With the Kobe Marathon 2026 (Sunday, November 15, 2026) six days away, estimated OCC (based on OTA-listed inventory) for the race eve of November 14 across 23 business hotels (3,982 rooms) in Kobe’s Chuo Ward had already reached 96.0% at 45 days out, and the latest observation at 6 days out still stands high at 97.3%. Race night, November 15, tells a different story: its 88.9% at 7 days out was almost level with the previous Sunday, November 8, at the same point (89.5%). The 5.0pt lead over the prior week at 45 days out had shrunk to 0.7pt by 30 days out, suggesting that race-night demand was largely exhausted early. This article compares the booking curves for three nights — race eve, race night and the following Monday — against the prior week, and sets out the actions still available for race night in the final week.
Scope: Chuo Ward, Kobe, Hyogo — business hotels N=23 properties (3,982 rooms); city hotels N=11 properties (1,569 rooms; stay dates from November 13). Prices in this article are estimated settled ADR (the settled price level estimated from OTA and other sales data, tax-exclusive equivalent); occupancy is an estimate based on OTA-listed inventory. Both are defined at the end of the article. Data as of November 9, 2026.
• ADR (average daily rate) = an estimated settled room rate (tax-exclusive equivalent), calculated by applying a category-specific adjustment factor to the lowest plan price each property publishes on OTAs (per room, two guests, tax-inclusive). Checked against property-level results disclosed by listed hotel REITs (91 properties, latest 3 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 or ward) is the median of the estimated settled rates of the properties covered (the level of a typical property in the area).
• Wherever listed prices are mentioned, they are per-room rates for two guests (tax-inclusive).
• OCC (occupancy) = the share of rooms sold out of total rooms in the area (an estimate based on OTA sales inventory).
• Source: MetroEngines Research (continuous tracking of OTA public rates for approximately 168,000 properties in Japan, of which approximately 27,000 have verified activity)
- — Race eve, November 14, was at 96.0% at 45 days out (estimated OCC, business hotels N=23 properties, 3,982 rooms) — 12.3pt above the previous Saturday. It peaked at 98.8% at 19 days out and has since eased to 97.3% at 6 days out.
- — Race night’s lead over the previous Sunday went from +5.0pt to −0.6pt. It narrowed to +0.7pt by 30 days out, and at 7 days out the night stood at 88.9%, in line with the previous Sunday (89.5%).
- — The final-week actions are “reselling returned inventory” and “multi-night and extended stays around race night”. The previous Sunday added +6.2pt in its last week, so last-minute demand may still remain for race night.
Event basics — 61,000 applicants for 20,000 places, with both start and finish in Chuo Ward
First, the size of the demand pool. According to Hyogo Prefecture, 61,173 runners entered the full marathon of the Kobe Marathon 2026 for a field capped at 20,000, so places were allocated by lottery. Applicants comprised 20,441 from within Hyogo, 37,899 from other prefectures and 2,833 from overseas (as of June 1), meaning just over 60% live outside the prefecture. The Kobe Shimbun later reported that at the organizing committee’s general meeting in August, overseas applicants reached a record 3,364, 95% of them from Asia.
What matters for hotel location is the course. According to Kobe City, from the 2025 race the course starts in front of Kobe City Hall, turns at Okura Kaigan in Akashi City and finishes at Kobe Harborland, and the 2026 race follows the same route. Because both the start and the finish are in Chuo Ward, hotels in the ward are a natural first choice for participants from afar on both race eve and race night. From here we look at which stay dates that demand actually built up on, and when.
Race eve, November 14 — 96.0% at 45 days out, with a little inventory coming back over the last two weeks
The defining feature of race eve is that the sale was all but decided by 45 days out. Business hotels on November 14 stood at 96.0% at 45 days out, 12.3pt above the previous Saturday, November 7, at the same point (83.7%). The night then climbed to 98.0% at 30 days out and to the observation-period peak of 98.8% at 19 days out, but the latest reading at 6 days out is 97.3%, 1.5pt off the peak. The previous Saturday kept building right to the end, landing at 96.8% at 1 day out — so even on the same day of the week, the curve shapes differ clearly: race eve “fills early and gives a little back,” while an ordinary Saturday “keeps building to the last day.” A comparison of Kobe against the Kanazawa and Yokohama race eves at the same lead time is covered in Kanazawa Marathon 2026 Eve: OCC 91% at LT45; Yokohama, Kobe Compared.
Chuo Ward, Kobe — business hotels N=23 properties, 3,982 rooms. Estimated OCC (based on OTA-listed inventory); the horizontal axis is days remaining until the stay date (45 days out → 1 day out).
