There are several counter-arguments against the emphasis on three-act structure. The most thought-provoking may be whether it’s an example of apophenia, the tendency people have of seeing patterns where none exist, of reordering random events to fit a narrative. In this scenario, someone came to the wrong conclusion, and others were influenced by him into believing that there is a meaningful pattern to many narratives.
While many two-hour stories do fit a three-part template (hero introduced, hero develops new goal, hero faces greatest challenge), perhaps it could just as easily be two-act structure, or four-act structure. If a film has five parts, someone trying to determine the structure might decide that the first part constitute act one, the next two parts constitute act two and the finale is act three. In an eight-part film, it could be decided that the first two parts are act one, the middle four are act two and the final two are the third act. In both cases, there will likely be a major change to the narrative at some point in the first quarter, and at some point when there’s about a quarter of the film left. But there are other beats that don’t fit as conveniently into the assumed structure.
This could build on the idea of the self-fulfulling prophecy. Now writers are working on three-act films, even if it’s totally unnecessary. But because they don’t know any better, the hero faces some sort of decision towards the end of the first quarter, and some sort of crisis right before the final challenge, when there’s a quarter of the film left. And since some of these writers will do a good job, the assumption will be that this template worked, and subsequent screenwriters will stick to it.