Roger's article is confined to center and average sharpness across the frame. And if that had been the metric I had used, indeed most of the eight GM 12-24 zooms I have tried would have passed. But that article doesn't talk about 'worst corner' sharpness. If you look at worst-corner sample variation you will see a large increase in the number of bad copies, as we noted in the thread related to that article.
My comments in the current thread about the substantial corner-sharpness sample variation with the GM are very much in line with how I shoot, rather than looking for flaws. Corner sharpness matters to me because I focus stack for sharp details everywhere in the frame, and I like to have the option to print large (I wouldn't own an A7r4 otherwise). I also shoot astro, so corner performance wide open is also relevant.
My impression comparing 8 copies of the 12-24 GM with 4 Sigma 14-24 zooms is that the Sony has larger corner-sharpness variation, even if I exclude focal lengths below 14mm. And the Sigmas I tried all cleaned up very well by f6.3, which has not been the case with the GM.
p.s. when looking at centering with the GM at its wide end, it's highly recommended to use corner focus due to field curvature.