It seems, that this possible problem already came up when __build_class__ got implemented, see issue1681101. The second and third version of the patch at this issue (file7874 and file7875) contains a comment: "XXX Should we do the "winner" calculation here?". But the next version, which contains the C implementation of __build_class__ still uses simply the first base. The "winner calculation" probably refers to the algorithm determining the proper metaclass in lines 1950-1976 of typeobject.c (in type_new).