There's an implication in your argument that the science posted in the journal is of a good enough quality to be published - otherwise why wouldn't you apply the same standard you are applying here? Doing a bad job isn't limited to just reproduction.

Considering the reproducibility crisis, is it your position that people are worse at reproducing research than they are at conducting it? An alternative interpretation would be that people have more incentive to publish hastily-made "failed to reproduce" papers than they have "findings" papers, but I don't think that's the case either, especially considering the reply from Science to this case.