Which, incidentally, is probably a better theory than dark matter. For example it can produce the same results without the problem of undetectable matter.
On the other hand, determining the local time dilation factor based on all mass beyond the local area is essentially not possible. We can talk about how the great voids have less time dilation than galaxy clusters, sure. But what about the universe as a whole? Our universe could be embedded in a larger one that contributes to time dilation in ours and we could never sense that. Time dilation at cosmological scales is relative for this reason.
> that we literally can't calculate it (the age of the universe) anymore
crash with our observation of the CMB?
I don't see how us being unable to calculate a quantity from one set of observations could possibly clash with another set of observations (the cmb).
What am I missing?
However, other things are suggesting we might not be seeing the whole universe just by looking as far away as possible. It could be we can see some regions on the CMB that have already expanded outside of our observable universe. These regions aren't just "even fainter and we need a better telescope", they're "the last photon from that region that will ever reach us came and left billions of years ago."
Therefore, there might not be one singular hubble constant, there might be two. One that applies to our local observable universe, and one that applies to the entire universe.
It could be that the universe is 26 billion years old: https://academic.oup.com/mnras/article/524/3/3385/7221343?lo...
And because at great enough distances(/times), expansion is faster than light, and we simply can't see a significantly different epoch of the universe just by looking deeper.