A large, undiscovered ninth planet lurks in the outer solar system, shepherding the orbits of distant icy bodies
Where the evidence lands: UnresolvedThat an undiscovered planet of several Earth masses orbits the Sun far beyond Neptune, on a distant and highly elongated path, and that its gravity is responsible for the observed alignment and clustering of the orbits of certain distant Kuiper Belt objects.
Believed by: A substantial share of professional planetary astronomers treat it as a leading explanation worth testing, while a vocal group of researchers argues the evidence is consistent with selection bias; the idea also has a large popular following
The full story
What is documented
Start with what is not in dispute. Beyond Neptune lies the Kuiper Belt, a broad ring of icy bodies, and among its most distant members are a handful of objects on enormous, elongated orbits. Some of them, like Sedna, discovered in 2003, never come anywhere near Neptune, so the giant planets we know about cannot easily account for how they got where they are.
In 2014, astronomers Chad Trujillo and Scott Sheppard noticed something else: several of these extreme objects seemed to share a similar orientation in space, as though nudged into step by a common influence. They floated the idea that an unseen super-Earth-mass perturber might be responsible.
Two years later, in January 2016, Caltech's Konstantin Batygin and Michael Brown turned that suggestion into a detailed, quantitative hypothesis. Their computer models indicated that a distant planet on a particular kind of orbit could produce and maintain the observed pattern. They named it Planet Nine. Crucially, they did not see a planet; they inferred one from the behavior of other objects. That distinction, between a documented pattern and an inferred cause, is the whole of this case.
The case for a hidden planet
The strongest form of the argument is genuinely compelling, which is why so many professional astronomers take it seriously. It begins with a real anomaly rather than a hunch. Detached orbits like Sedna's are hard to explain, and the apparent alignment of the most distant objects is exactly the fingerprint a large, far-off planet would be expected to leave.
There is also a proud historical precedent. In the 1840s, Neptune was predicted mathematically from small irregularities in the orbit of Uranus, and then found almost exactly where the calculations said it would be. Inferring an unseen planet from gravitational clues is not a fringe move; it is one of the great success stories of celestial mechanics.
And the hypothesis does more than explain the one pattern it was built on. Its proponents argue it also predicts a separate population of highly tilted distant objects and can account for a slight tilt in the plane of the solar system, so a single planet ties together several loose threads at once.
Neptune was a point of light no one had seen until mathematics said where to aim the telescope. Planet Nine asks whether the sky still has one more of those waiting.
Best of all, the claim is testable. Batygin and Brown published where to look, and a new generation of surveys is now looking. That is what separates this from an unfalsifiable belief: it makes a bet that observation can win or lose.
Where the skeptics push back
The central objection is disarmingly simple: the clustering may not be real. We do not survey the whole sky evenly. Telescopes search particular patches, often near where objects are brightest and easiest to catch, and distant bodies are only visible for the small fraction of their long orbits when they swing closest to the Sun. Both effects can manufacture the appearance of a pattern where the underlying population is actually uniform.
In 2021, Kevin Napier and colleagues put this to the test. Combining objects from three different surveys and carefully modeling where each had looked, they found the sample statistically consistent with no clustering at all. If there is nothing to explain, there is no need for a planet to explain it.
The sample is also tiny. The conclusions rest on a couple of dozen objects, and with numbers that small the statistics are fragile, swinging with each new discovery and with the assumptions built into the bias model. Reasonable analysts have looked at nearly the same data and reached opposite conclusions.
Even the detached orbits have rivals. A close pass by another star in the solar system's youth, or the collective gravity of a self-gravitating disk of small bodies, could lift objects like Sedna without any surviving ninth planet. More exotic proposals, including a small primordial black hole, have also been floated. None is clearly better than Planet Nine, but their existence shows the data do not force one answer.
Then there is the plainest point of all: a decade of dedicated searching has not produced the planet. Absence of detection is not proof of absence, but each empty search shrinks the space where it can hide.
The 2025 candidate, and why it is not a confirmation
In April 2025 a team led by Terry Long Phan and Tomotsugu Goto reported an intriguing result. By comparing two far-infrared sky surveys taken 23 years apart, the 1983 IRAS mission and the 2006 AKARI mission, they searched for a faint source that had crept slowly across the sky, as a very distant planet would. They flagged one candidate that appeared in one survey and not at the same spot in the other.
Headlines treated it as a possible discovery, but the researchers themselves were careful. A single candidate from archival data needs direct follow-up with a modern telescope to confirm it is a real, moving object rather than an artifact or an unrelated source.
