The theoretical factors influencing human female facial attractiveness – symmetry, averageness,and sexual dimorphism – have been extensively studied. However, through improved methodologies, recent studies have called into question their links with life history and evolutionary utility. The current study uses a range of statistical and methodological approaches to quantify how important these factors actually are in perceiving attractiveness, through the use of novel analyses and by addressing methodological weaknesses inherent in the literature. Study One examines how manipulations of symmetry, averageness, femininity, and masculinity affect attractiveness using a two-alternative forced choice task, revealing that increased masculinity is unattractive, but increased femininity is not, and large effects observed for averageness. Study Two utilises a naturalistic ratings paradigm, finding similar effects of averageness and masculinity, but no effects of femininity and symmetry on attractiveness. Study Three applies a random forest machine learning algorithm and geometric measurements of the factors from faces to predict perceived attractiveness, finding that averageness and dimorphism are useful features capable of relatively accurate predictions. However, the factors do not explain as much variance in attractiveness as the wider literature suggests. Implications for future research on attractiveness are discussed.