The whole point of the article that we're commenting on is that it might not be matter. The root problem is that the observed doesn't line up with the expected. Is the problem with the observed or with the expected? We don't know. You're assuming the problem is with the observed. This article is saying we can modify our expected values with a tweak to the model.

We should name the phenomenon after the problem, not after one of the not-yet-proved-but-expected-to-be solutions.