Dave, (first of all I've not read the new rule book description nor flown any "clover" for a long time, butt....)
Assuming the four "round" loops' segments must be of equal size and be flown in the same airspace as the "old" version...:
I would think the judging of the maneuver would not "begin" until the initial vertical climb reaches the point of tangency of the top right loop segment to the vertical arc's path.
The loops are still notionally the same size and fit in the same space. The one consequential change is that it removes the latitude/longitude sort of definition. So you start on the downwind side, turn up (somehow undefined) to get yourself to vertical climb dead downwind, when you hit 45 degrees and you are going in a straight line, the judging starts. Then you do more or less the same maneuver as before, ending when you cross 45 vertically again.
To me, it's a wash, not much easier or harder than before. The two issues I have seen is that it took 10-20 flights to figure our where you start the first loop on the way up, and, the overwhelming tendency to finish the overhead 8, go a 3/4 lap, then start climbing up to 38.5/42/??? degree level flight after having done it the old way 20,000 times.
How you get it vertical on entry, to get to 45 and vertical, seems to be a matter of variation right now, some people jam a tight corner like starting a wingover on the wrong side of the circle, others (your two proteges' included) generally enter pretty gently to maintain energy. Obviously we think we are right, you are going to climb vertically into the wind, and starting to the right of downwind, so you can use all the energy you can get. In good conditions it doesn't matter, matters in the wind a little bit.
It's not a huge change. I like the change to projected angles (vice lat/long) from a hypothetical standpoint, but in practice you had to do it that way anyway before, too.
Brett