Yearlong Sweepstakes
I’d like to take this opportunity to lay out the details of the yearlong sweeps, for the purpose of either confirming them or adjusting them before our first tournament. While everyone is more than welcome to add their $.02, only the members of the Executive Committee (currently Myself, Michael Miller, Robert Maxwell, David Saulet, Danny Cantrell) get an official vote.
I will lay out the Yearlong Sweeps as it existed last year, and then I’m thinking in the next week or so we’ll call a vote via e-mail of those committee members.
There are three divisions of awards. As of last year, novice and senior distinctions are not made because coaches are expected to put competitors in their correct division, making the competition theoretically equal for everyone. The argument is that enough tournaments disallow very experienced debaters to enter Novice divisions to make coach abuse (putting senior level debaters into novice) negligible.
The first division is the Lincoln Division, which goes to the school with the highest percentage of wins. This one is the toughest to describe. Prelims are simple in that each competitor simply has their wins divided by the number of prelim wins. For example, a debater that goes 1-3 at a four prelim tournament would have a 25% winning percentage for the Lincoln Division. Final rounds become more complicated. In order to prevent punishing debaters for losing percentage points for breaking, all final rounds in which a debate competes are assumed to be 3-0 wins. If the debater does not progress beyond the first elimination round, only the initial 3-0 round is tabulated. For example, Debater A went 2-2 in prelims, and did not break. Debater B went 3-1 in prelims and lost according to all three judges in the first elim round. Debater C went 3-1 in prelims and won on a 2-1 in the first prelim, but lost 0-3 in the second prelim. According to the current tabulation system, A would get a 50% (2 wins, 4 rounds), B would get an 86% (6 wins, 7 rounds), and C would get 90% (9 wins, 10 rounds). To illustrate why this system was used, if we went strictly by wins and losses on every ballot, A would still get 50%, B would get 43% (3 wins, 7 ballots), and C would get 50% (5 wins, 10 ballots). As you can see, that system would sometimes lower someone’s percentage for breaking. The current system argues that prelim losses are important, but debaters are rewarded for proceeding to the next elim round by a higher win percentage. This division is designed to reward smaller teams who have more quality debaters.
The second division is the Douglas Division, which describes the number of adjusted wins. This is similar to the above calculation, but ignores the number of rounds involved. Every team simply gets a win for every prelim win, and 3 wins for each appearence in an elim round. This division is designed to reward larger teams who choose to support the activity with larger squads and participation.
The third division is similar to the Lincoln division, but for individual debaters as opposed to teams. The third division awards the SoCal-LD Debater of the Year award.
All three divisions are restricted by participation, and to be eligible a team or competitor must compete at a minimum of 33% of possible tournaments. Recall that only tournaments that take place in Southern California are tabulated, though schools from outside Southern California are tabulated. Last year, no school from outside the region went to the requisite 33% of tournaments (which turned out to be a minimum of three).
Last year, I tabulated the results. I felt comfortable including my own school in the award lists, since the data from which the results were drawn are publically accessible and transparent. I certainly don’t have to do the tabulation, but am willing to do it again. It really wasn’t too hard, once I set up the spread sheet to produce the proper document I posted at various intervals last year.
So, there it is. Can we make this better?
Did you enjoy this post? Why not leave a comment below and continue the conversation, or subscribe to my feed and get articles like this delivered automatically to your feed reader.
Comments
Very good ideas.
For #1, that tabulation would result in a 3-1 competitor who broke on points but lost 0-3 in the first elim round, receiving the exact same record as someone that went 3-1, and didn’t break on points. That makes sense to me. I think I like this change. It still prevents punishments for breaking, but makes prelim rounds more important. It would also keep the SQ policy of giving more points for more difficult tournaments, which have more elim rounds.
For #2, this is also a fine idea. It would not likely have a huge effect on the tabulation, since no debater attended more than 5 SoCal LD tournaments last season, though that may be more likely this year. Only three debaters (including yourself) attended 5. This would probably just result in a few people having their bottom tournament removed.
My only suggestion would be simplify the formula a bit – how about 1 of 1 ballot to the loser of each elim round and 2 of 2 ballots for the winner of each elim round? Sometimes data is not reliable for outround decisions. Having calculated NPDA sweeps for 3 years I think the easier the formula the better results.
So a student who goes 3-1 and doesn’t break gets a 75%. A student who goes 3-1, breaks, and loses goes 4-1 (80%). A student who goes 3-1, breaks, wins 1 elim, and then drops, gets (3-1 + 2 (elim 1 win) + 1 (elim 2 loss) = 6-1 (85%). A student who goes 3-1 and wins 2 elims would = 7-1 (87%).
Also, remember that this is a cumulative score. So a student who goes 4-0 and two elims gets 8 total ballots (8/8 = 100%) but that would be added to another tournaments where the student could go 3-1 and get 7 ballots – so (15/16) over the course of two tournaments.
I think the 5 total tournaments for each student is fine too – count only the top 5.
Sorry, the comment form is closed at this time.
I posted this in an earlier thread, here it is again. Two observations:
1) Elim rounds are ridiculously overweighted. If you can win just one or two elims your ballot count for that tournament could easily exceed 8 or 9 (or 80%/90%). That isn’t realistic at all.
Instead of just giving 3 free ballots for every elim round regardless of outcome, consider the following scheme:
* Give 2 out of 2 possible ballots for every outround win with 80% or more judges voting for you (i.e. a 3-0, 4-1, or 5-0 decision). Give 0 out of 0 possible ballots to the loser (there is no penalty for losing this round, but no reward either).
* Give 1 out of 1 possible ballot to both the winner and loser for any outround with a decision split less than 80% (2-1, 3-2, 6-3, etc).
This largely minimizes the massive outround ballot giveaway.
2) Use an NPTE style scoring system for both the Lincoln division and the top overall debater standings. By this, I mean that only the top five tournaments for each debater should be factored into the rankings. This is good because a) it minimizes the effect of one or two bad tournaments, and b) it incentivizes competing in as many tournaments as possible to accumulate the best stats.