Alright, here are some preliminary thoughts about the regionals pairing system.
What I would do, if I had cards, is reconstruct the pairings and then use 8 as the cutoff, as if this were a regional, and see how that changed it.
This was something I was wondering about as well, so I did the re-pairing last night. Even assuming 8 bids (SLU used a 5 "bid" cutoff, since there were 5 team awards), there would not have been any change to the Round Four Pairings. With 5 bids, the Last Bid Record going into Round 4 was 4-2; that stays the same with 8 bids (in fact there were 10 teams with a record of 4-2 or higher). That resulted in 0 teams automatically in, 4 teams definitely out, and a number of teams on what I'll call "life support" at 2-4.
The fact that there were 0 automatically in doesn't surprise me. My guess would be this will almost always be the case. With the level of parity that has developed in AMTA over the last few years, I would expect several teams to stay in that 4-2 bubble, usually enough that 4-2 will be the last bid record. It's possible that would fall to 3.5 or 3 ballots, which could free up some 6 or 5.5 win teams to move down into bracket two. Granted this is just my hunch, but I don't think that will happen very often, especially at the Opening Round Championship sites.
What I think this pairing system does is more or less set up a "win and you're in" situation for teams at the Last Bid Record. Because of the High-Low pairing, it makes that win easier for teams who already come in with a lead in the Fourth Round (as witnessed by the 5-1 teams below). If you're on the outside looking in, that won't be a good position to be in.
I have to admit that I don't quite understand pairing systems, but from looking at the tab summary of the invitational something does not seem quite right. The top 3 teams who all went 7-1 were 5-1 after round 3, and hit three teams who were each 2-4. Going further past that, the 4th and 5th place teams who went 6-1-1 and 6-2, each hit teams that were 3-3. When not even a single team in the Top 5 of a tournament hits a team with a winning record in the 4th round, there is a problem.
This was my initial impression, but after looking at it a little longer I don't think that the pairings were quite a unfair as you suggest. In order to properly assess the evenness of the pairings, you have to compare the records of the teams as they entered Round Four. Here are the pairings, with the top ten teams (going into Round Four) with their record and the record of the opponent that they faced. (Left doesn't mean Plaintiff or Defense, just the higher ranked team.)
924 (5-1) v. 952 (2-4)*
519 (5-1) v. 613 (2-4)*
660 (5-1) v. 445 (2-4)*
633 (4-1-1) v. 528 (3-3)*
768 (4-2) v. 489 (2-4)*
488 (4-2) v. 632 (3-3)
925 (4-2) v. 468 (2-4)*
490 (4-2) v. 517 (3-2-1)
612 (4-2) v. 720 (3-3)
470 (4-2) v. 444 (3-2-1)
The pairings with an asterisk indicate a pairing where there was more than one ballot difference between the two teams. One reason that the pairings looked the way they did was certainly that Round Four is paired high-low. The size of the bracket contributed to this, since with more teams included, the lower the record of the lowest teams will be. However, at this tournament the Round Four side constraints contributed as well. Most of the 4-2 teams were on the side of the bracket needing plaintiff, if I remember correctly. This makes it less likely that a team will hit a team with a similar record.
Even with the factors discussed above, 4 of the pairings were fairly even, with a difference less than or equal to one ballot. One more had a difference of 1.5 ballots. Two had a difference of two ballots. The ones that I think give people trouble are matchups at the top, three 5-1 teams versus three 2-4 teams. That's a big downside if you're trying to determine which of those 5-1 teams is the best, but that's not the purpose of this pairing system. The problem would be comparing one of the 5-1 teams to one of the 4-2 teams that had a "better" opponent in Round 4 (measured in terms of previous success in the tournament).
As I said before, I think this more or less sets up a "win and you're in" system. That's exactly how this turned out at SLU. 10 teams came in with a record of 4-2 or better. 8 of those teams won both ballots. 2 split (488 and 470). The end result was that the 8 teams that did win both ballots were the 8 top ranked teams in the tournament. Those 8 would have gotten bids if this were a regional tournament. I think that's a fair outcome.
The one unresolved question is whether it's fair to have the 5-1s hit the 2-4s. Again, I'm not sure you'll always see such a big discrepancy in records. This is only one tournament, so it's hard to generalize. However, I don't have any problem with the 5-1s seeing opponents with lower records than the 4-2s. Going into Round Four, the 5-1s had identified themselves as better than the 4-2s in terms of their performance against the competition. Someone had to hit the 2-4s.
Yeah. I don't remember which one, but when I first looked over the results, I thought I saw that one 7-1 team hit three 2-6 teams. That just doesn't seem right.
MTSU hit 472 (1-7), 489 (2-6), 924 (7-1), and 445 (2-6). Round Four is the round that I think you may have a legitimate beef with, but for the reasons discussed above I don't think it's so bad. Round 3 was definitely an even matchup, as the results indicated. As for Rounds 1 and 2, it's really just a fluke. Round 1 is paired randomly, nothing you can do about that. Round 2 is paired high-high and although 489 may not have ended with the best record, they were 2-0 going into Round 2. Their performance after hitting 660 is nothing that you can take into account when pairing. If anything, MTSU was hurt by it's opponents in Rounds 1 and 2, since their CS was the lowest of the 7-1 teams.
These thoughts aren't really very flushed out, but maybe they'll be a starting point for further discussions. I have another question about the Round 3 pairing, but since it's not really specific to this tournament, I'm going to throw it in the Numbers section.
