It's not just the CMB, it's the CMB + LCDM. You know, that theory that posits a form of matter that we have consistently failed to find and that has technically been falsified a number of times, but which was then adjusted to work around those problems, and yet problems still remain. You might have "slight" bit of unaccounted for error sneaking in there.
I personally wouldn't trust the CMB calculation at all until the dark matter issue is resolved.
Isn’t there loads of evidence for dark matter including direct evidence like gravitational lensing? We’re just not sure what it is, though primordial black holes look like a strong candidate.
> Isn’t there loads of evidence for dark matter including direct evidence like gravitational lensing?
There is an equivalent boatload of evidence for modified gravity that can't be explained by particle dark matter [1], but one set of observations is hand waved away as a minor inconvenience while the other set is taken as definitive proof.
If gravity doesn't work the way we expect, then lensing in galaxy collisions isn't necessarily telling us what we think it is.
But setting that aside, if we take even the most naive modified gravity, MOND, then lensing does require a form of particle dark matter, like sterile neutrinos, but the total amount of dark matter required to explain the evidence is only double the amount of visible matter, rather than almost 10x the amount of visible matter as in LCDM. This has significant implications for calculating the Hubble constant from the CMB [2].
I personally wouldn't trust the CMB calculation at all until the dark matter issue is resolved.