Comparing ONE recruiting class to ONE recruiting class doesn't tell you much, I mean, there is some difference between recruiting classes even within the same system of generation. I would think you have to see a good number of recruiting classes before you can make any concrete observations. Of course, if the AVERAGE of all players went up by three points, but the standard deviation went down, then the natural result would be that devision one would go down a little bit, Division thereen would go up a bit, and division two probably stays similar. . however, the average difference between the top fo division one and the bottom of division one. . .
However, say you take that as a representative sample: Recruits 1 - 150 went down, the rest stayed almost even.
1 - 50 went down by an average of 25.5
51 - 100 went down by an average of 17 points.
101 - 150 went down by 6.5 points.
151 - 200 went down by 2
201 - 250 went down by 2.5
251 - 300 by an average of 3 points.
So the average distance between the top recruits and the remainder has, indeed, gone down. I imagine if you plotted out the entire spectrum from d1 to d3, you would see a slight raise in average rating. . however, a raise of three points, while over a large sample it would show, is small enough to get lost in the noise of the random variation in the entire set from one season to the next. Absolute rating is meaningless, only relative rating. . and the elite recruits HAVE come down to be slightly closer to the non elite recruits.