Source: MetroEngines Research; compiled by HotelBank Editorial Team
The pullback after the peak is small, but 1–2pt of inventory returning to the market on a near-sold-out night is not trivial. Some bookings secured early appear to have been released, which gives even properties that thought they had sold out race eve room to resell recently returned rooms to race-eve demand. Snapshot values for the key stay dates are shown in the table below.
| Stay date | Category | 45 days out | 30 days out | 21 days out | 14 days out | 7 days out | Latest observation |
|---|---|---|---|---|---|---|---|
| November 7 (Sat) | Business | 83.7% | 88.6% | 90.5% | 91.9% | 94.5% | 96.8% (1 day out) |
| November 8 (Sun) | Business | 71.9% | 79.1% | 83.8% | 85.6% | 89.5% | 95.7% (1 day out) |
| November 14 (Sat) race eve | Business | 96.0% | 98.0% | 98.1% | 98.2% | 97.1% | 97.3% (6 days out) |
| November 15 (Sun) race night | Business | 76.9% | 79.8% | 84.8% | 87.0% | 88.9% | 88.9% (7 days out) |
| November 16 (Mon) night after | Business | 60.1% | 65.9% | 74.8% | 81.3% | — | 85.2% (8 days out) |
| November 21 (Sat) first day of 3-day weekend | Business | 89.3% | 98.3% | 99.6% | 99.7% | — | 99.4% (13 days out) |
| November 14 (Sat) race eve | City | 92.1% | 95.0% | 95.0% | 93.8% | 93.3% | 93.2% (6 days out) |
| November 15 (Sun) race night | City | 82.8% | 86.1% | 89.3% | 89.7% | 88.1% | 88.1% (7 days out) |
| November 16 (Mon) night after | City | 70.6% | 73.9% | 79.5% | 81.1% | — | 82.9% (8 days out) |
Estimated OCC (based on OTA-listed inventory). Business: N=23 properties, 3,982 rooms. City: N=11 properties, 1,569 rooms for stay dates from November 13 (N=10 properties, 1,479 rooms for November 7 and 8). “—” indicates a point not yet reached.
Source: MetroEngines Research; compiled by HotelBank Editorial Team
Race night, November 15 — the lead faded by 30 days out, and at 7 days out it matched the previous Sunday
What about race night? Business hotels on November 15 added +12.0pt, from 76.9% at 45 days out to 88.9% at 7 days out. Compared with the previous Sunday, November 8, at the same points, however, the lead moved from +5.0pt at 45 days out to +0.7pt at 30 days out, +1.0pt at 21 days out and +1.4pt at 14 days out, then flipped to −0.6pt at 7 days out. City hotels likewise went from +7.4pt at 45 days out to −0.4pt at 7 days out — notably, both categories trace the same shape.
Estimated OCC for November 15 (Sun) minus the value for November 8 (Sun) at the same number of days out (pt). Business: N=23 properties, 3,982 rooms. City: N=11 properties, 1,569 rooms for November 15; N=10 properties, 1,479 rooms for November 8.
Source: MetroEngines Research; compiled by HotelBank Editorial Team
The takeaway is that race-linked bookings for race night were largely in before 45 days out, and from 30 days out the night has been building at the same pace as an ordinary Sunday. That said, the picture is not all negative. The previous Sunday, November 8, added +6.2pt in its final week, from 89.5% at 7 days out to 95.7% at 1 day out, so a similar amount of last-minute demand may still remain for November 15. Compared with the Sunday two weeks earlier, November 1, which stood at 82.3% at 7 days out, race night is entering its final week from a high level for a Sunday. How to close the gap between race eve and post-race night is examined with the same framework in our analysis of Okayama Marathon 2026 in Kita Ward, Okayama City. Looking back at October Sunday nights, business hotels landed across a wide range: 78.7% on October 4, 91.6% on October 18 and 86.3% on October 25 (all observed the day before the stay date).
The night after, November 16 (Mon), is at 85.2% at 8 days out, only 2.8pt above the previous Monday, November 9, at the same point (82.4%). Monday is also a day driven mainly by business travel, so it is hard to read the race effect as extending broadly into the following night. Meanwhile, the following Saturday, November 21, is the first day of a three-day weekend and already stands at 99.4% at 13 days out — above race eve on November 14. Late November in Chuo Ward thus has two peaks a week apart: marathon eve and the three-day weekend. For how the three-day-weekend uplift plays out by category in Hyogo, see our article comparing Hyogo’s October weekends across four categories’ booking curves.
City hotels — race eve eased from a 96.6% peak to 93.2%; race night at 88.1%
For city hotels (N=11 properties, 1,569 rooms), the post-peak pullback was larger than for business hotels. November 14 rose from 92.1% at 45 days out to 96.6% at 17 days out, then fell 3.4pt to 93.2% at 6 days out. November 15 also gave back 2.6pt, from a peak of 90.7% at 18 days out to 88.1% at 7 days out, so for both race eve and race night the immediate issue is “reselling returned inventory” rather than “building further.”
