copyists errors (called glosses)
Um, “glosses” are not copyists’ errors: they are editorial explanations of obscure elements of a text. John 1:41’s “the Messias, which is, being interpreted, the Christ” is a gloss.
They can precisely date a writing, usually to within a decade. They can tell, give or take fifty miles, where the author learned to write. By studying word usage and frequency they can tell if two works were written by the same person (which is how we know beyond all reasonable doubt that the guy that wrote Romans did not write Hebrews).
Sorry, but it is nowhere near that precise, or that simple. When a date for an ancient text is derived from internal evidence alone (i.e. from the form of language which it uses), the margins of error for time and space are often
measured in centuries and whole kingdoms (and the space one is even worse for widely-propagated languages, like Latin). As for word usage being proof of the same or different authorship, it is subject to the most basic of analytical problems: limited sample sizes make
author identification difficult or even impossible.
While stylometry works reasonably well for consistent, prolific, modern writers, life becomes much more complicated when dealing with more experimental but less fruitful authors, even when they are quite recent.
Studies of known authorship demonstrate that one author can have a range well beyond that of another (fig 1 - meaning that using the latter as a guide for “maximum plausible variety” would lead to disqualifying texts from the former’s catalogue of works), or have an individual range greater than the distance between that author and another (fig 3 - meaning that the works of one author could be misattributed to the other). Further, even
good usage of it fails to quell motivated doubts.