Here's an orthogonal idea which would make a great project - scrape IMDB for all creative roles in the credits - actors, directors, writers, set designers, make up, music, etc. -- all of them. Regress them against budget, revenue, margin and revenue/budget.

Importantly it would allow for groundbreaking and genre bending movies to be assessed.

See who's adding value and who's not. And see how well that predicts your outcome.