Cepheids vs. Supernovae: A New Battle over the Hubble Constant ✛
Context
The Hubble constant debate resembles a tug-of-war between the early and late Universe. Planck, looking at the cosmic microwave background, pulls down toward 67.4. SH0ES, relying on Cepheids and supernovae, pulls up toward 73.2. Between them lies a 5.8-sigma discrepancy that cannot be dismissed as chance. This work joins efforts to test whether the error hides in the method rather than in physics. Nearby studies refine the expansion rate using different objects: twin supernovae, gravitational waves, and supernova echoes. All seek a crack in the distance ladder, but this one takes a sideways step: it discards supernovae entirely and examines what remains.
Methods
The authors selected 35 galaxies that host both Cepheids and Type Ia supernovae, but used only the Cepheids and redshifts. They built a Bayesian model that directly predicts the apparent magnitudes of Cepheids from their periods and metallicities, then links distances to redshifts via the Hubble law. Peculiar velocities were accounted for using the Manticore-Local reconstruction, which provides an ensemble of possible velocity fields based on the BORG algorithm. Galaxy selection into the sample was modeled as a cut on supernova apparent magnitude with a smooth threshold. To validate the method, they ran mock catalogs and confirmed that the H0 estimate is unbiased.
Results
Assuming the sample was selected by supernova brightness, the Hubble constant is 71.1 ± 1.4 km/s/Mpc. If selection was by redshift, the value rises to 72.5 ± 1.4. Mixed scenarios yield intermediate results. The central value is 1.3 sigma lower than SH0ES, and the discrepancy with Planck is 2.8 sigma. Compared to a previous analysis of the same data, the uncertainty has shrunk by 41 percent. Using Manticore-Local instead of the older Carrick reconstruction reduced the error by an additional 18 percent.
Implications
This work stands out from neighboring studies by completely discarding the third rung of the ladder, the supernovae. Other studies either refine H0 using supernovae or combine different indicators, but here, for the first time, the same data are analyzed with rigorous Bayesian selection and a modern velocity reconstruction. This allows testing whether supernovae introduce a systematic error. The result shows that even without them, H0 remains high, though slightly lower than in the full SH0ES analysis. Thus, the problem lies not in supernovae per se, but in something deeper, either in the Cepheids or in the physics.
Future development
The method can be extended to other distance indicators: the tip of the red giant branch, Mira variables, and surface brightness fluctuations. With more Cepheid galaxies and improved velocity models, precision could reach one percent. The proposed Habitable Worlds Observatory could add 24 new Cepheid galaxies, enough to achieve the target precision. Further development of Bayesian reconstructions of the local velocity field, such as BORG, will reduce systematics from peculiar motions.
Impact
The result will influence cosmology by refining the local value of H0 and constraining dark energy models. It is also important for calibrating supernovae and understanding systematics in the distance ladder.
Next steps
Include more Cepheid galaxies from other surveys and jointly calibrate different distance indicators. Improve selection models by explicitly defining criteria for future observational programs.
Key open problems
In cosmology, the cause of the Hubble tension remains unresolved. This work tests whether part of the discrepancy stems from systematics in supernova calibration and shows that even without them, the tension persists at the 2.8-sigma level. This narrows the field for new physical explanations.
The knowledge machine's view
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
The study revises the distance to the 35 nearest galaxies using only Cepheids and redshifts, without Type Ia supernovae. A rigorous Bayesian approach and the Manticore-Local model of local velocities are applied. The resulting Hubble constant is 71.1 ± 1.4 km/s/Mpc, which is lower than SH0ES but higher than Planck.
Where its strength is
The strength is that the second rung of the distance ladder alone can achieve competitive precision if selection effects and peculiar velocities are properly accounted for. This allows isolating the influence of supernovae and testing whether they are the source of systematics.
Where it could go next
- It might be interesting to compare the result with independent estimates from gravitational waves, which do not rely on the distance ladder. A nearby paper in the archive measures H0 from neutron star mergers.based on: Gravitational waves measure the expansion rate of the universe arXiv:2605.27227
- Another nearby paper corrects peculiar velocities in a different way, which could help assess systematics from velocity field reconstruction.based on: A Tailwind for Six Runners: How Peculiar Velocities Skewed the Hubble Constant arXiv:2608.06247v1
- One could compare with an approach using supernova echoes for distance measurements to check consistency among different indicators in the local universe.based on: Supernova Echoes: A Fresh Measure of the Hubble Constant arXiv:2606.25205
Worth trying
- Test the peculiar velocity model from paper 2608.06247v1 on the same 35 galaxies and compare how H0 changes.
- Check how including data on twin supernovae from 2512.02391 as a systematic control affects the result.
- Use the Bayesian framework for a joint analysis with gravitational wave data to narrow down H0 without the distance ladder.
Why this may matter
If Cepheids without supernovae yield an H0 higher than the Planck value, it strengthens the case for new physics rather than systematics in the distance ladder. Such an independent test is important for resolving the crisis in cosmology.
Nearby works in our archive
- The Cosmic Tuning Fork: Why the Universe Plays Two Different Notessimilarity 0.58
- Refining the Expansion Rate of the Universe with 'Twins' Among Type Ia Supernovaesimilarity 0.57
- A Tailwind for Six Runners: How Peculiar Velocities Skewed the Hubble Constantsimilarity 0.54
- Supernova Echoes: A Fresh Measure of the Hubble Constantsimilarity 0.53
- Gravitational waves measure the expansion rate of the universesimilarity 0.53
- Neutrinos Reveal the Universe's Expansion Ratesimilarity 0.51
The archive is sparse around this work: the closest review is only 0.58, against a usual 0.64. It looks like few have drilled nearby.
Cepheids pulsate with periods ranging from days to months, and their period reveals their true luminosity. This law was discovered by Henrietta Leavitt in 1912 while working with photographic plates of the Magellanic Clouds.
In Carl Sagan's novel 'Contact', the protagonist uses Cepheids to calibrate distances when searching for a signal from space. Here, Cepheids also serve as calibration, but for testing cosmology itself.
Key numbers
- H0 (SN selection): 71.1 ± 1.4 km/s/Mpc
- H0 (redshift selection): 72.5 ± 1.4 km/s/Mpc
- Uncertainty reduced: 41%
- Tension with Planck: 2.8σ
- Number of Cepheid galaxies: 35