September Saturdays are printing on the low side: the calendar, and the day-of-week shape. In 2026, Respect for the Aged Day falls on Monday, September 21 and the Autumnal Equinox on Wednesday, September 23, which makes Tuesday, September 22 a “citizens’ holiday” in between — so Saturday, September 19 through Wednesday, September 23 forms a five-day holiday, the first Silver Week in eleven years (confirmed against the Cabinet Office’s official list of national holidays). September 19, the first day of that run, is running extremely high with 24 days remaining: 97.5% across all properties, 98.2% at business hotels and 96.9% at ryokan, with the sold-out property rate (estimated) at 95.7% for business hotels and 84.0% overall. Early-September Saturdays look relatively soft because September’s demand is leaning heavily toward the five-day holiday starting September 19.
| Period / property type | Mon | Tue | Wed | Thu | Fri | Sat | Sun | Period avg. |
|---|---|---|---|---|---|---|---|---|
| July 2026 — All properties (N=190) | 80.7% | 83.8% | 82.8% | 83.1% | 83.1% | 91.7% | 83.6% | 84.0% |
| July 2026 — Business hotels (N=47) | 80.3% | 86.1% | 85.2% | 85.3% | 83.7% | 93.1% | 82.1% | 85.1% |
| July 2026 — Ryokan (N=73) | 76.5% | 76.0% | 77.8% | 76.4% | 78.3% | 88.7% | 82.1% | 79.2% |
| August 1–27, 2026 — All properties (N=190) | 88.6% | 90.9% | 91.4% | 92.0% | 93.7% | 93.8% | 87.2% | 91.0% |
| August 1–27, 2026 — Business hotels (N=47) | 89.4% | 93.3% | 93.7% | 94.2% | 95.3% | 94.9% | 86.2% | 92.3% |
| August 1–27, 2026 — Ryokan (N=73) | 84.9% | 84.7% | 87.3% | 86.4% | 90.2% | 91.1% | 87.6% | 87.4% |
Source: MetroEngines Research; compiled by the HotelBank Editorial Team. August covers only the completed dates of August 1–27 (three Fridays, four of every other weekday).
July was a month with a pronounced Saturday spike: 91.7% on Saturdays across all properties against 80.7% on Mondays, an 11.0pt spread. August 1–27 includes the Obon period, which lifts the weekdays too, so Saturday’s 93.8% sits against a weekday range of 88.6–93.7% and the day-of-week relief is comparatively flat. With Obon over, September is the stretch where the shape reverts to July’s “only Saturday spikes” pattern, which makes how Saturday inventory is handled a strong lever on the whole month. Of the four Saturdays from August 22 to September 12, only two are in the 90s at the latest reading — business hotels on August 22 (93.8%) and September 5 (94.3%) — while the other two, August 29 at 84.9% and September 12 at 78.0%, fall short of 90%.
For revenue managers running business hotels and ryokan in Tottori — implications and an action plan
Operator’s-eye insights
1. Read your own curve by stretch, not by total. On August 22 business hotels added +10.8pt and ryokan +22.4pt, so on totals alone ryokan look like the bigger movers — but what drives decisions is the allocation. Business hotels put +6.5pt into T-30→T-14; ryokan put +10.1pt into T-14→latest. To benchmark your own pace against the market, log the three points at T-45, T-30 and T-14 every week in the same format and look at which stretch’s gradient diverges from the market. Comparing totals alone cannot tell you whether you are behind or simply shaped differently.
2. In the business segment, identify the “settles at T-14” dates first. Dates that reached the 90s by 14 days out (92.4% on August 22, 93.5% on September 5) then added only +1.4pt and +0.8pt. On dates of that shape, the market as a whole has almost no room left to sell through in the final fortnight. Put the other way round: the contest is in the two weeks from T-30 to T-14, and if you arrive at T-14 having done nothing in that stretch, the result for the date is largely fixed.
3. In the ryokan segment, there is room to redesign the final two weeks as a “leakage window”. On both August 22 and August 29, ryokan moved +10.1pt and +10.0pt from T-14 to the latest reading, so the market moves substantially in that window. If minimum-stay requirements or closed-out inventory are still in place during it, you can end up as the only property with nothing in the market precisely while the market is moving.
4. Treat early September as a different product on each date. September 5 and September 12 are 18.7pt apart at T-45 for business hotels, while September 19 (day one of the five-day holiday) is at 98.2% with a sold-out property rate (estimated) of 95.7% and 24 days still to go. Because the depth of demand differs so much between Saturdays in the same month, a uniform month-level inventory rule leaves money on the table on the heavy dates and misses sales on the thin ones.
