Mates Question

Try using parallel or angle relationship first to orient the leg, then use distance relationship to a point or vertex on the leg.

Sounds like something is slightly out-of-square.

A surface mate eliminates 3 dofs: 2 axial and 1 linear, leaving 2 linear one axial dof. The edge mates you mention still leave one linear dof, unless edge mates also constrain length. But, no, you shouldn't be over constrained. Something else is going on. Why don't you just try mating surfaces instead of edges. Offsets are still possible.

If he is using a distance mate from one of the edge faces then he may be trying to remove one linear and one rotational degree of freedom. Since that rotational dof was already removed with the surface mate he is over constraining which may not become apparent until the second edge distance mate it attempted.

Try edge distance mates to a point on the end of the leg.

You mated the Top surface of leg to Bottom surface of table. Did you make sure the leg edges are mated to the Bottom edges of the table (not the Top edges of table). The description you gave "Now I try to mate the edges of the leg to be one inch from the front and side edges of the top." hinted this might be the case.

Kman