A Number That Should Be Simple
Imagine two teams of surveyors tasked with measuring the height of the same mountain. Both teams are meticulous. Both use well-tested equipment. Both produce results they trust completely. And yet, when they compare notes, their numbers disagree by enough to matter. Not by a rounding error. By a gap that has only grown wider the more carefully each team has worked.
That is, in rough outline, the situation confronting cosmology right now. The number in dispute is called the Hubble constant, denoted H₀ and pronounced “H-naught.” It describes how fast the universe is expanding: specifically, how much faster a galaxy recedes from us for every additional megaparsec of distance between us and it. One megaparsec is roughly 3.26 million light-years. The Hubble constant is typically expressed in kilometers per second per megaparsec, and the two main ways of measuring it are producing answers that differ by somewhere between 5 and 10 percent, depending on which datasets you compare.
That gap is the Hubble tension. It sounds technical, even arcane. But its implications are anything but. If the discrepancy is real and not a product of systematic error, it means that our standard model of cosmology, the so-called Lambda-CDM model built on cold dark matter and a cosmological constant representing dark energy, is missing something fundamental. It may mean the universe contains physics we have not yet imagined.
Two Roads to the Same Number
To understand why physicists are losing sleep over this, you need to understand where the two competing measurements come from.
The first method is called the cosmic distance ladder. It works by chaining together overlapping distance measurements across increasing scales. The first rung uses parallax: the tiny apparent shift in a nearby star’s position as Earth orbits the Sun, which gives a direct geometric distance. The second rung uses Cepheid variable stars, which pulsate at rates directly linked to their intrinsic brightness. By knowing how bright a Cepheid truly is and comparing that to how bright it appears, astronomers can calculate how far away it is. The third rung uses Type Ia supernovae, thermonuclear explosions of white dwarf stars that reach a fairly consistent peak brightness, making them useful as what astronomers call “standard candles” across truly cosmic distances. Attach redshifts to those supernovae and you can measure how fast the universe is expanding locally.
The SH0ES (Supernova H₀ for the Equation of State) team, led by astronomer Adam Riess of Johns Hopkins University, has refined this ladder over many years. Their most recent measurements, incorporating data from the James Webb Space Telescope, peg H₀ at approximately 73 kilometers per second per megaparsec, with an uncertainty of about 1 percent.
The second method is entirely independent. It starts from the Big Bang itself. In the early universe, sound waves propagated through the hot plasma of matter and radiation, leaving a characteristic imprint on the distribution of matter called baryon acoustic oscillations (BAO). Those imprints, frozen into the cosmic microwave background (CMB), the faint afterglow of the Big Bang, act as a “standard ruler.” By measuring the angular size of temperature fluctuations in the CMB, and feeding those measurements into the Lambda-CDM model, physicists can calculate what the Hubble constant should be today.
The Planck satellite collaboration, using observations of the CMB made between 2009 and 2013, derived a value of approximately 67.4 kilometers per second per megaparsec. The difference between 67.4 and 73 may not sound dramatic in everyday terms. In physics, with uncertainties of about 1 percent on each side, it represents a disagreement at the level of roughly 5 sigma. In particle physics, a 5-sigma discrepancy is the threshold for claiming a discovery. Here, it signals a potential crack in the foundations of the standard model.
The Systematic Error Problem
Before you can claim the universe is broken, you have to rule out the possibility that the measurements themselves are broken. Physicists are, by temperament and training, deeply suspicious of their own results. The history of science is littered with confident announcements that turned out to reflect instrumental artifacts or analytical assumptions rather than real phenomena.
The low-H₀ camp, anchored in CMB observations, has a potential vulnerability: it is not measuring the expansion rate directly. It is inferring H₀ by fitting the Lambda-CDM model to the CMB data. If the model is subtly wrong, the inferred Hubble constant will be wrong too, without anyone’s measurement being in error per se. The Planck team has searched exhaustively for systematic errors in their instrument calibration, foreground subtraction, and statistical methods. So far, no smoking gun has emerged.
