--- In quizbowl_at_y..., "nephelococcygia" <dthorsle_at_u...> wrote: > --- In quizbowl_at_y..., briman14200 <no_reply_at_y...> wrote: > > Why do all tournaments insist on Swiss Paired second rounds. I > think > > that once teams qualify for playoffs we should go to a single > > elimination format. It would be far more exciting and would make > > every single game relevant during the afternoon. Also you would be > > far more likely to see some surprising teams going deep into > > tournaments even winning. In the event that a "surprising team" wins something, it should be because they earned it, not because a system is set up in such a way that treats such results as an ipso facto "good" thing. If we really wanted that kind of result, why not just have playoff games be three tossups in length? That would be really "exciting" and would get more "surprising teams" winning. More seriously, Dave's explanation works pretty well, namely, that one outlier of a round can serve to eliminate lots of teams (whether the "top" teams or otherwise) and make everything that came before it irrelevant. The only reason it's done in AC tournaments now is due to either question or time shortages. What is seemingly more common, however, is a situation of "de-facto single elim." Dave's same logic can hold true for round-robin pools if you have a tiered system where teams are sorted by placement and separated into groups (top, middle, bottom, etc, assuming no cross-bracket games) and only one or two teams are taken from each group. Sometimes there would be a three-way tie within a group with one loss, and one of those one-loss teams gets eliminated. It's arguably _worse_ than single-elim playoffs since at least in single-elim playoffs, teams are on notice going in that a loss in that game means they go home, and that the contents of the round with the ability to eliminate can be more closely monitored. (Whereas in a de-facto single elim situation, it could be on any round of the tournament, and is frequently the starkest outlier.) It's particularly unfair in multi-bracket situations where other teams elsewhere can get playoff berths with multiple losses while a one-loss team is eliminated. For some reason, such three-way ties at the top were more common in CBI tournaments than elsewhere on the circuit, though plenty of circuit invitationals employed the same setup. Again, TDs might be excused if there were question shortages and/or time pressures. I was somewhat saddened, however, to find that de-facto single elim is still rearing its head at national championship tournaments, who largely don't have to deal with byes, submitted rounds, severe time pressure, or moderator crunches. How do you fix the problem? If you're going to have a playoff, you should probably take more than two teams from each pool, if feasible (assuming you have a field large enough so that four per bracket isn't excessive ; it makes little sense to let, say, 8 of 12 teams into the playoffs, or 16 of 20.) But if your field is 48 or 36 or even 24, taking two per pool is little better than single-elim, if at all. Failing that, you could have the tiebreaker matches the way NAQT did in their D2 ICT this year. True, it only goes halfway, to the extent that 1 1/2 losses could send you packing, but single+half elimination is preferrable to being knocked out by a single loss in a random round, particularly given that the half-round is very much not randomly "chosen." Or you can set up a system of cross-bracket, schedule-evening games that lasts long enough for all but a small handful (four is a good number) of teams that would have at least two losses. -Tim
This archive was generated by hypermail 2.4.0: Sat 12 Feb 2022 12:30:45 AM EST EST