Live stats from Springfield

[dtaylor4]

Tomorrow I will be doing stats from the Springfield VJV tournament. Each division has 24 teams. Varsity is using an NAQT set, and I think JV is using the Auburn house-write. I'll post a link here and/or on Facebook once I get round 1 entered, for those who are sitting at home with little to do.

[jonah]

I had thought the JV division was using the KC set, but NAQT's website says it's using IS-91A. I asked one of the Auburn people about it and he (I think it was Zahed?) said that maybe KC was for their league tournament or something.

[dtaylor4]

I had thought the JV division was using the KC set, but NAQT's website says it's using IS-91A. I asked one of the Auburn people about it and he (I think it was Zahed?) said that maybe KC was for their league tournament or something.

I do recall someone from Auburn saying that.

Also, I promise not to tweet answers, nor will I put them in a facebook status. I will also issue a verbal threat to all players and coaches who even think about doing so.

[CometCoach72]

We are looking forward to a great day of competition. I hope everyone travels safe and has a good time- good luck to everyone participating.

[dtaylor4]

Yea, sorry this didn't happen.

Unfortunately, most of my time was spent deciphering the sometimes barely-intelligible scribble of high schoolers putting in minimal effort for extra credit. By the time the playoffs were done, I only had FOUR rounds entered in each division. The format was: five card system rounds involving 24 teams. 20/20 NAQT, no powers, no negs. Weird timing rules. Two time-outs per game, and NO PLAYER CAP. If there's one part of quizbowl in this part of the state I hate more than anything, it's seeing players literally play ONE TOSSUP in a game, or have a coach play TEN players in a single game, or have someone sub out, only to be subbed back in later.

Ok, maybe I hate several aspects.

[David Riley]

Yeah, I saw a weird manifestation of that at Masonic State one year: the coach subbed out his best player when they could have easily won the match had he remained--they lost. Later, I asked the coach why, and he replied "because everybody wins".

[CometCoach72]

I can only say from the limited view of what I saw and what I heard from others, but I thought Mr. Adkins ran an excellent tournament. Almost all of the matches we played were close and enjoyable. The students seemed to have fun (the most important part).

I never really heard who won, which was the only bad part, but I know that will come up at some point in time soon.

Donald & Jeff, it was nice finally putting together faces and names. I'm looking forward to going over the final results whenever they became available.

[garciaja]

I thought it was an interesting way to run a tournament and obviously took a lot of organization, which Springfield did a great job of doing. The moderators were very competent of the rules and that worked out great. Champaign Centennial's Varsity team was in a 4 way tie for first with New Berlin, Litchfield, and PORTA all at 6-2. After having to hear people exclaim CIRCLE OF DEATH, it was determined that the way the tie would be broken was by points per bonus.

While I'm biased because our team came out on the bottom of that tiebreaker, I wondered why tournaments don't first go with the head to head tiebreaker. In a three way tie the situation is usually that three teams have beaten each other (Stevenson, WWS, and Centennial at D&G). However, in this case Centennial had beaten the other three teams that were 6-2, thus (and here's my bias) I thought it would be an easy pick as to who the better team was. We probably shouldn't have gotten ourselves into the situation of losing 2 games, but I guess I wonder why points per bonus is the default tie-breaker. We played bounce-back and no bounce-back at two separate tournaments and both of them were points per bonus tie breakers.

I guess I haven't been around the block enough and I don't understand the logic behind it. Wouldn't total points be a better measure, or quality of opponents played? Someone care to explain?

[Matt Weiner]

While I'm biased because our team came out on the bottom of that tiebreaker, I wondered why tournaments don't first go with the head to head tiebreaker. In a three way tie the situation is usually that three teams have beaten each other (Stevenson, WWS, and Centennial at D&G). However, in this case Centennial had beaten the other three teams that were 6-2, thus (and here's my bias) I thought it would be an easy pick as to who the better team was. We probably shouldn't have gotten ourselves into the situation of losing 2 games, but I guess I wonder why points per bonus is the default tie-breaker. We played bounce-back and no bounce-back at two separate tournaments and both of them were points per bonus tie breakers.

I guess I haven't been around the block enough and I don't understand the logic behind it. Wouldn't total points be a better measure, or quality of opponents played? Someone care to explain?

This has been discussed a lot, and the idea is that head-to-head rewards the wrong things. A four-way tie is very unusual and may not be the best place to demonstrate this, but consider the simple two-way tie for first as follows in an eight-team round robin bracket:

Team A 6-1 (lost to Team C)
Team B 6-1 (lost to Team A)
Team C 5-2 (lost to Team B and Team D)
Teams D and below 3+ losses

The head-to-head tiebreaker, which would favor Team A, has two problems in this situation:
1) Team B only lost to the other team tied for first, but Team A lost to a third-place team that took a loss to someone out of contention entirely. Team A's loss was worse, but they get rewarded for that.
2) Team A already received credit for their win over Team B to create the 6-1 tie in the first place. To double-count the game that already created the tie, when the point of a round-robin is that your record against the field is what is meaningful, is arbitrary and unfair.

