How to "seed" prelim brackets

Dormant threads from the high school sections are preserved here.
Locked
User avatar
cchiego
Yuna
Posts: 891
Joined: Fri Jun 06, 2008 7:14 pm
Contact:

How to "seed" prelim brackets

Post by cchiego »

After tournaments this past weekend, I've seen a rise in complaints about improper prelim bracketing. I'd like to start a discussion of what can be done to address the issues inherent with trying to "seed" prelim brackets.

As TD, you have only limited information to make these choices. Some schools will have a strong track record, but simply underperform. Others will be missing key players. Still other schools might be a total surprise and have some player who spent the previous summer reading the entire online packet archive. Thus here's what a tournament director should, at a minimum, do when setting up the format for a tournament:

1. Be familiar with previous tournament results. Look carefully at the stats and see how well the teams did, especially at tournaments similar to the level you're hosting at. Check the NAQT website results as well as the ones posted on this forum and pay attention to how well the schools in your area usually do. This may seem like common sense, but it's a crucial first step.
2. Attempt to have as much geographic/team diversity as possible. If you get three teams from out-of-state at a tournament, split 'em up. If you get 4 teams from the same school, try to scatter them across the various brackets. This is just common courtesy for teams who travel a long distance to play.
3. Try to use some kind of "wildcard" mechanism that helps account for good teams that got screwed over by unlucky bracketing (esp. in large tournaments). The best way to do this is probably PPB. Reserve a spot or two in the championship bracket for such a team.
4. Make every effort to keep the number of teams competitive for the "championship bracket" as large as possible. For instance, if you only have 4 rounds to work with for playoffs, you could have two "championship" playoff brackets of 4 teams each and then have a final between the winners of each bracket. Teams that are still in the hunt are much more likely to be willing to stay these extra rounds too, especially in HS where teams have a propensity to leave early.
5. (Optional) Try to figure out the quizbowl "background" of new schools that randomly pop up come tournament time. While more often than not they'll probably be bottom-bracket feeders, it can't hurt to find out more about their background since they might turn out to be a surprise (it has happened). Plus if they're brand new to "good quizbowl," you can help point them to online resources and other good tournaments in the area. No TD really has to do this, but it would be nice and helpful for the game as a whole.

At the same time, coaches and team leaders have responsibilities as well:

1. Be upfront with TDs about missing players, particularly if their absence will have a major effect on your team's performance. It's not enough to say "we'll be missing two of our A-team members" if you'll be missing your lead scorer and only science person for instance. More information is always useful in this case.
2. Not try out "unusual" compositions of teams without informing the TD. If you put your "A" team players on your "C" team without telling anyone, it's a nightmare for the brackets. Similarly if you're trying to "balance" your teams with a few good/new players on each team, let the TD know as well.
3. Accept the fact that sometimes, despite the best good-faith efforts of TDs, your team might end up getting screwed.

JV tournaments especially can be difficult to seed, particularly if there are a number of new players and teams that don't play very often. For that, perhaps the card-swapping format used at HSNCT could be replicated on a small-scale to help make these matches more fair and competitive. I know a few tournaments have in fact used this format for JV tournaments (Chattahoochee being one of them) and it seemed to work well, so I think this might be a good way to resolve the information asymmetries at the JV level.
Chris C.
Past: UGA/UCSD/Penn
Present: Solano County, CA
Locked