Soph. OT 523
ἀλλ᾽ ἦλθε μὲν δὴ τοῦτο τοὔνειδος τάχ᾽ ἂν
ὀργῇ βιασθὲν μᾶλλον ἢ γνώμῃ φρενῶν.
Perhaps it was a sudden gust of anger
that forced that insult from him, and no judgement.
— David Grene 1942
In the Bristol reprint of the 1897 revised edition of Soph. OT by R.C. Jebb there is a line in middle of a long note on S.OT 523 reading:
We cannot take τάχ᾽ ἂν as = 'perhaps' and take ἦλθε as a simple indic[ative].
I read that comment (note) against Cooper's treatment of the same passage (vol 3, 54.3.10a p.2415). Cooper claims that this is not best read as a vague hypothetical potential scenario, rather a contrary to fact scenario. I don't know what Jebb means as a "simple indicative." Is he suggesting that this indicative is functioning as an optative or … ?
Jebb note from Perseus is from a different edition:
 ἀλλὰ ... μὲν δὴ cp. Soph. Trach. 627.
ἦλθε ... τάχ᾽ ἂν, “might perhaps have come.” ἧλθεν ἂν is a potential indicative, denoting for past time what ἔλθοι ἂν denotes for future time. That is, as ἔλθοι ἂν can mean, “it might come,” so ἦλθεν ἂν can mean, “it might have come.” ἦλθεν ἂν does not necessarily imply that the suggested possibility is contrary to fact; i.e., it does not necessarily imply, ἀλλ᾽ οὐκ ἦλθεν. Cp. Dem. 37.57 “πῶς ἂν ὁ μὴ παρὼν ... ἐγώ τί σε ἠδίκησα;” “how was I likely to do you any wrong?” [This was the view taken in my first edition. Goodwin, in the new ed. of his Moods and Tenses （1889）, has illustrated the “potential” indicative with ἄν （sect. 244）, and has also shown at length that ἦλθεν ἂν does not necessarily imply the unreality of the supposition （sect. 412）. This answers the objection which led me, in a second edition, to suggest that ταχ᾽ ἂν was here no more than τάχα, and that the usage arose from an ellipse （ἦλθε, τάχα δ᾽ ἂν ἔλθοι）. In Soph. OC 964 ff. also I should now take ἦν ... τάχ᾽ ἂν as = “perchance it may have been.”]
This edition appears to clear up the ambiguity in the 2nd edition. I looks to me like Cooper doesn't agree with Jebb. The recent commentary (Dawe on S.OT 2nd ed. 2006) ignores this issue entirely. He probably considered the question settled a long time ago.