Prolepsia forecastYes / no
Will Chinese forces seize or land troops on any territory controlled by Taiwan (including Kinmen, Matsu or Pratas) before 1 January 2028?
As of 17 November 2026, Prolepsia puts the chance of yes at 10%.
10%chance of yes
How Prolepsia reasons
The report Prolepsia wrote with this forecast.
How forecasts are made · Track record
TL;DR
The forecast is 10.2% Yes and 89.8% No for a qualifying Chinese landing or seizure during the specified window, ending January 1, 2028, at 00:00 Taiwan time. Continued coercion without personnel reaching land is the central expectation; a brief offshore-island operation is the main risk.
Key evidence
- The invasion timetable is not fixed. The March 2026 U.S. intelligence assessment said Chinese leaders did not then plan to invade Taiwan in 2027 and had no fixed unification timetable (ODNI). Reuters reported on October 1 that U.S. officials considered invasion before 2028 increasingly unlikely, citing readiness disruption and incentives to await Taiwan’s next presidential election (Reuters). Neither finding excludes a small offshore operation.
- Pressure remains below the land threshold. Taiwan reported 13 Pratas expulsion operations involving 16 Chinese coast-guard vessel-occurrences during 2026 through October 31. The release describes maritime confrontation and jurisdictional claims, not personnel ashore (Taiwan Coast Guard).
- Brief landings are an observed tactic elsewhere. Reuters reported that four Chinese coast-guard personnel briefly landed on Sandy Cay, with no lasting Chinese presence established afterward. That feature is not Taiwanese-administered, but the operation demonstrates a method relevant to this question (Reuters).
- Taiwan is improving interception. A renovated Pratas wharf accepted on July 1 enabled deployment of a 100-ton patrol boat, according to a September 23 release (Taiwan Coast Guard).
Uncertainties
The largest unknown is political authorization for a limited landing, not China’s ability to deliver a small team. A coercive inspection, flag-display operation, or raid could satisfy the criteria without a wider war. Conversely, even a blockade or bombardment would not qualify without personnel reaching qualifying land or a territorial seizure.
Attribution also matters. Suspicious civilian arrivals are not enough: credible evidence must establish PRC direction and a qualifying coercive purpose. Public reporting cannot exclude every covert operation. Target-specific landing preparations or orders to extend enforcement ashore would be much stronger warning signs than another maritime patrol.
Read the full report
TL;DR
The forecast is 10.2% Yes and 89.8% No for a qualifying Chinese landing or seizure between November 17, 2026, and the January 1, 2028 deadline. Continued maritime and aerial coercion without personnel reaching land is the central expectation. Watch for orders and preparations to put security personnel ashore on offshore territory, rather than treating every exercise as an invasion signal.
Context
This question has a much lower threshold than an invasion of Taiwan. One PRC-directed armed or security operative deliberately reaching Taiwanese-administered land during an unauthorized coercive operation is enough. The personnel do not need to fire, remain ashore, or succeed. Taiwan’s administration extends beyond its main island to offshore territories including Penghu, Kinmen, Matsu, Pratas, and Taiping (CRS, October 20, 2026).
The distinction is physical presence on land. Patrols, maritime boardings, blockades, missile attacks, and drone overflights do not qualify by themselves. Nor do civilian arrivals or ordinary espionage. The analysis therefore separates China’s capacity to apply pressure from its willingness to authorize a coercive landing—and from the evidence needed to establish that such a landing occurred.
Evidence
The historical record establishes that offshore seizures are a real pathway, but it does not supply a reliable modern annual frequency. Yijiangshan and the Dachen islands in 1955 are relevant territorial-conquest precedents. Their military and political setting differs sharply from the present, however, and they cannot be treated as interchangeable trials for today’s risk (historical island-seizure analysis). There is also no comprehensive, consistently defined record matching this question’s inclusion of brief, small, or covert coercive landings. That prevents a defensible calculation based simply on years without a reported seizure.
The closest recent behavioral analogue is smaller. Reuters reported on April 28, 2025 that imagery showed four Chinese coast-guard personnel landing by inflatable boat on Sandy Cay and displaying a flag. Philippine personnel later found no Chinese presence, and Manila rejected claims of a lasting seizure (Reuters). Sandy Cay is not qualifying Taiwanese-administered territory. Its relevance is operational: a short sovereignty-assertion landing is an observed tactic. Its limitation is political: landing on a contested sandbar is not equivalent to confronting a Taiwanese garrison.
