When I said "tilt the mast forward", what I meant was, tilt the pivot axis forward because otherwise, if it is perpendicular to the equatorial plane, it will track a circle that is eccentric to the Geosynchronous circle, but if you tilt the pivot axis forward between one half and one degree, the intersection becomes an ellipse that tracks the arc more closely. Remember in Calculus, when they demonstrated that the intersection of a cone and a plane is an elipse ? Well, a polar mount with a declination angle sweeps out a flattened cone, but the intersection is still an elipse. I think in one of Dr. Frank Baylin's mistake-laden texts, he says the maximum tracking error of a modifid polar mount compard to a true polar mount is about one-half as much error.