Kentucky '10-'11

[Scott]

Absurd.

I feel awful for you guys and Pikeville.
No matter what happens one of the top 10 teams in the state is going to get knocked out the first night.

[Rococo A Go Go]

A few observations on the questions:

1) Most of the non-math questions are 2-5 lines long, with the bulk of them being about 3 lines long. While reading or judging this weekend I was at times reminded of an NAQT A-set (lengthwise), which is at least better than what KAAC questions have been in the past.

2) There questions seem to be varying in pyramidality, but the regional set was better than the district set on this front. Most likely the best questions are reserved for the State Finals, so I'm eager to see if that theory is correct. Some of these questions (especially Social Studies and Arts and Humanities) do seem to be adhering fairly nicely to pyramidality, but obviously there are occasional (sometimes glaring) exceptions.

3) The Science is all over the place, I'm not entirely sure what's going on there. Some of the questions were quite good, while occasionally they were decidedly non-pyramidal. The Math had some nice theory questions, but the computation questions are just computation questions. I'd love to see mathcomp reduced (or maybe even eliminated) in favor of more theory but I can't forsee that ever happening, there are too many math teachers and education officials that would absolutely throw a fit.

4) Spelling questions need to go away and grammar questions need to be reduced. Way too much literature in the world that could be asked about that we're missing out on, and spelling questions are just annoying at this point. They're all two line questions that start with a sentence full of ten dollar words and then use the second sentence to give a definition of the word in the previous sentence that means a certain thing. It's as pyramidal as you can make a spelling question, but I'd rather writers spend their time coming up with interesting clues for a non-spelling question written about a word.

5) Overall, inconsistency is still ruling the day but the questions are starting to show significant leaps forward. Compared to where KAAC questions were 5 years ago, the ones today are amazing. Hopefully these trends in the right direction continue.

Also, I'd like to note (pertaining to what Scott said at the beginning of his post) that in my opinion we can probably treat what's still called "quick recall" as quizbowl. It's not up to the standards of good quizbowl yet, but there are plenty of worse formats and questions that have been discussed on this board.

[Faiyad]

Off topic but...
Also, the new outlier system for the written composition scoring caused an individual from our school who had two firsts and two mid scores to place behind three individuals with three mid scores and a low score. I know the intentions are good, but it doesn't really work out that well.

What are you trying to say exactly? It sounds like you are saying that you wanted to keep the outlier and place him/her lower rather than use the modified, higher score.

[Rococo A Go Go]

Off topic but...
Also, the new outlier system for the written composition scoring caused an individual from our school who had two firsts and two mid scores to place behind three individuals with three mid scores and a low score. I know the intentions are good, but it doesn't really work out that well.

What are you trying to say exactly? It sounds like you are saying that you wanted to keep the outlier and place him/her lower rather than use the modified, higher score.

The situation involved one of our students. who placed behind three students who each had one of their scores modified by an outlier, while our student's score did not have any scores ruled as outliers.

[Scott]

About the outlier system, I am saying that the intentions are good, but the system does not work as it is intended.
For example...

A student who had the following ranks 1st, 2nd, 4th, 9th would finish above a student ranked 1st, 2nd, 4th, 8th.
Because the 9th would qualify as an outlier and the 8th would not.
Does that make sense?

[skkid11]

About the outlier system, I am saying that the intentions are good, but the system does not work as it is intended.
For example...

A student who had the following ranks 1st, 2nd, 4th, 9th would finish above a student ranked 1st, 2nd, 4th, 8th.
Because the 9th would qualify as an outlier and the 8th would not.
Does that make sense?

I'm not familiar with this system. How exactly does this work? Do they eliminate any sub scores from total score calculation that are distant from other sub scores? I'm not sure I see the logic behind this system. It would seem to me that the flat out simplest, and best way to calculate a comp score would be to just average all the sub scores and be done with it.

[Scott]

I'm not familiar with this system. How exactly does this work? Do they eliminate any sub scores from total score calculation that are distant from other sub scores? I'm not sure I see the logic behind this system. It would seem to me that the flat out simplest, and best way to calculate a comp score would be to just average all the sub scores and be done with it.

The take the subscores that are more than twice as high as any other score and average that subscore with the average of the other three subscores.
It has good intentions but is rather a headache if you ask me.

It results in cases where students are better off to receive a 9th place than an 8th as I previously mentioned.
With our student, she had two graders rate her 1st, a 4th and an 7th (or something like that).
Her average would have placed her 2nd, but three students with 3 scores in the 4-6 range and a single very low score were adjusted ahead of her.
I'm not saying that her paper was better, but it shows that the system is flawed.

Definitely agree that a total/average would be simplest and best.
That is the way it has been done every year up until now.