The clearest measured trend concerns maritime presence around Pratas. AMTI’s study covering January 2021 through December 2025 identified 24 distinct Chinese law-enforcement vessels and 94 activity days. The annual activity-day observations reported here are zero in 2021, 25 in 2024, and 60 in 2025; separate annual values for 2022 and 2023 are not specified. The same research identified five patrols by March 2026 (AMTI, March 30, 2026). These observations show rising exposure. They do not measure landing attempts, and incomplete AIS coverage limits what they reveal about total presence.
More recent official reporting confirms repeated encounters, not a transition ashore. Pratas tallies rose from 11 operations involving 14 vessel-occurrences on October 16, to 12 and 15 on October 20, and 13 and 16 on October 31 (October 16 reporting, October 20 reporting, October 31 Coast Guard release). These are cumulative operations and vessel-occurrences, not unique ships or landings. The final encounter involved an AIS-dark approach and competing jurisdictional claims. The reported one-nautical-mile separation was between ships, not distance from shore.
The pressure is geographically broad. Taiwan described the June 11 intrusion by two Chinese government vessels around Taiping as unprecedented (Taiwan foreign ministry). Around Kinmen, the Coast Guard reported six formation entries in the two weeks preceding its November 9 release; that day’s encounter ended with the Chinese ships departing (Taiwan Coast Guard). I read this as wider exposure to escalation, not evidence that a landing decision has been made.
The strongest restraint evidence concerns intent. The March 2026 U.S. intelligence assessment said Chinese leaders did not then plan a Taiwan invasion in 2027 and had no fixed unification timetable (ODNI). Reuters’ October 1 report said U.S. officials increasingly considered invasion before 2028 unlikely, citing military-purge disruption and Beijing’s interest in Taiwan’s 2028 election (Reuters). These assessments weigh against a major campaign. They do not rule out a limited offshore landing, and anonymous sourcing limits confidence in the more recent account.
Capability is less reassuring than intent. The Pentagon’s 2024 report assessed that operations against small Taiwan-held islands such as Pratas or Itu Aba could require little overt preparation beyond routine training. It also identified substantial political costs, including stronger Taiwanese resistance and international opposition (Pentagon report). This supports meaningful limited-operation risk without establishing that Beijing intends to act.
Taiwan’s response adds friction. Its September 23 release confirmed that Pratas’s renovated wharf had been accepted on July 1, enabling a 100-ton patrol boat to be stationed there (Taiwan Coast Guard). That improves nearshore interception. It does not make a landing impossible. Enforcement disputes also create potential pretexts for Chinese intervention, so stronger enforcement has both protective and escalatory effects.
The latest daily defense bulletin reported five aircraft sorties, six naval vessels, and two official vessels during November 16–17, without announcing personnel ashore (Taiwan defense ministry). This is a limited observation, not exhaustive proof of absence. Taken together, the evidence supports continued pressure below the land threshold as the central outcome.
What's non-obvious
A low invasion risk is not the same as a negligible landing risk. The question includes small offshore features, coast-guard personnel, failed insertions, and immediate withdrawals. A brief coercive landing could qualify while falling far short of the campaign discussed in most invasion reporting. Readiness problems that constrain a major amphibious assault need not constrain delivery of a small security team to the same extent.
The opposite mistake is to treat every patrol as progress toward an inevitable landing. Repeated maritime coercion can be a chosen alternative to taking land. It asserts jurisdiction and tests Taiwan’s responses without necessarily accepting the costs of a territorial attack. Patrol counts therefore measure exposure and pressure, not independent chances that mechanically accumulate into a landing. The decisive missing step is authorization to cross from presence at sea to coercive presence ashore.
Uncertainties
Political authorization is the largest gap. The available evidence establishes neither an authenticated landing directive nor a target timetable. Verified orders for compulsory island inspections, target-specific disembarkation preparations, or shoreward launches of security teams would address that gap far more directly than another exercise or sovereignty broadcast.
Attribution is a separate problem. Research on small-boat arrivals documents access vulnerabilities, but does not establish that the arrivals were PRC-directed coercive operations (CMSI study). Nationality, suspicious behavior, or former military service alone is insufficient. A covert operation would require credible evidence of government direction and qualifying purpose.
Coverage remains incomplete, especially for tiny features and clandestine activity. Public reporting does not provide a complete landing-event census, continuous surveillance of every island, or verified target-specific preparations. That uncertainty cuts both ways: silence cannot prove safety, while ambiguous arrivals cannot establish a qualifying event. The forecast remains 10.2% Yes and 89.8% No.
