Hubble Tension or Distance Ladder Crisis?

Abstract

We present an up-to-date compilation of published Hubble constant ($H_0$) measurements that are independent of the CMB sound horizon scale. This compilation is split in two distinct groups: A. Distance Ladder Measurements sample comprising of 20 recent measurements, primarily from the past four years, utilizing various rung 2 calibrators and rung 3 cosmic distance indicators. B.One-Step Measurements sample including 33 measurements of $H_0$ that are independent of both the CMB sound horizon scale and the distance ladder approach. These 33 measurements are derived from diverse probes such as Cosmic Chronometers, gamma-ray attenuation, strong lensing, megamasers etc. Statistical analysis reveals a significant distinction between the two samples. The distance ladder-based sample yields a best fit $H_0=72.8\pm 0.5$ km s${}^{-1}$ Mpc${}^{-1}$ with $\ddot{I}{‡}^2/dof=0.51$ indicating some correlations. The one-step measurements result in $H_0=69.0\pm 0.48$ km s${}^{-1}$ Mpc${}^{-1}$ with $\ddot{I}{‡}^2/dof=1.37$ indicating some internal tension. If two outlier measurements are removed (TDCOSMO.I-2019 known to have systematics and MCP-2020) the best fit of the one step sample reduces to $H_0=68.3\pm 0.5$ km s${}^{-1}$ Mpc${}^{-1}$ with $\ddot{I}{‡}^2/dof=0.95$, fully self-consistent and consistent with sound horizon based measurements. A Kolmogorov-Smirnov test yields a p-value of 0.0001 suggesting that the two samples are fundamentally distinct, with a probability of less than 0.01\% that they are drawn from the same underlying distribution. These findings suggest that the core of the Hubble tension lies not between early and late-time measurements, but between distance ladder measurements and all other $H_0$ determinations. This discrepancy points to either a systematic effect influencing all distance ladder measurements or a fundamental physics anomaly affecting at least one rung of the distance ladder.

44 pages, 4 figures 3 tables. Updated data with three additional H0 measurements (total 53). Expanded Table 3 and the $\ddot{I}{‡}^2$ considering additional subsamples further testing the robustness of results and conclusions. Comments added, typos corrected, references added. The numerical analysis files used in this work are publicly available at https://github.com/leandros11/distladder/