Tension-Dependent Variance Suppression in a Composite Hubble Parameter

We analyze a curated set of contemporaneous local and early-universe measurements of the Hubble constant to study how statistical tension between datasets affects parameter inference. Defining a composite parameter �, we find that while the mean value of � remains stable across tension regimes, its variance exhibits strong, statistically significant suppression as inter-dataset tension increases. Using variance ratio tests, robust non-parametric methods, correlation analysis, and permutation tests, we show that this variance suppression is unlikely to arise from random fluctuations. The results are consistent with tension-dependent regularization in joint cosmological inference, rather than measurement error or new physical effects. This work reframes aspects of the Hubble tension as a property of the inference process under dataset conflict and provides a diagnostic framework for assessing stability and uncertainty in composite cosmological parameters.