As an example only, and not to call anyone out or try to denigrate them, here is an example of the A/B/C ranking system using the best guess I have at the "uniform" scores.
The original ranking (presuming EXP and ADV are properly aligned with no overlap, ranking all Experts ahead of any Advanced:
1 'Fitzgerald ' 608
2 'Buck ' 607
3 'Dvarvydis ' 599
4 'Aron ' 592
5 'Nunes ' 586
6 'Just ' 584.5
7 'Pomposo ' 565
8 'Wright ' 545
9 'Shorts, L. ' 525
10 'Moore ' 522
11 'Schulz ' 520
12 'Dean ' 501.5
13 'Cunha ' 497.5
14 'Shorts, D. ' 497
15 'Schulz ' 489.5
16 'Wong ' 484
17 'Scholtes ' 474.5
18 'Constantine' 448.5
19 'Stimpson ' 70
What I did was take the BEG/ADV/EXP entries, and corrected the EXP scores to be more-or-less consistent with the Advanced, by scaling them down by a constant factor of .92. The idea was to create a consistent slope for those not considered outliers.
The dotted line is what I was trying to match between the classes. This was only necessary to get an equuvalent score for expert, as if we had flown all the flights for the same judge panel. In a real-life case, you wouldn't need to do that, since all the fliers fly at once on Saturday.
Once that is done, presume that we all flew together on Saturday, got the scores as adjusted. Then, with 19 entrants, break it up into 7, 7, and 5, then on Sunday, yo have 3 separate groups, A, B, and C, that are only competing with each other:
1 'Fitzgerald ' 577.6000 A
2 'Buck ' 576.6500 A
3 'Dvarvydis ' 569.0500 A
4 'Aron ' 562.4000 A
5 'Nunes ' 556.7000 A
6 'Just ' 555.2750 A
7 'Pomposo ' 536.7500 A
8 'Shorts, L. ' 525.0000 B
9 'Moore ' 522.0000 B
10 'Schulz, J. ' 520.0000 B
11 'Wright ' 517.7500 B
12 'Dean ' 501.5000 B
13 'Cunha ' 497.5000 B
14 'Shorts, D. ' 497.0000 B
15 'Schulz, Z. ' 489.5000 C
16 'Wong ' 484.0000 C
17 'Scholtes ' 474.5000 C
18 'Constantine' 448.5000 C
19 'Stimpsom ' 70.0000 C Note that this has nothing to do with what happens at the next contest, does not "establish" someone in a class, it just means on this particular Saturday, this is how you came out against the other people who showed up. Next time, it might be different, based on how you flew. Obviously we can't conclude what would have happened when you did the A, B, and C flyoffs, because we didn't actually do that, but it gives a taste of how it would work.
At the end, you give out 1-2-3 for A, 1-2-3 for B, and 1-2-3 for C, so there are actually two competitions - one, to get in this highest group you can (because, conceptually, #1 for B is actually 9th place), then, trying to beat out the other guys in your group to get the #1 trophy for your group.
Brett