> AFAIK the previous models that all assumed that Type 1a supernovae were not affected by the age of the progenitor stars had no actual analysis to back that up; it was just the simplest assumption.

Why would you assume this? It's not correct.

Type 1a supernovae aren't even assumed to be "standard candles" as is often claimed: rather, they're standardizable, i.e. with cross-checks and statistical analysis, they can be used as an important part of a cosmological distance ladder.

A great deal of analysis has gone into the development of that distance ladder, with cross-checks being used wherever it's possible to use them.

They look at surface brightness fluctuations in the same galaxies, Tully-Fisher distances[1], tip of the red giant branch distances[2], and even baryon acoustic oscillations[3]

Is it possible that this one single paper has upended all that? Theoretically. Is it likely? No.

[1] https://en.wikipedia.org/wiki/Tully%E2%80%93Fisher_relation

[2] https://en.wikipedia.org/wiki/Tip_of_the_red-giant_branch

[3] https://en.wikipedia.org/wiki/Baryon_acoustic_oscillations