I would like to add a perspective that partially supports what Peter and the Tyndale House GNT is doing.
The whole enclitic and accent system has been artificially regularized and imposed on our written NT. For example, the genitives αὐτοῦ αὐτῆς αὐτοῦ are required to have an accent in the regularized system. However, as a reader becomes sensitive to word order they will begin to perceive that a word like αὐτοῦ is often moved forward and behind a dominating (not a grammatical head word but a clause-level main word) exactly like enclitics. If an enclitic were to move to the same place we might recognize it as reinforcing the item to which it is enclitic. We don't usually recognize this with αὐτοῦ because it is not enclitic "by definition." My claim is that that definition is artificial and does not recognize how the language was being used.
A second example of this phenomenon of an artificially imposed accent system is the βαρύτονος grave accent. Basically, the grave accent marks a potential, suppressed, acute accent. That works fine in many contexts where a word final acute accent is suppressed (written as grave and not pronounced) in order to allow a longer string of words to flow together as a phonological phrase. However, that artificial rule was almost certainly not applied in speech to words that were fronted and placed immediately before a verb to mark a/the salient point of information in the clause/phrase. Such a fronted, focal item almost certainly involved an ever-so-slight pause and would have retained its acute accent in real speech. This has been discussed in this forum in the past but it is rarely or not-at-all discussed in most Greek teaching so let me use the closest contextual example as an illustration.
Matt. 5:25 ἴσθι εὐνοῶν τῷ ἀντιδίκῳ σου ταχύ, ἕως ὅτου εἶ μετ᾿ αὐτοῦ ἐν τῇ ὁδῷ, μήποτέ σε παραδῷ ὁ ἀντίδικος τῷ κριτῇ καὶ ὁ κριτὴς τῷ ὑπηρέτῃ καὶ εἰς φυλακὴν βληθήσῃ·
In Matt 5.25 the σε is placed in front of the verb and can and should be read as a marked Focus. It is without an accent and humorously (I speak to Greek grammarians) treated as an enclitic to μήποτε. In real speech I have no doubt that it was fully accented, probably with an ever-so-slight pause and increase in volume and/or pitch. However, as an enclitic this does not fully illustrate the point discussed above about grave accents hiding the real pronunciation.
Later in 5.25 we see καὶ εἰς φυλακὴν βληθήσῃ. The phrase εἰς φυλακὴν has been fronted as marked, salient information, in order words, as a Focus structure. The editors have marked φυλακὴν with a grave according to the regularization rules. However, as a marked Focus it was almost certainly intended to be read and spoken as καὶ εἰς φυλακήν βληθήσῃ, with full accentuation on the syllable κήν.
What is my point in all of this? That the accent rules were an artificial application to written texts that did not truly reflect the pragmatics of actual speech. I have no doubt that they represent default speech patterns in Greek from the ancient period (classic and second sophistic). But real speech undoubtedly differed from plain-Jane, milk-toast communication. (Apologies to all the extraordinary Jane's in the world.) When the word order patterns in Greek are understood, and when the natural Focal patterns of many many languages are considered, then it is highly probable that perceptive readers of Greek texts read Focal words with grave accents as if they had a focal acute accent. And words like αὐτοῦ, when used as an enclitic to reinforce a main word/phrase, would be pronounced as if they did not have an accent.
My take home assumption/rule-of-thumb: τίς and τις were not accented as simple acute and "enclitic" according to lexeme but were pronounced according to their function in the sentence and in many cases "anyone" had a normal accent, while an enclitic phrase like ἄνθρωπός τις would not. Of course, in many contexts "anyone" may have had no accent, just like an acute is suppressed and marked by a grave as a "potential accent" in case of slowing down. And the rule was/became to write "anyone" without any accent as if it was always an enclitic, even though it wasn't. On the other hand, τίς ῾who?' always carried its accent, and resisted a grave accent, as a special category as an inherently Focal item.