I don't know exactly what the format of this tournament was, which affects the issue of total points vs. bonus conversion. In a round-robin where both teams play the same set of opponents, total points is the best paper tiebreaker since it takes into account all aspects of the game. When the strength of opponents differs, total points is far less fair because scoring 200 points against Team X may not mean the same thing as scoring 200 points against Team Y. Since bonus conversion is an opponent-independent statistic, it's the thing to fall back on in that case.

Ideally, all ties would be played off on full or partial packets; I don't know the packet or time situation of this tournament, so that may not have been feasible. When playing off the tie doesn't work, PPG is the stat to use for ties created by round-robin scenarios, and bonus conversion is best for other formats.

[AlphaQuizBowler]

This thread has some good discussion and data regarding different tiebreaking methods. It does refer to 2-team ties, not 4-team ties, but I think most of the ideas should still apply.

[garciaja]

Ideally, all ties would be played off on full or partial packets; I don't know the packet or time situation of this tournament, so that may not have been feasible. When playing off the tie doesn't work, PPG is the stat to use for ties created by round-robin scenarios, and bonus conversion is best for other formats.

That was part of the problem. Our team had to get some of our kids back for a band concert thing and we couldn't playoff or anything. Does the bonus conversion matter when the bonus parts are bounce-back though. I guess I see the positive and negative, but points per bonus still seems pretty arbitrary. Also, this was not a round-robin, but teams played their first 5 matches against teams with the same or similar records in the morning and then were seeded into the afternoon for three games. Our problem was not winning our first playoff game when we were seeded #1. Oh well. I do appreciate Springfield running a high quality tournament downstate.

[Irreligion in Bangladesh]

What format got you guys in a 4-way 6-2 tie in the first place?

[garciaja]

What format got you guys in a 4-way 6-2 tie in the first place?

The first round was randomly drawn and we drew Decatur LSA. After our round we were given a card and told to go to our next room where we were going to play a team with the same record as us. We then played and beat New Berlin, Macomb, Litchfield, and PORTA. After the 5 rounds were were seeded 1 in the championship bracket of 8. We lost our first match to SHG, beat Jacksonville, and then lost our last match (very badly) to Glenwood.

We beat the other three teams and I assume that their other losses were to each other.

[dtaylor4]

Ideally, all ties would be played off on full or partial packets; I don't know the packet or time situation of this tournament, so that may not have been feasible. When playing off the tie doesn't work, PPG is the stat to use for ties created by round-robin scenarios, and bonus conversion is best for other formats.

That was part of the problem. Our team had to get some of our kids back for a band concert thing and we couldn't playoff or anything. Does the bonus conversion matter when the bonus parts are bounce-back though. I guess I see the positive and negative, but points per bonus still seems pretty arbitrary. Also, this was not a round-robin, but teams played their first 5 matches against teams with the same or similar records in the morning and then were seeded into the afternoon for three games. Our problem was not winning our first playoff game when we were seeded #1. Oh well. I do appreciate Springfield running a high quality tournament downstate.

For the bounceback tiebreaker, we stated in the coaches' meeting that we were only using BC from controlled bonuses. Bounceback points were removed in calculating this, despite the issues this created with our scorekeepers.

[TheTeen]

And if I am not mistaken, points from the afternoon rounds were the only ones used, correct? We we had 2 losses in the morning (To New Berlin and Centennial (10 - 30 point loss margins)), which placed us number 6 in the final bracket. During the afternoon, we went on to beat Macomb, Porta, and New Berlin placing us, also, into the tie. Because only final bracket scores were used and we were the only team involved in the tie to win all 3 of our afternoon matches may have been the reason we won. It was, overall, a great tournament though. I quite enjoyed it.

[mrgsmath]

Yes the difficulties came when it was announced at the coaches meeting at the beginning of the tourney that only 7-1 ties would result in a play-off. All others would use the bonus total method. It was not anticipated that a 5-0 would lose the first match in the afternoon to an 3-2 team., nor was it expected that they would then lose again later. It was a very well run tournament, and only the result was unfortunate.

[dtaylor4]

Stats for the Varsity division are finished. The four teams in the square dance of death have qualified for the NSC and the HSNCT. The files have been sent to Jonah for posting to the IHSSBCA website.

[jonah]

prelim stats

overall stats

[CometCoach72]

Thanks Donald. Glad that you were able to decipher everything!