If you've followed cosmology news, you've probably heard about the "Hubble tension" - the puzzling fact that two very reliable ways of measuring how fast the universe is expanding give different answers. One method, looking at the cosmic microwave background (CMB - the leftover glow from the Big Bang), suggests the Hubble constant (H₀) is about 67 km/s per megaparsec. The other, using nearby supernovae and a step-by-step "distance ladder," gets around 73 km/s/Mpc. That's a difference of roughly 8-10%, and it's been driving cosmologists nuts for years. This short paper suggests a possible way out, without needing exotic new physics. It builds on work by Thomas Buchert and collaborators showing that the universe might not be infinite but instead have a 3-torus topology - a finite, looped shape about 3-4 times bigger than what we can see. In such a universe, there's a natural limit to the longest possible wavelength fluctuations in the CMB. I propose that this creates a subtle split: the CMB measurement (which relies heavily on frequency shifts across the whole sky) gets biased lower when we analyze it assuming an infinite universe, while local measurements (tied more to direct wavelength and distance calibrations on smaller scales) see the "true" higher rate. Both are correct - they're just looking at different parts of the same cosmic picture.
George Maise (Fri,) studied this question.