So far it's showing the truth of a rule of thumb that AussieGunwriter posted a few years ago. If your groups show two and one or in this case three and one, seat deeper. If they are evenly distributed but larger than they should be, seat shallower. You can see that flyer in #2, 3 and 4, and it's still there in #6 but it gets closer and closer to the main group.
One problem I have is implicitly trusting these samples of one each but they do show a definite progression which matches the idea of accuracy nodes being a sine wave, coming and going at different depths. With that in mind there's a good chance that if I was to start seating closer to the lands that group represented by #1 would start to get tighter.
