I think one objection to going straight in is that a chip is peeled off by each flank of the tool, both chips try to occupy the same space and wedge up. But it obviously worked for you.
A half-way house, with the tool set at right angles to the job can be achieved, if the topslide is set parallel to the bed. Each time the cross slide is advanced to deepen the cut, the topslide is advanced towards the chuck by half the amount by which the cross slide was advanced. That makes the leading flank of the tool do most of the cutting, as with the angled approach. It does need a bit of concentration (for me, anyway) to keep track, and if the cross slide dial is marked up in "reduction in diameter" rather than actual depth of cut, then the top slide advance should be a quarter, not a half, of what the cross slide dial shows.
Andy