Comparison of the H₀ measurements by the TDCOSMO collaboration (and its predecessor H0LiCOW) in chronological order. The measurements at the top by Wong et al. (2020) and Millon et al. (2020b) are “assertive” in the definition of Treu et al. (2022). They are based on power-law and composite models to describe the mass density profile of the deflector galaxies, thus implicitly breaking the MSD and obtaining ∼2% precision. The TDCOSMO-4 results from Birrer et al. (2020) are “conservative” (Treu et al. 2022). They are based on the same data but obtain larger uncertainties by introducing the internal MSD parameter λ_int, and constraining it with unresolved stellar kinematics. The Birrer et al. (2020) measurement in combination with SLACS is shown as a dashed line for historical purposes but should not be used anymore, as it was later discovered that the stellar velocity dispersions based on low SNR (∼9 Å⁻¹) spectra from SDSS suffer from systematic errors and covariance (Knabel et al. 2025a). The new measurements presented in this work are shown in the bottom panel. They are “conservative” in terms of uncertainties and constrain the MSD using new stellar kinematics based on high SNR JWST-NIRSpec, VLT-MUSE, and Keck-KCWI spectra, as well as an improved methodology (Knabel et al. 2025b).