The high-H₀ camp has a different vulnerability: the cosmic distance ladder involves multiple steps, and errors can compound. Cepheid measurements in particular rely on calibrations that themselves depend on other assumptions. Critics pointed for years to possible contamination of Cepheid samples by crowded stellar fields, a concern that the James Webb Space Telescope was specifically positioned to address. When JWST confirmed and sharpened the Cepheid-based distances in 2023 and beyond, the high value of H₀ did not budge downward. Riess and colleagues published results showing that Hubble Space Telescope and JWST measurements of Cepheids are consistent, and the tension remained firmly in place.
Alternative distance indicators have entered the conversation as cross-checks. Surface brightness fluctuations in elliptical galaxies, the tip of the red giant branch (TRGB), and gravitational wave “standard sirens” from neutron star mergers all offer independent rungs on the ladder. Results from TRGB measurements have at various times come in slightly below the SH0ES value, closer to 70 or 71 kilometers per second per megaparsec, which some researchers interpreted as a possible sign that systematic errors were at least partially responsible. But even 70 to 71 sits in uncomfortable tension with 67.4. The discrepancy has shrunk in some analyses and remained stubborn in others, but it has not gone away.
What Physics Could Explain It
If the tension is real, the universe needs new physics. Cosmologists have not been shy about proposing it.
One broad category of solutions targets the early universe. The CMB-based inference of H₀ is sensitive to the size of the sound horizon at the time of recombination, roughly 380,000 years after the Big Bang, when electrons and protons first combined into neutral hydrogen and the universe became transparent. If something exotic increased the expansion rate just before recombination, it would shrink the sound horizon, and the CMB-inferred Hubble constant would be revised upward. This idea goes by the name “early dark energy”: a hypothetical form of energy that was briefly important in the early universe before fading away. Several groups have proposed specific early dark energy models. So far, none has found clear confirmation in CMB or large-scale structure data, and some analyses find that solving the Hubble tension this way tends to worsen the fit to other observations.
Another category modifies dark energy itself. In the standard Lambda-CDM model, dark energy is a cosmological constant: a fixed energy density of empty space that drives accelerating expansion. Its equation-of-state parameter, w, equals negative one. If dark energy is instead a dynamical field, w could be different from negative one, or could evolve over time. Some analyses of large-scale structure and supernova data have hinted at deviations from w = negative one, though the evidence has waxed and waned across different datasets.
In 2024 and again in March 2025, the Dark Energy Spectroscopic Instrument (DESI) collaboration released results from their massive spectroscopic survey of galaxies and quasars. Their baryon acoustic oscillation measurements, combined with CMB and supernova data, suggested that dark energy might not be a simple cosmological constant after all. The 2025 results put the preference for evolving dark energy at roughly 2.8 to 4.2 sigma, depending on which supernova sample is used. That is short of the 5-sigma discovery threshold, but it attracted serious attention. Whether evolving dark energy would do anything about the H₀ tension is a separate and still open question, since the DESI results are about the expansion history rather than a direct measurement of the local expansion rate.
Other proposals include modifications to general relativity at cosmological scales, the possibility that dark matter has interactions beyond pure gravity, or the existence of additional relativistic particle species in the early universe. Neutrino physics has entered the frame: extra “sterile” neutrino species could, in principle, affect the sound horizon in ways that shift the inferred Hubble constant. Each proposal comes with its own set of predictions and constraints.
Why It Matters Beyond the Numbers
Cosmology is sometimes portrayed as a purely academic pursuit, a beautiful but distant enterprise with no practical stakes. The Hubble tension challenges that framing.
The Hubble constant is not merely a number about a number. It encodes the age of the universe (a Hubble constant of 67.4 implies an age of about 13.8 billion years, while 73 would imply somewhat less), the large-scale geometry of space, and the ultimate fate of cosmic expansion. More fundamentally, the tension is forcing a reckoning with what we mean when we say we understand the universe.