More striking, Brown, one of Planet Nine's own architects, pointed out that the candidate's implied orbit looks too inclinedto be the Planet Nine his models predict. So even if the source is confirmed as a distant body, it would not automatically vindicate the original hypothesis; it might instead be a different object entirely. It is a reminder that “a planet was found” and “the predicted Planet Nine exists” are not the same statement.
The coming test
What makes Planet Nine unusual among the ideas catalogued here is that it may not stay unresolved for long. The Vera C. Rubin Observatory in Chile, with its enormous camera and a plan to image the entire southern sky again and again, is beginning a decade of observations designed, among many other goals, to hunt for exactly this kind of faint, slow-moving object.
Astronomers involved in the search have said that if Planet Nine exists in the predicted region, Rubin has a good chance of turning it up within the first years of the survey, and that even a null result would sharply constrain where it could still be. Either the planet appears, or the room for it to hide keeps shrinking.
That is why this file rates the existence claim as unproven rather than debunked or substantiated. The clustering is documented; the models are published and serious; and the planet has never been seen. Some of the sharpest specialists suspect the pattern is a mirage of selection bias, while others expect a detection any year now. Both camps are doing science, and both agree on the thing that matters here: the question is open, and the sky itself is about to answer it.
Watch
What's still unexplained
- Is the observed clustering of distant objects a real physical pattern or an artifact of where surveys have searched? Independent analyses still reach opposing conclusions, and a larger, less biased sample is needed to decide.
- If a planet is required, what are its actual mass and orbit? Estimates have shifted over time, and different models point to somewhat different regions of sky, which complicates the search.
- The 2025 infrared candidate remains unconfirmed, and its apparent orbit may be incompatible with the predicted Planet Nine, leaving open whether it is Planet Nine, a different distant object, or noise.
- Can the Vera C. Rubin Observatory settle the question this decade? Astronomers expect its survey to either detect the planet or tighten the constraints enough to make continued belief hard to sustain.
Point by point
The claim: The orbits of distant Kuiper Belt objects cluster in a way that only a massive hidden planet can explain.
What the record shows: The clustering is a real feature of the catalogued objects, and gravitational shepherding by a distant planet is one credible explanation for it. But it is contested. Napier and colleagues (2021) found that once the specific patches of sky each survey examined are folded in, the combined sample is statistically consistent with no clustering at all. The pattern that looks striking on a plot may partly reflect where telescopes happened to point, so the observation supports the hypothesis without confirming it.
The claim: The alignment is so improbable by chance that a planet is the natural conclusion.
What the record shows: Batygin and Brown originally estimated the raw clustering had well under a one percent chance of being coincidental, and later argued that a bias-corrected significance still favored a perturber. Critics reply that the significance depends heavily on the small sample size and on how observational selection is modeled. With only a couple of dozen relevant objects known, the statistics remain sensitive to assumptions, which is why the field has not reached consensus.
The claim: A candidate planet has already been spotted in infrared survey data.
What the record shows: In 2025 a team reported one promising slow-moving source by comparing infrared sky surveys taken 23 years apart. It is unconfirmed and awaits direct follow-up with instruments like DECam. Notably, Brown has said the candidate's implied orbit is too inclined to be the Planet Nine his own models call for, so even a confirmed detection there would not straightforwardly validate the original hypothesis.
The claim: Only a real planet can account for detached bodies like Sedna that never approach Neptune.
What the record shows: Detached, high-perihelion objects are a genuine puzzle, and a distant planet is one way to lift their orbits away from Neptune. But it is not the only way. Proposed alternatives include a passing star early in the solar system's history, a self-gravitating disk of leftover small bodies, and other primordial processes, none of which require a ninth planet to exist today.
The claim: If the planet were not there, no other model could reproduce the same effects.
What the record shows: Several competing ideas have been published, from a self-gravitating outer disk to more exotic proposals such as a small primordial black hole, alongside the simplest alternative that there is no genuine clustering to explain. These competitors are not obviously better than Planet Nine, but their existence shows the observations do not force a single conclusion.
Timeline
- 2003-11Michael Brown, Chad Trujillo, and David Rabinowitz discover Sedna, a distant body on a detached orbit that never comes near Neptune. Its perihelion sits far beyond the reach of the known giant planets, raising an early question about what could have placed it there.
- 2014-03Trujillo and Sheppard announce the discovery of 2012 VP113 and note that it and other extreme trans-Neptunian objects share a similar argument of perihelion. They suggest the pattern could be maintained by an unseen super-Earth-mass perturber in the distant solar system.