Chuo Ward, Kobe — city hotels. N=11 properties, 1,569 rooms for November 14 and 15; N=10 properties, 1,479 rooms for November 8. Estimated OCC (based on OTA-listed inventory); the horizontal axis is days remaining (45 days out → 1 day out).
Source: MetroEngines Research; compiled by HotelBank Editorial Team
As a price yardstick, consider the estimated settled ADR (monthly) for Hyogo Prefecture as a whole. Comparing finalized months year on year, business hotels kept rising: +2.8% in August, +1.9% in September and +5.0% in October. City hotels ran below the prior year at −8.7% in August and −7.8% in September, then recovered to ¥13,900 in October, flat with the prior year (−0.0% YoY). November, which includes the marathon, is still an estimate at this point, so a comparison against finalized figures must wait until the month closes.
| Hyogo | Aug 2025 | Aug 2026 | YoY | Sep 2025 | Sep 2026 | YoY | Oct 2025 | Oct 2026 | YoY |
|---|---|---|---|---|---|---|---|---|---|
| Business hotels | ¥7,884 (N=168) | ¥8,104 (N=167) | +2.8% | ¥7,678 (N=169) | ¥7,826 (N=168) | +1.9% | ¥7,657 (N=167) | ¥8,040 (N=167) | +5.0% |
| City hotels | ¥14,632 (N=47) | ¥13,366 (N=47) | −8.7% | ¥13,808 (N=46) | ¥12,737 (N=47) | −7.8% | ¥13,945 (N=47) | ¥13,940 (N=47) | −0.0% |
Estimated settled ADR (tax-exclusive equivalent). All comparisons are between finalized (past) months. Estimates for November 2026 based on current sales are ¥11,200 for business hotels (N=162) and ¥17,300 for city hotels (N=48), but these may change by month-end, so a simple comparison with finalized figures must wait for the month to close.
Source: MetroEngines Research; compiled by HotelBank Editorial Team
For revenue managers running business and city hotels in Chuo Ward, Kobe — implications and action plan
1. Race eve is not in a “sold out” phase but a “resell returned inventory” phase. The market’s race eve rose from 96.0% at 45 days out to as high as 98.8%, and has recently eased back to 97.3%. For city hotels the pullback was from 96.6% to 93.2%. The first step is to check whether your own property has seen inventory return over the same period. Returned rooms are inventory that can fully capture the strength of race-eve demand; deciding the conditions under which to sell them comes before any rush to move price.
2. Race night is better treated as “Sunday last-minute demand” than as event demand. With the lead over the prior week essentially gone by 30 days out, it makes sense to track race night over the final week with the same yardstick as an ordinary Sunday. The previous Sunday’s +6.2pt in its last week suggests last-minute demand may remain, but whether you capture it depends on how your property presents itself.
3. The gap between race eve and race night can be read as room to fill with multi-night stays. The difference between race eve at 97.3% and race night at 88.9% is also the difference in whether the same runners choose a second night. For participants who want to rest after finishing rather than travel, or those coming from overseas who want to add sightseeing the next day, a race-night stay has value beyond an ordinary Sunday night. Building conditions that communicate that value is likely to work better here than raising or lowering prices.
4. Treat the two late-November peaks separately. The first day of the following three-day weekend, November 21, has already reached 99.4% and is filling on demand unrelated to the marathon weekend. It is important not to carry any shortfall from the marathon weekend into pricing decisions for the following week.
| Timing | Action | Decision trigger (figures from this article) | Aim |
|---|---|---|---|
| Today – November 11 | Resell inventory that has returned for November 14 to race-eve demand. Consider setting aside some rooms conditioned on a two-night stay on the 14th and 15th | The market’s race eve eased from 98.8% (19 days out) to 97.3% (6 days out). If returned inventory has appeared at your property too | Use race-eve strength as leverage to fill race-night vacancies |
| Today – November 11 | Prepare a race-night plan designed for post-race stays (late check-in, luggage storage, conditions allowing a relaxed departure the next morning) | If your property’s booking progress for November 15 is below the market’s 88.9% at 7 days out | Resell race night as “a night to rest after the race” |
| November 12 – 14 | Track daily pickup for November 15 against the same yardstick as the previous Sunday | The previous Sunday, November 8, went from 89.5% at 7 days out to 95.7% at 1 day out (+6.2pt). If your pickup falls short of this | Detect last-minute shortfalls early and feed them into revisiting conditions |
| November 12 – 14 | For race-eve guests checking out on November 15, consider offering extended use so runners can return to freshen up after the race | Properties where race eve is high (business 97.3%, city 93.2%) and rooms sit empty during race-day daytime | Turn daytime rooms into an additional touchpoint |
| November 15 – 16 | Prepare information on 15th–16th consecutive stays and sightseeing extensions in languages that also reach overseas participants | The night after, November 16, is at 85.2% at 8 days out, only +2.8pt above the previous Monday at the same point (82.4%) | Explore room to fill Monday vacancies with runners’ extended stays |
| Looking ahead to next year | Save this year’s race-eve and race-night curves alongside your own results and use them as the starting point for next year’s pricing calendar | This year’s pattern: the race-night lead shrank from 5.0pt at 45 days out to 0.7pt at 30 days out | Move race-night decisions to before 30 days out |
Summary — three yardsticks for measuring event eve and event night separately
The curves for Chuo Ward ahead of the Kobe Marathon 2026 yield three yardsticks usable in future years. First, race eve is largely decided by 45 days out. A lead of 96.0% for business hotels, +12.3pt over the previous Saturday, shows that the time to set prices and conditions comes before 45 days out. Second, the race-night lead was short-lived. With +5.0pt at 45 days out shrinking to +0.7pt by 30 days out, race-night measures should ideally be fully deployed by 30 days out, after which pace management is the same as for an ordinary Sunday. Third, inventory comes back late even on nights that are nearly sold out. The race-eve pullbacks of 1.5pt for business hotels and 3.4pt for city hotels show that reselling work remains in the final week.
