Academic studies in recent decades have generally failed to find any conclusive evidence to support studio bosses’ faith in stars’ pulling power. Our own analysis suggests only that a few of them do add a bit to box-office receipts. Number-crunchers at Epagogix, a company in London, use an algorithm to project box-office takings of films based on their story elements—including the use of special effects, a surprise ending or a cool location. And they reckon that as long as the stars look good and can act, they make scant difference, with at best a very few exceptions. It helps to have a damsel in distress, but it does not really matter which damsel.
Source: Fading stars | The Economist