That grades are a better indicator of law school success than the LSAT (no shit?) doesn't change the fact that transfer students provide a solid empirical foundation that shreds the premise of this article.
Point 1: Student X is at a T50, at the top 5%. Transfers to a T6, stays in the top 5-10%, or even improves to top 1%.
Without evidence to the contrary, it is more reasonable to assume, from point 1, the following:
Point 2: Student Y is at a T50, at median. In alternate-world where the median student would be able to transfer to a T6, Student Y would be at median;
than it is to assume what the article assumes, i.e.:
Point 3: Student Y is at a T50, at median. In alternate-world where the median student would be able to transfer to a T6, Student Y would be significantly below the median.
Annecdotally, Point 1 is true for far, far more transfer students than any variation (i.e., student transfers, falls to the top 20% or falls to median). Would be interesting to see an actual study done on that; but the article's failure to do so, in the face of a large amount of annecdotal evidence that supports the idea that transfer students do not, in fact, slip in their rank, renders the premise of the article highly suspect.
Edit: Also, you're off your rocker if you think large firms do any appreciable hiring outside of OCI. They don't. HTH. If a firm isn't interviewing at your school, you are not going to get a job there.