I'm going to try and give an explanation for why, with everything else being equal, reversing the direction of rotation of the brake cam does make a difference to the effectiveness of a cam driven single leading shoe brake.
One working edge of the cam is further from the pivot of the shoes than the other working edge of the cam. It is the outer edge that is further. The outer working edge bears against the outer section of the shoe rubbing pad. This section of the rubbing pad is further from the shoes pivot than the contact section for the other shoe rubbing pad. This means that a rotating force on the camshaft will provide greater force against the drum for one shoe than the other shoe.
On a single leading shoe brake during braking, the geometry of the shoe pivot and the drum is such that one shoe is pulled against the drum and the other shoe is repelled. The one that is pulled in is called the leading shoe and the other is called the trailing shoe. If the direction of rotation of the drum is reversed, then the shoes swap roles.
It is a small effect, but if the bike has the cam rotation direction such that the leading shoe has the smaller of the two forces against the drum, then reversing the rotation direction will increase the force applied to the leading shoe which should make the brake more powerful, with everything else the same.
It's interesting to note which way bikes have the cam rotation which way as standard