Translated onto a timeline, that becomes the following sequence. Every decision trigger is tied to a market-side figure presented in this article.
| Time horizon | Action | Decision trigger | Objective |
|---|---|---|---|
| Today – this week Saturdays 0–7 days out | Take a day-by-day stock check of inventory for Saturday, August 29 and confirm nothing is being held back by consecutive-night conditions or cut-off settings | Your own 8/29 is running below the market’s latest level (business 84.9% / ryokan 74.5%) | Capture the stretch in which the market is moving +6.7pt (business) / +10.0pt (ryokan) from T-14 to the latest reading |
| Today – this week | For Saturday, September 5, narrow the intended guest profile for the remaining rooms and decide in advance how inventory will be released through to the date | The business segment in the market has reached 94.3% with 10 days to go and a sold-out property rate (estimated) of 64.4%, while you still hold double-digit vacant rooms | Avoid dumping inventory cheaply on the assumption this is a thin date, when it is not |
| Within two weeks Dates reaching T-14 | Set a T-14 checkpoint for Saturday, September 12 and decide whether to lift minimum-stay and consecutive-night restrictions | The market level for 9/12 at the latest reading (17 days out) is business 78.0% / ryokan 73.8%, below where 8/22 stood at the same point | Put your inventory on the ryokan-type wave of “+10pt in the final two weeks” |
| Within two weeks | Treat Saturday, September 19 as day one of the five-day holiday and lock down remaining rooms and receiving capacity (housekeeping, staffing) early | The market’s sold-out property rate (estimated) with 24 days to go is business 95.7% / ryokan 81.7% / all properties 84.0% | Avoid a situation where inventory is throttled for operational reasons on a date with deep demand |
| Toward next month Groundwork from T-45 | For Saturdays from October onward, begin fixed-checkpoint tracking that logs T-45, T-30 and T-14 every week in the same format | Your T-45→T-30 gradient runs in the opposite direction to the market (8/22: business +2.9pt / ryokan +6.7pt) | Build an in-house yardstick that separates “behind” from “differently shaped” |
| Toward next month | Switch to date-level inventory rules and revisit uniform month-level minimum-stay and cut-off settings | A double-digit-point spread opens up between Saturdays in the same month at the market’s T-45 level (18.7pt between 9/5 and 9/12) | Reduce both leakage on heavy dates and missed sales on thin ones at the same time |
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
How far will the remaining inventory go? Landing scenarios and a two-axis sensitivity grid
So far this has been about checkpoints we can observe. For August 29, September 5 and September 12, which have not yet reached their stay dates, we can work backwards from stay dates already completed in Tottori to ask how far they could still build from here.
Only one ratio is needed: the remaining-inventory absorption rate. Starting from estimated OCC at a given checkpoint, it measures what share of the remaining inventory disappeared by the day before, defined as absorption rate = (estimated OCC the day before − estimated OCC at the checkpoint) ÷ (100 − estimated OCC at the checkpoint). We measured it across the 61 stay dates falling between June 1 and July 31, 2026, and took quartiles (p25 / p50 / p75) by property type and by checkpoint. The projected landing is calculated as latest estimated OCC + absorption rate × (100 − latest estimated OCC), so by definition it cannot exceed 100%. Checkpoints are matched to each target date: September 12 uses the 17-day distribution, September 5 the 10-day, and August 29 the 3-day.