Recording these three alongside your own results should make it easier to decide, ahead of time at next year’s race, “when to set prices, when to change conditions and when to resell.”
About the data
| Estimated OCC | Occupancy based on OTA-listed inventory = 100 − 100 × OTA-listed remaining rooms ÷ total rooms. Covers a fixed cohort of properties that list at least 30% of their total rooms on OTAs (the same set of properties for each stay date; the N in tables and notes is the number of properties). This is an estimate based on how OTA-sold inventory is absorbed and differs in definition from actual room occupancy (rooms not listed are counted as sold, so it tends to run high). |
| Booking curve | Based on observations from 45 days before the stay date to the latest available. Each value is the estimated OCC for that stay date on the day it was observed. |
| Estimated settled ADR | A settled price level (tax-exclusive equivalent) estimated from OTA and other sales data (lowest plan level × category-specific factor, ensemble of multiple channels). Past months are finalized values; current and future months are estimates based on current sales. Median error against published operating results: 6.6%. This article uses monthly values for Hyogo Prefecture as a whole. |
| Coverage N= | Estimated OCC: business hotels in Chuo Ward, Kobe, Hyogo N=23 properties, 3,982 rooms (common to all stay dates); city hotels N=11 properties, 1,569 rooms (stay dates from November 13) / N=10 properties, 1,479 rooms (November 7 and 8). October Sunday-night values: business hotels N=22–23 properties. Estimated settled ADR: Hyogo business hotels N=162–169 properties, city hotels N=46–48 properties (varies by month; shown by month in the table). |
| Data as of | Data as of November 9, 2026. Sales and inventory change daily, so the figures in this article are a snapshot at the time of collection. Days of the week follow the 2026 calendar. |
References and sources
| Kobe City, “Kobe Marathon 2026 is scheduled for November 15, 2026” (「神戸マラソン2026」は2026年11月15日に開催を予定しています, February 6, 2026) | city.kobe.lg.jp |
| Hyogo Prefecture, “Runner entries for Kobe Marathon 2026, the second year of the new course, top 60,000!” (新コース2年目となる「神戸マラソン2026」ランナーエントリー数が6万人を突破！, June 12, 2026) | web.pref.hyogo.lg.jp |
| Kobe Shimbun NEXT, “Kobe Marathon 2026: overseas runner applications hit record 3,364, 95% from Asia” (神戸マラソン2026 海外ランナーの応募が過去最多更新 3364人、95％がアジアから) | kobe-np.co.jp |
Related Articles
- Kanazawa Marathon 2026 Eve: OCC 91% at LT45 (+10.6pt vs. Previous Saturday) — Yokohama and Kobe Compared at the Same Lead Time
- Japan’s Autumn Marathons 2026: Race-Eve Hotel Rates Up to +49.1% — Kanazawa, Mito, Toyama and Kobe
- Okayama Marathon 2026 (Nov 8): Race Eve 88.5% OCC vs Race Night 67.7% in Kita Ward — Closing a 20pt Gap in 10 Days
- Shimonoseki Kaikyo Marathon: Race Eve Oct 31 at 90.4% OCC at T-45, 94.3% Now — How to Sell Race Night (83.6%) and Two Nights Before (77.7%)
- Hyogo October Weekends: Holiday Lift Only for Ryokan & Resorts (+5.4pt) — Booking Curves for Four Categories
- Hyogo Booking Curves: 3.6x Pickup Gap by Category, Ryokan Sellout Rate 28.4%
- 12 Marathons Oct–Feb: Rooms Within 5km of the Start Measured — Kanazawa 13,964 vs Shonan 305