Welcome to the Speedsolving.com, home of the web's largest puzzle community! You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.
Assume that one layer, say the bottom layer, is solved. Then there are 10 different cases with these probabilities:
Solved case : 1/24
(UF UB)(UL UR): 1/24
(UF UL)(UB UR): 2/24
(UL UB UR) : 4/24
(UR UB UL) : 4/24
(UF UR) : 4/24
(UL UR) : 2/24
(UF UL UR UB) : 4/24
(UF UL UB UR) : 1/24
(UF UR UB UL) : 1/24
Of course the same probabilities if the upper layer is solved instead of the bottom layer. So to calculate the probability of each edpermutation you must multiply the "probability of the U-layer" with the "probability of the D-layer". For example: (UF UL UB UR) (DL DR) has probability (1/24)*(2/24).
Yeah, that's the math that didn't add up for me. I assumed that the 7th case had rotational symmetry and therefore a mirror image (I didn't realize it could be obtained by a rotation)