Forecast history
| Date | Forecast |
|---|---|
| 10% chance of yes |
Sources Prolepsia read
- CRS, October 20, 2026www.everycrsreport.com
- BY ANDREW CHUBB FEBRUARY 2023 | | Taiwan Strait Crises: Island Seizure Contingenciesasiasociety.org
- Philippines says China has not seized disputed South China Sea reefwww.streetinsider.com
- Patrolling Pratas: From Peripheral Concern to a New Normal | Asia Maritime Transparency Initiativeamti.csis.org
- October 16 reportingwww.taipeitimes.com
- October 20 reportingwww.cna.com.tw
- 海洋委員會海巡署金馬澎分署全球資訊網-海巡新聞-中國海警匿蹤侵擾東沙限制水域 海巡巡護七號併航監控強勢驅離www.cga.gov.tw
- MOFA strongly condemns brazen and unprecedented intrusion into waters around Taiping Island by Chinese government vessels, urges international community to jointly counter China’s illegal actionsen.mofa.gov.tw
- ODNIarchive.dni.gov
- US sees Chinese invasion of Taiwan unlikely next year despite 2027 timeline, sources say By Reuterswww.investing.com
The question
This question asks whether forces acting on behalf of the People's Republic of China (PRC) will seize or physically land personnel on any land territory under the effective administration of Taiwan (the Republic of China, ROC) between November 17, 2026, and December 31, 2027, inclusive. The deadline is January 1, 2028, at 00:00 Taiwan Standard Time (UTC+8). Qualifying territory includes Taiwan's main island and all outlying islands, including Penghu, Kinmen (Quemoy), Matsu, Wuqiu, Pratas (Dongsha), and Taiping Island (Itu Aba). The question concerns territory actually administered by Taiwan as of November 17, 2026, rather than all territory claimed by the ROC. Chinese forces include the People's Liberation Army (PLA), People's Armed Police, China Coast Guard, and other armed or security personnel acting under the direction or operational control of PRC authorities. The question covers both military amphibious operations and coercive landings conducted by Chinese government forces, whether as part of a broader invasion, a limited territorial seizure, or an isolated military operation. As of November 17, 2026, the PRC continues to claim sovereignty over Taiwan and its associated territories but does not exercise effective control over them. Taiwan administers its main island and several strategically important offshore territories, some of which lie close to mainland China. The US Congressional Research Service's August 2026 report confirms Taiwan's administration of Penghu, Kinmen, Matsu, Pratas, and Taiping Island. China has substantially increased military pressure around Taiwan in recent years, including military exercises, aircraft operations, and maritime patrols. Taiwan's Coast Guard Administration has documented repeated Chinese Coast Guard incursions into restricted waters around Kinmen and Pratas during 2025 and 2026. For example, Taiwan reported additional Chinese Coast Guard incursions near Kinmen in July 2026. These maritime activities illustrate the potential for escalation but do not themselves constitute a qualifying landing or territorial seizure. Sources include the Congressional Research Service (August 5, 2026), Taiwan Coast Guard Administration (July 21 and July 29, 2026), and US Department of Defense assessments of Chinese military capabilities. The forecast specifically concerns the escalation from maritime or aerial pressure to the physical insertion of Chinese government forces onto Taiwanese-administered land, or the forcible establishment of Chinese control over such territory, before the specified deadline.
