Since Syd Field published his book Screenplay in 1979, the three-act structure has become a dominant tool for analysing screenplays and films. While the formulaic nature and constricting effects of this ‘paradigm’ have often been criticized by adherents of independent films, until now its explanatory suitability to classical and mainstream filmmaking has rarely been questioned as such. After commenting on the history of the concept, the article examines how well the model works in practice by statistically evaluating a large number of studies by different authors. This comparison reveals that there is a surprisingly low consensus on how to divide films into three acts. The discrepancies seem fundamentally due to the imprecision and vagueness of the paradigm’s core ideas, and thus neither the functions ascribed to acts nor the concept of plot points appear to be valid tools for ensuring consistency in analysis. Hence mainstream film needs to be seen as more complex – and structural analysis more a matter of interpretation – than the model allows for.