If you check out some of the references in the Neilsen, Guffanti, Sarkar (NGS) paper, you will find
that their claim is not new (see NGS refs 13, 19). For instance Jun-Jie Wei and Xue-Feng Wu
located at Purple Mountain Observatory in mainland China make
exactly the same claim using a different supernova data set (SLS)
Wei & Wu's SLS data set contains 252 SNs and the NGS data set (JLA) contains 740 supernovas.
What these and several other papers seem to be saying is that both the data and the corrections
to the data seem to depend on the model you assume. If you calculate both Dark-Matter Universe (DMU)
and Hubble Constant expansion Universe (HCEU), you can fit them BOTH to the facts.
However the more facts you add the more you have to fudge the DMU predictions. Astonishing that you can fit
the same data with radically different assumptions.
However, as the various authors point out, given a theory with one free parameter (HCEU) and a theory
with 6 free parameters (DMU), the theory with fewest free parameters is favored
Occam's Razor rules.
It would be interesting to see what the response of Perlmutter et al is to these recent publications that seem
to cast their prize-winning work into serious doubt.
I don't like to jump to conclusions until the other side has had its say. So far I have not seen any criticism of these
recent results. Please alert me, Jack, if you see a publication by Perlmutter and his dark energy colleagues that
claims to blow Wei, Wu and Sarkar out of the water.