- 2016-01-20Batygin and Brown publish “Evidence for a Distant Giant Planet in the Solar System” in The Astronomical Journal. Their simulations indicate that a planet of about ten Earth masses on an anti-aligned, eccentric orbit could produce the observed clustering, which they calculate has a low probability of arising by chance.
- 2016The hypothesis makes further predictions to be tested: a distinct population of very high-inclination and even retrograde distant objects, and a slight tilt of the solar system's planetary plane relative to the Sun's equator, both of which Planet Nine could help explain.
- 2019Brown and Batygin publish a review, “The planet nine hypothesis,” refining the estimated parameters to a smaller and closer planet of roughly five Earth masses at a few hundred astronomical units, and arguing the clustering remains significant after accounting for observational bias.
- 2021-03Kevin Napier and colleagues publish “No Evidence for Orbital Clustering in the Extreme Trans-Neptunian Objects” in The Planetary Science Journal, reporting that a combined sample from three surveys is consistent with a uniform, unclustered population once the direction of each search is taken into account.
- 2020-2023Wide-field searches, including work with the Dark Energy Survey and Pan-STARRS, rule out portions of the predicted parameter space without finding the planet, narrowing but not eliminating where it could hide.
- 2025-04A team led by Terry Long Phan and Tomotsugu Goto reports a candidate slow-moving far-infrared source by comparing the 1983 IRAS and 2006 AKARI surveys. The candidate requires follow-up to confirm, and Brown himself notes its inferred orbit does not match the Planet Nine his models predict.
- 2025-2026The NSF and DOE Vera C. Rubin Observatory in Chile begins its Legacy Survey of Space and Time. Astronomers describe its deep, repeated all-sky imaging as the most decisive test yet: if Planet Nine exists in the predicted region, Rubin has a strong chance of finding it within the first years of operation.
From the case file
The actual records: declassified, released, or leaked. We link straight to each document in its official archive, so you never have to take our word for it. Read the originals yourself.
A Sedna-like body with a perihelion of 80 astronomical units
The discovery paper for 2012 VP113, which noted that several extreme trans-Neptunian objects share a similar argument of perihelion and first suggested an unseen super-Earth-mass perturber could maintain the pattern. It is the observational and theoretical starting point that Batygin and Brown later developed into the Planet Nine hypothesis.
Read the document: Nature →The planet nine hypothesis
An invited review of the distant solar system's orbital structure and thousands of dynamical models, refining the proposed planet's mass and orbit and arguing the observed clustering remains significant after accounting for observational bias. It is the fullest published statement of the case for the planet, distinct from the claim that it actually exists.
Read the document: ScienceDirect (Physics Reports) →Other case files that cite the same sources
Unresolved. Planet Nine is a genuine, testable scientific hypothesis, not a fringe belief. In 2016 Caltech astronomers Konstantin Batygin and Michael Brown proposed that a distant planet of several Earth masses could explain why a handful of far-flung Kuiper Belt objects share strangely aligned orbits. The documented record is the observed clustering and the published models. The rated claim is narrower: that such a planet actually exists. As of 2026 it has never been directly detected, and reputable astronomers disagree over whether the clustering is real or an artifact of where telescopes have looked. On that specific existence claim the verdict is unproven: a serious prediction awaiting a decisive observational test, not a settled discovery and not a hoax.
Sources
- 1.Evidence for a Distant Giant Planet in the Solar System, The Astronomical Journal (Batygin & Brown) (2016)
- 2.Hypothetical Planet X, NASA Science (2024)
- 3.Caltech Researchers Find Evidence of a Real Ninth Planet, Caltech (2016)
- 4.No Evidence for Orbital Clustering in the Extreme Trans-Neptunian Objects, The Planetary Science Journal (Napier et al.) (2021)
- 5.Planet Nine Could Be a Mirage, Scientific American (2021)
- 6.Astronomers searching for Planet Nine find possible hints of a different planet, Science (AAAS) (2025)
- 7.A Search for Planet Nine with IRAS and AKARI Data, arXiv (Phan, Goto et al.) (2025)
- 8.Vera C. Rubin Observatory Begins Its Long-Awaited All-Sky Survey, Sky & Telescope (2026)
- 9.Is there a 9th planet out there? We may soon find out, National Geographic (2025)
Help us investigate
This is a living case file. If you spot an error or know evidence we missed, tell us, and weigh in on where you land.
Where do you land?
Cast your read on this one.
Comments
Add your take. Comments are read and approved by a human before they appear, so keep it on topic and civil. Please do not accuse named, living people of crimes.