I have a weird suggestion about the prelim format for this tournament.
The last few years, SSNCT prelims have not used a card system, instead dividing the field into pools of 11 and having them play a round robin, with teams finishing 6-4 or better making playoffs. This is by no means a
bad format, but it strikes me as suboptimal for a few reasons:
-- SSNCT is very hard to seed correctly, meaning that some brackets will unavoidably result in (potentially significantly) easier or harder paths to the playoffs.
-- Tying playoff progression to final record, rather than placement in a bracket, exacerbates this issue to some extent. Different brackets (which are theoretically composed of teams of the same average skill) send different numbers of teams to playoffs, dependent on the patterns of who upset whom.
-- Teams (especially those near the bottom and top of the field) get to play significantly fewer games against opponents of similar skill level. For the top teams, this may not be a big problem, but for the bottom teams, coming to nationals and getting to play only one or two games in which you are not massively blown out is not a good experience. I know multiple coaches of teams at the lower end of the field last year commented on their disappointment with the format.
However, with the current field sizes, using a 10-round card system is three or four rounds too much -- there would be a lot of games in the last few rounds between teams with significant differences in record, and a lot of rematches in the top and bottom of the field, leading to similar issues with unfairness of strength of schedule (except this time even nearly perfect seeding would not ameliorate it, because it would be down to luck of which cross-record games had which outcomes). We saw these issues the last time SSNCT ran with a card system (of the top 10 teams in the open division in 2019, only two didn't have any rematches in prelims, 6/10 had at least two rematches, and one team played three rematches!)
My proposal is to make the pre-lunch games and post-lunch games two separate, 5-game, card system tournaments.
In the simplest version of this, teams would play 5 games by card system before lunch. Then, after lunch, teams would be reseeded by their pre-lunch records/statistics, receive new cards, and play another 5 games by card system. At the end, as usual, teams with 6 or more wins would make the playoffs. This would have the following benefits:
-- I have not mathematically verified this, but I think it should be possible to guarantee avoiding rematches within a 5-game card system, or at least make them very rare. This means that most teams would play zero rematches (although some teams at either end of the bracket would probably play one rematch), a significant improvement over the 10-round card system.
-- This format would be significantly more seeding-resilient; the card system in general is quite seeding-resilient, and teams would literally be seeded for the second half based on their actual performance at the tournament -- one of the best seeding methods I can think of!
-- While teams wouldn't get the full benefit of a 10-round card system in terms of playing opponents of similar skill (in particular, rounds 1 and 6 would both be characterized by blowouts, like round 1 at other card-system-based NCTs), teams at the ends of the field would get a lot more meaningful prelim games than they would under a round-robin system.
The system could probably also be engineered to seed the afternoon by the cards teams hold in the morning, which would make afternoon seeding slightly less accurate but make it possible to significantly reduce the likelihood of rematches in the afternoon (and reduce the need for handing out new cards during lunch!). This would also allow for slight tweaks like "do a 4-game card system first to seed for a 6-game card system", should something like that be preferred.
I also think this format would work well with any field size that's a multiple of 8 (or even 4), providing more flexibility for NAQT in terms of the increments by which they could expand the field.
Here's a spreadsheet showing how this might go for some field sizes. It breaks pretty nicely -- the 60-team case is a little bit ugly but still has 22/30 of the last round games between teams of the same record and none more than 1 game off.