How it resolves
The question resolves YES if, between November 17, 2026, and December 31, 2027, inclusive, at least one of the following occurs: (1) Armed personnel belonging to the PLA, People's Armed Police, China Coast Guard, or another PRC-directed armed or security organization deliberately land, disembark, or are inserted onto any Taiwanese-administered island or other land territory as part of a military, paramilitary, coercive, or unauthorized security operation. (2) PRC forces forcibly seize, occupy, or establish effective physical control over any portion of Taiwanese-administered land territory, regardless of the duration of control or whether Taiwan subsequently recaptures it. (3) PRC forces deploy personnel onto Taiwanese-administered land through amphibious assault, helicopter insertion, airborne operations, or another military insertion method, even if the operation is unsuccessful, the personnel are immediately captured, or they withdraw shortly afterward. A qualifying landing must involve at least one PRC government-directed armed or security operative physically reaching land and must be connected to an unauthorized military, paramilitary, or coercive operation. The personnel need not fire weapons, engage Taiwanese forces, or establish lasting control. An unopposed but unauthorized military landing qualifies. A landing by armed personnel acting under covert PRC government direction also qualifies if that affiliation and purpose are established by credible evidence. The question resolves NO if no qualifying landing or territorial seizure occurs before January 1, 2028. Specifically, Chinese military exercises, naval deployments, coast guard patrols, violations of Taiwan's territorial waters or airspace, blockades, missile launches, aerial bombardment, drone overflights, cyberattacks, and maritime confrontations do not qualify unless accompanied by a qualifying landing or territorial seizure. Unarmed civilian arrivals, accidental incursions, shipwrecks, emergency landings, routine authorized visits, diplomatic exchanges, and personnel entering with Taiwan's permission for non-coercive purposes do not qualify. Civilian fishing vessels entering Taiwanese waters or civilians landing without evidence of a PRC-directed armed or coercive operation do not qualify. Resolution will rely primarily on official statements from Taiwan's Ministry of National Defense, Coast Guard Administration, and other relevant Taiwanese authorities, supplemented by PRC government or military statements, US Department of Defense reporting, and credible international news organizations such as Reuters, Associated Press, BBC, and AFP. Independent evidence, including authenticated imagery, satellite observations, and corroborated eyewitness reporting, may also be considered. A public acknowledgment by either government is sufficient but not necessary. A YES resolution requires credible evidence establishing that a qualifying event occurred. Where accounts conflict, preference will be given to independently corroborated reporting supported by physical evidence or multiple reliable sources, rather than unsupported assertions from either side. A single uncorroborated government allegation will not automatically establish a YES outcome. Events occurring before the deadline may be confirmed afterward. The question will ordinarily be resolved using evidence publicly available by January 31, 2028, allowing time for reporting and verification. If no qualifying event is established by that date, the question resolves NO unless credible evidence indicates that a potentially qualifying event occurred and remains under active investigation. In that exceptional case, resolution may be deferred until February 29, 2028. If the available evidence remains insufficient to establish a qualifying event, the question resolves NO. The question should be annulled only if extraordinary circumstances make a meaningful determination of the outcome impossible, rather than merely because reporting is incomplete or disputed.
Fine print
All dates are interpreted using Taiwan Standard Time (UTC+8). The event must begin within the forecast window; a landing that occurred before November 17, 2026, does not qualify merely because Chinese personnel remain present afterward, unless a separate qualifying seizure or landing occurs during the forecast period. The definition of Taiwanese-administered territory is based on effective administrative control as of November 17, 2026. It includes small islands, reefs, and other naturally occurring land formations under Taiwan's effective administration, provided there is land above water on which personnel can physically land. It does not include territorial waters, submerged reefs, artificial structures at sea that are not part of Taiwanese-administered land territory, or territories claimed by Taiwan but administered by another government. A successful territorial seizure is not required. A single unauthorized military landing involving even a small number of PRC personnel can resolve the question YES. Conversely, a large-scale naval blockade or sustained aerial bombardment without the physical landing of personnel or seizure of land does not qualify. Landing on the territory of a Taiwanese-administered island qualifies even if the location is an uninhabited beach, reef, or remote military outpost. An attempted landing that fails before any personnel reach land does not qualify unless Chinese forces separately establish physical control over Taiwanese-administered land. Personnel operating disguised as civilians qualify only if credible evidence establishes that they were acting as PRC-directed military, paramilitary, or security operatives conducting a coercive operation. Ordinary espionage, intelligence gathering by undercover agents, political influence operations, and unauthorized civilian visits do not qualify without a corresponding armed or coercive landing operation. The question concerns actions attributable to the PRC government, not independent actions by private Chinese citizens. The nationality of participants alone is insufficient to establish a qualifying event. The question does not require a formal declaration of war, a sustained military occupation, casualties, armed resistance, or international recognition of a change in territorial control.
More public forecasts
Will Taiwan's Ministry of National Defense report more than 153 Chinese military aircraft around Taiwan in a single daily report between 1 January and 31 December 2027?
21%chance of yes
Prolepsia forecast
Will Dubai Land Department record a sale of a single residential apartment or villa worth at least AED 1 billion between 1 November and 31 December 2026?
1%chance of yes
Prolepsia forecast
Will the United States carry out air or missile strikes on Iranian territory between 1 January and 31 December 2027?
88%chance of yes
Prolepsia forecast
Will Xi Jinping be confirmed for a fourth term as General Secretary of the Chinese Communist Party before 1 January 2028?
96%chance of yes
Prolepsia forecast
Will the EU and China agree minimum prices that replace EU tariffs on Chinese electric cars before 31 December 2027?
26%chance of yes
Prolepsia forecast
Will China fire ballistic missiles into waters around Taiwan during a military exercise between 1 January and 31 December 2027?
21%chance of yes
Prolepsia forecast