I recently watched a video done by Dr. Michael S. Heiser that reviewed so-called "exegetical fallacies." Of course, the scholar, who put such terminology on the map was D.A. Carson--the author of a book with that title.
Exegesis (especially used in the biblical sense) means to explain, interpret or unfold a text (see John 1:18); the act of exegesis involves drawing meaning from (out of) a text, not reading sense/meaning into it. Exegetes base their explanations on close readings of textual sources in their respective biblical language, whether that language is Hebrew-Aramaic, Greek, Latin or other. Nevertheless, like any task, there are right ways to undertake exegesis: bad exegesis usually entails committing exegetical fallacies or one sort or another. In logic, "fallacy" is normally defined as "a mistake in reasoning." E.g., argumentum ad hominem.
Other definitions for fallacy include:
a) a mistaken belief, especially one based on unsound argument.
b) a failure in reasoning that renders an argument invalid.
c) faulty reasoning; misleading or unsound argument.
Two kinds of exegetical fallacies are grammatical and presuppositional/historical fallacies. For example, to understand the Greek aorist as the "action that happens once in the past" tense is fallacious, since many uses of the aorist do not fit this definition. Furthermore, aspect research now suggests that aorists are default Greek verbal forms; and they likely portray action as a whole. See Matthew 4:9; Philippians 2:12; 1 John 2:1-2.
Historical fallacies might occur when exegetes try to reconstruct historical circumstances for Gospels or NT Epistles. We do not have access to precise historical conditions for the ecclesiae at Corinth, Philadelphia or Pergamum, apart from NT records. So while reconstructions of history might be interesting, they can lead to fallacies when imposed on the text from outside. But many other traps await interpreters of Scripture.