Well, a beautiful position would usually earn a plus in the grade of execution. Good speed, good centering, significantly more than the required number of revolutions would also add pluses.
I expect that Nikodinov's and S. Hughes's laybacks as performed ca. 1999-2002 would often have earned +1 and sometimes +2. They weren't as fast as Zhang's or Nagasu's or sometimes Cohen's (or, e.g., Nathalie Krieg's or Lucinda Ruh's), and didn't have as many as most of those (Cohen didn't often hold hers any longer in those days, but she competed under the new system she did).
They would have been level 1.
The new (this year) feature for holding a position for 8 revolutions would mean they would only have to do one more variation -- speeding up while in position would be an option -- to earn level 2.
Skaters like Zhang and Nagasu (and Krieg, back in the day) who could do all of that with good basic layback positions and *also* keep the spin going faster and longer through other difficult variations should get credit for being able to do even more than those who just had good positions basic with good centers.
Maybe it would be possible to define a good classic layback position -- the angles of the free leg and the degree of back arch -- so that it could count as a "difficult position" for one more feature. The trouble is that it would be difficult to define it objectively, but I would certainly support an effort to do so.
And I'd also like the GOEs to be worth more than the difference between one level and the next, so it would be more valuable to do a level 2 layback with +2 GOE than to do level 4 with 0 GOE.
Medusa said:
It's definitely not outlawed -- it's not an illegal move at all, and it would be considered, and rewarded if well done, in the transitions component. If done immediately before a jump, it could also add to the GOE.
But you're right that it's not specifically recognized and explicitly rewarded and would just take up time without increasing the level of a step sequence, so a skater reading through the rules who doesn't already have a good split jump in his or her repertoire wouldn't see any reason to develop one.
What's happened with the rules as written is that some moves or variations that used to be rare are explicitly mentioned in the documentation, or have been inferred by skaters and by the technical specialists to count as features, and so have become much more common than they used to be.
Other moves/variations that may have been fairly common a few years ago explicitly do not count as features or have not consistently been credited as such, or they do count explicitly but they're more difficult to achieve successfully than other variations that earn the same reward, so they become less common or seem to disappear completely.
As the rules are tweaked each year, some things that were less common a few years ago under COP become more common, and vice versa.
I would recommend some major tweaks to explicitly reward even more types of moves and variations and to encourage more variation in long program structure -- I think that would benefit skaters by giving them more options to play to their own strengths in earning points, and it would benefit audiences by giving more variety of programs to watch.
Also, some moves fall in and out of fashion through the gradual development of the sport, quite apart from any rule changes. If a popular champion uses a given move to good effect, then pretty soon we'll see lots of other skaters at all levels including it as well. If the top skaters at the time don't happen to do it, or do it well, or if the ones who do the move don't seem to be rewarded for it compared to more obviously rewarded moves that could fill the same purposes in a program, it will fade out of use. This effect is true across all eras.
. 