| Stay date | Property type | N | Latest estimated OCC | Absorption rate p25 / p50 / p75 | Pessimistic | Mid | Optimistic |
|---|---|---|---|---|---|---|---|
| 2026-09-12 (Sat) 17 days out | Business hotel | 46 | 78.0% | 36.2% / 46.0% / 55.2% n=61 | 86.0% | 88.1% | 90.2% |
| Ryokan | 73 | 73.8% | 41.8% / 45.2% / 49.8% n=61 | 84.7% | 85.6% | 86.8% | |
| All properties | 190 | 77.7% | 38.4% / 45.3% / 49.8% n=61 | 86.3% | 87.8% | 88.8% | |
| 2026-09-05 (Sat) 10 days out | Business hotel | 45 | 94.3% | 28.1% / 39.0% / 49.0% n=60 | 95.9% | 96.5% | 97.1% |
| Ryokan | 73 | 75.7% | 39.3% / 41.8% / 47.3% n=61 | 85.3% | 85.9% | 87.2% | |
| All properties | 189 | 88.2% | 32.8% / 39.2% / 45.5% n=61 | 92.1% | 92.8% | 93.6% | |
| 2026-08-29 (Sat) 3 days out | Business hotel | 47 | 84.9% | 18.0% / 25.4% / 33.7% n=60 | 87.6% | 88.7% | 90.0% |
| Ryokan | 72 | 74.5% | 32.5% / 34.6% / 39.4% n=61 | 82.8% | 83.3% | 84.5% | |
| All properties | 190 | 83.0% | 23.4% / 27.7% / 32.2% n=61 | 87.0% | 87.7% | 88.5% |
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
The reading is simple: all properties on September 12 land in the 86.3–88.8% range, September 5 in 92.1–93.6% and August 29 in 87.0–88.5%. The one to watch is business hotels on September 12. Their latest estimated OCC of 78.0% is the lowest of the four dates, but because 17 days is a comparatively long runway, the spread in absorption rates is wide — pessimistic 86.0% to optimistic 90.2%, a 4.2pt band. That is also the date where actions have the most room to bite. Business hotels on September 5, by contrast, span only 1.2pt from a pessimistic 95.9% to an optimistic 97.1%; moving inventory there changes the outcome barely at all.
Assumptions and limits of this calculation. June and July, the months over which absorption was measured, are peak demand months in Tottori, so applying them directly to the post-Obon and early-September shoulder period skews optimistic. In practice it is safer to anchor on the pessimistic column (p25) and treat mid and optimistic as upside. This is also not a forecast — it is only a projection of what happens if the range of absorption actually observed in the past repeats unchanged. New events, group blocks being released and weather-driven late cancellations all sit outside this distribution.
Stepping away from specific dates, here is a two-axis grid for applying the same logic to your own checkpoint. The vertical axis is the level you are observing at the latest reading, the horizontal axis is the absorption rate on remaining inventory, and the intersection is the projected landing.
| Latest observed level | Absorption rate 20% | Absorption rate 30% | Absorption rate 40% | Absorption rate 50% | Absorption rate 60% |
|---|---|---|---|---|---|
| Latest observed 70% | 76.0% | 79.0% | 82.0% | 85.0% | 88.0% |
| Latest observed 75% | 80.0% | 82.5% | 85.0% | 87.5% | 90.0% |
| Latest observed 80% | 84.0% | 86.0% | 88.0% | 90.0% | 92.0% |
| Latest observed 85% | 88.0% | 89.5% | 91.0% | 92.5% | 94.0% |
| Latest observed 90% | 92.0% | 93.0% | 94.0% | 95.0% | 96.0% |
Source: compiled by the HotelBank Editorial Team (calculated on the same definitions as Table 5). Shaded cells mark the band where Tottori’s measured quartiles cluster.
The shaded cells correspond to the central band of absorption rates measured in Tottori (18.0–55.2% depending on property type and checkpoint). Place your own latest level on the vertical axis, then read across toward the pessimistic (left) side of the horizontal axis, and you get the floor for “where this lands if we do nothing”. Only when that floor falls short of target is it worth deploying inventory-side actions such as lifting minimum stays or relaxing consecutive-night restrictions. The action plan in Table 4 is a sequence built on top of that floor test.
Summary — three yardsticks to take away
1. Read the allocation across stretches, not the total pickup. On August 22 every property type added more than 10pt, but business hotels put +6.5pt into T-30→T-14 while ryokan put +10.1pt into T-14→latest. The same amount of pickup, with the decision timing two weeks apart.
2. The level reached at T-14 determines the headroom that follows. Dates that crossed into the 90s by 14 days out (92.4% on 8/22, 93.5% on 9/5) then moved only +1.4pt and +0.8pt over the final fortnight. T-14 is more useful as the branch point for choosing what to do in the last two weeks than as a checkpoint for confirming the result.
3. Do not lump the Saturdays of one month together. At T-45, business hotels stood at 85.8% on September 5 against 67.1% on September 12 — 18.7pt apart — while September 19, day one of the five-day holiday, was at 98.2% with 24 days still to run. September in Tottori is a month where the depth of demand differs sharply from date to date.
About the data
- Definition of estimated OCC (OTA-listed inventory basis): occupancy on an OTA-listed inventory basis = 100 − 100 × OTA-listed remaining rooms ÷ total rooms. It is an estimate based on how inventory offered on OTAs is absorbed, and is defined differently from actual room occupancy (it reads higher).
- Booking curves: based on observations from 45 days before the stay date through to the latest reading. The stay dates covered are August 22, August 29, September 5 and September 12, 2026 (with September 19 for reference), all Saturdays.