Lambda-CDM has been spectacularly successful. It predicted the detailed structure of the CMB before those structures were measured. It correctly predicted the statistical distribution of galaxies on large scales. It accommodated the discovery of accelerating expansion in 1998 through the cosmological constant. It is, by most measures, one of the best-tested models in cosmology. If the Hubble tension breaks it, the successor theory will have to explain everything Lambda-CDM explained, plus the new discrepancy.
That is a high bar. But it is also how science progresses. The tension between Newtonian mechanics and the measured orbit of Mercury was once considered a curiosity. It turned out to point toward general relativity. The ultraviolet catastrophe in classical physics, the divergence between theory and observation in blackbody radiation, seemed technical and obscure until it unlocked quantum mechanics. Physicists who have spent careers on Lambda-CDM are understandably cautious about declaring it broken. But they are also, by now, taking the Hubble tension with complete seriousness.
Wendy Freedman of the University of Chicago, who has led multiple independent calibrations of the distance ladder over her career, and her collaborators have used the TRGB and other indicators, including JWST observations, to arrive at intermediate values near 70 kilometers per second per megaparsec. They argue that these results are consistent with the CMB value, while other researchers counter that they sit in the middle of the tension rather than resolving it. The dispute over which calibration to trust is part of what keeps the tension unresolved.
Where the Search Stands
The investigation is entering a new phase. Several major observational programs are either underway or recently completed, each designed to sharpen one or more rungs of the distance ladder or constrain the early-universe physics.
DESI, mounted on the 4-meter Mayall Telescope at Kitt Peak National Observatory, completed its originally planned five-year survey in 2026, having mapped more than 40 million galaxies and quasars. Its BAO measurements provide an independent check on both the sound horizon and the late-universe expansion history. The Vera C. Rubin Observatory in Chile, after years of construction and commissioning delays, began its Legacy Survey of Space and Time (LSST) in June 2026, which is expected to discover large numbers of Type Ia supernovae and monitor huge numbers of variable stars, providing new data at unprecedented volume.
JWST continues to be a workhorse. Its near-infrared sensitivity cuts through the dust that complicated Hubble Space Telescope observations of Cepheids in distant galaxies. Each new galaxy added to the JWST Cepheid sample is further evidence against the crowding-bias hypothesis, though Freedman’s group disputes the SH0ES calibration.
Gravitational wave astronomy offers perhaps the most exciting independent probe. When the LIGO, Virgo, and KAGRA detectors observe a neutron star merger that is also seen in light by electromagnetic observatories, the gravitational wave signal gives a direct distance measurement requiring no calibration chain. The electromagnetic counterpart gives the redshift. Combining the two gives H₀ with no rungs to climb. The first such measurement, from the event GW170817 detected in 2017, gave H₀ near 70 kilometers per second per megaparsec, with wide uncertainties. As more events accumulate, that uncertainty will shrink. If the gravitational wave H₀ converges clearly toward either 67 or 73, it will be powerful evidence.
The Simons Observatory, a new CMB telescope in the Atacama Desert in Chile, will measure the polarization of the cosmic microwave background with far greater precision than Planck. It will either confirm the low value with tighter error bars, further sharpening the tension, or reveal previously hidden features that change the analysis.
The Long View
The Hubble tension has been debated for more than a decade. It has survived the scrutiny of dozens of independent teams, multiple instruments, and the arrival of JWST. As of late 2026, no consensus resolution has emerged. The community is divided not between believers and skeptics of the tension itself, which is now broadly accepted as real, but between those who think refined systematics will ultimately close the gap and those who think new physics is already peeking through.
The honest answer is that we do not yet know which camp is right. What we do know is that the universe is presenting us with a measurement that does not fit neatly into our best theory. That is, historically speaking, the most productive situation in physics. Every crisis of this kind has eventually resolved into a deeper understanding.
For the astronomers and cosmologists spending their careers on this problem, the tension is not a source of despair. It is an invitation. The universe has always been larger, stranger, and more generous with surprises than any model we have built to describe it. If the Hubble tension is pointing toward something genuinely new, the field that discovers what that something is will have earned its place in the long story of humanity trying to measure the sky.