Thanks, for pointing out the mistake, it was actually diameter I had in mind while using radius. But still I can't imagine how the ball should move outwards. I'll put the question in a different manner. Suppose the ball was fixed to the groove during the course of rotation, all of a sudden the rotation is stopped by an external agency and the ball is made free to move along the groove simultaneously. The ball will then have a tendency to move perpendicular to the radius(or groove) due to inertia, but isn't this tendency hindered by the walls of the groove.