- Sold-out property rate (estimated): the share of properties for which no listed inventory can be confirmed on OTAs or similar channels for that stay date (an estimate).
- This article is built solely on the occupancy (estimated OCC) axis and does not cover any price-level metric.
- Breakdown of N: Tottori, all properties N=187–190 (varying by stay date; total rooms 8,161–8,489), business hotels N=45–47, ryokan N=71–73, resort hotels N=10. The day-of-week actuals (Table 3) use all properties N=190, business hotels N=47 and ryokan N=73.
- Number of observed properties: the count of properties with observable inventory tends to rise as the stay date approaches. For business hotels on August 22 it was 40 properties at T-45 and 45 the day before; for ryokan, 59 at T-45 and 71 the day before.
- Data as of August 27, 2026. Sales conditions and inventory change daily, so the figures in this article are a snapshot at the time of retrieval.
References and sources
■ Data sources
An observation series tracking OTA-published inventory daily from 90 days before the stay date through to the day before, covering accommodation properties in Tottori, N=187–190 (total rooms 8,161–8,489). Property-type splits are business hotels N=45–47, ryokan N=71–73 and resort hotels N=10. The day-of-week actuals (Table 3) cover only completed dates in July 2026 and August 1–27, 2026, with all properties N=190, business hotels N=47 and ryokan N=73. Public holidays were confirmed against the Cabinet Office’s official list of national holidays. Data as of August 27, 2026.
■ Calculation assumptions
The landing scenarios in Tables 5 and 6 apply the remaining-inventory absorption rate = (estimated OCC the day before − estimated OCC at the checkpoint) ÷ (100 − estimated OCC at the checkpoint), measured across the 61 stay dates falling between June 1 and July 31, 2026 in Tottori, with the quartiles (p25 / p50 / p75) assigned to the pessimistic, mid and optimistic cases. Checkpoints are matched to each target date, using the 17-day distribution for September 12, the 10-day for September 5 and the 3-day for August 29 (n=60–61 per cell). The projected landing is calculated as latest estimated OCC + absorption rate × (100 − latest estimated OCC), so by definition it cannot exceed 100%. Dates whose starting checkpoint had estimated OCC of 99.5% or higher, and dates with an absorption rate below −5% (cases where added inventory moved the denominator), were excluded from the distribution.
■ Limitations and caveats
Estimated OCC is an estimate based on OTA-listed inventory and is defined differently from actual room occupancy (it reads higher). Direct bookings, telephone reservations and group blocks are not captured. June and July, over which absorption was measured, are peak demand months in Tottori, so applying those rates to the post-Obon and early-September shoulder period skews optimistic; in practice we recommend reading from the pessimistic column. Tables 5 and 6 are not forecasts but an arithmetic projection of what the previously observed range of absorption would produce if it repeated unchanged. August 29 is a snapshot with 3 days remaining, September 5 with 10 days and September 12 with 17 days; none of them are final. Resort hotels have a small base at N=10, where inventory decisions at a single property can move the aggregate. This article is built solely on the occupancy (estimated OCC) axis and does not cover any price-level metric.
- Cabinet Office, “About National Holidays” — official holiday list (syukujitsu.csv) — https://www8.cao.go.jp/chosei/shukujitsu/syukujitsu.csv (confirms September 21, 2026 Respect for the Aged Day, September 22 citizens’ holiday and September 23 Autumnal Equinox Day)
- Tottori Prefecture official site, Tori-Net, “Events in September 2026” — https://www.pref.tottori.lg.jp/327484.htm
- Estimated accommodation inventory and occupancy: MetroEngines Research; compiled by the HotelBank Editorial Team
Related reading
- SHINE UP FES 2026: Tottori’s Demand Peak Sat 100km From the Venue
- Tochigi Booking Curves by Type: Resort 81.1% at T-45, Only +2.6pt Left
- Booking Curves Across 47 Prefectures: +4.1pt Median Pickup from T-45
- Hiroshima Booking Curves: 26.2pt Gap at T-45, Ryokan +10.9pt Late
- Ehime Saturday Booking Curves: Aug +14.7pt, Sep Flat at +1.2pt
- Nagasaki Early Sep Booking Curves: 9/12 OCC 82.7%, Ryokan 24.2pt Gap
- Ibaraki Sep Booking Curves: Peak Sep 19-21 at 74.9%, Sep 23 55.7%
- Okayama Booking Curves: 3 Checkpoints, Aug 8 Late-Surges +11.8pt