Not sure I understand the wax hate.
It is logical if you stop to consider this may have a mechanical component and is not a not a purely electrical issue (unless you are the guy who swapped node boards and had a different component fail each time, you either have a short or 2 bad boards).
At the moment the magnet engages it needs to overcome both the gravitational force [m g sin(theta) where theta is the pitch of the playfield] as well as the friction resistance of the ball interaction with the playfield. We know the strength of the magnetic field (B) is inversely proportional to the distance (r) from the core of the magnet (and directly proportional to the current passing through the wire), so for the ball to change direction we need the the force applied from magnetic field to be greater than the sum of the the gravitational force and the friction force at that distance which we can think of as the following:
B(r) > m*g*sin(theta)+Fk.
You can see here why reducing the friction will help. The more friction you have, the further that ball is going to travel away from the core of the magnet to where the magnetic field gets weaker the the less ability it has to overcome the angular momentum of the ball rolling.
pasted_image (resized).png