[Faiyad]

I think that the outlier system is fine, but the definition of an outlier is what is flawed. I think that twice the highest subscore is too low to be considered an outlier. Scott's example had scores that were fairly close together, so a simple average would have been acceptable. But if you had scores like 3, 7, 19, and 5, the outlier skews it terribly. Then it seems appropriate to modify the scores.

[Cody]

If the goal is to minimize the effect of "outliers" (where outliers anomalously low placements), then they should just use the harmonic mean, which greatly minimizes the effect of "low" scores (such as placing 20th) while greatly maximizing the effect of "high" scores (such as placing 1st).

[Faiyad]

If the goal is to minimize the effect of "outliers" (where outliers anomalously low placements), then they should just use the harmonic mean, which greatly minimizes the effect of "low" scores (such as placing 20th) while greatly maximizing the effect of "high" scores (such as placing 1st).

I like that, but I doubt that it'll be utilized by KAAC.

[Rococo A Go Go]

To switch gears a bit...

WKU is hosting the College Heights Academic Tournament on March 26th, and UofL is hosting the Kentucky NAQT Championship on April 23rd. Are there any other good quizbowl tournaments planned?

[alexdz]

So I know this has something to do with a post here from a few months ago, but I read it just now, and wanted to point out that there is precedent for a "statewide magnet school" type thing attending SCT. Fort Hays State University's program (KAMS) sent a team to Region 11 this year.

[Coach K]

To switch gears a bit...

WKU is hosting the College Heights Academic Tournament on March 26th, and UofL is hosting the Kentucky NAQT Championship on April 23rd. Are there any other good quizbowl tournaments planned?

Danville is hosting the middle school NAQT set on April 2nd and IS-100 on April 16th.

[quantumtheory]

If the goal is to minimize the effect of "outliers" (where outliers anomalously low placements), then they should just use the harmonic mean, which greatly minimizes the effect of "low" scores (such as placing 20th) while greatly maximizing the effect of "high" scores (such as placing 1st).

The harmonic mean would be something to look at but it actually could backfire.
Here is a real example: Student A, ranks 3, 5, 6, 11: Harmonic mean = 5.05747. Student B, ranks 14, 12, 10, 1: Harmonic mean = 3.18786. In the competition, Student A actually finished 5th and Student B finished 10th but using the harmonic mean, Student B would have finished 5th. The rankings shown are lined up by judge-by-judge which means that Student A out ranked Student B in three of four rankings but would have lost because of the calculation of the harmonic mean which factors in the inconsistency of the rank value of 1. I know some people will ask how Student B got a ranking of 1 but this is the kind of inconsistency you see every year, especially at region when there are 20 competitors.

I think the outlier system is flawed in that raw scores are converted to ranks and then the ranks are used to perform mathematical calculations. The top raw score on Composition is 92. If the first place student from one judge earns a 92 and the second place student from the same judge earns an 80, they are ranked 1, 2 regardless of the difference in their raw scores.

Checking the average of raw scores may be an option but I know there are some concerns about how easy or hard each judge scores.

I would like to see the number of judges doubled and then throw out the highest and lowest score of each student, add up the ranks, lowest score wins.

[jackatthedisco]

Any particular reason why for the past four years, Pikeville has been on the same side of the bracket as Dunbar?
yea, that's fair.

[Faiyad]

Any particular reason why for the past four years, Pikeville has been on the same side of the bracket as Dunbar?
yea, that's fair.

Oh Jack, always with the QQ. ;P.

That pools are pretty intense this year.

[Kahloon]

Any particular reason why for the past four years, Pikeville has been on the same side of the bracket as Dunbar?
yea, that's fair.

Exceptionally intuitive bracketing systems

[Rococo A Go Go]

KAAC State Finals (Quick Recall) Results:

1st- Dunbar
2nd- Johnson
3rd- Manual
4th- Russell

Quarterfinalists: Adair, Apollo, Grayson, Simon Kenton

More: http://kaac.com/ASAP/finalresults/state/default.htm

Questions were better again this year, everything but math is mostly pyramidal. Hopefully with the new format, next year's tossups will be up to standard quizbowl quality. I'd say that some of them are pretty good already, I especially noted some of the Social Studies questions as being adequately well written and pyramidal. That being said, I still think the format is kind of weird (especially the single elimination) and hopefully that can improve some in the future. Dunbar played really well today, and from what I saw they definitely deserved the State Championship. Anyway, congratulations to all the teams who placed and all of the written assessment, FPS, and composition winners.

[Kahloon]

I especially noted some of the Social Studies questions as being adequately well written and pyramidal.

Were you one of the writers?

[Rococo A Go Go]

I especially noted some of the Social Studies questions as being adequately well written and pyramidal.

Were you one of the writers?

No, but since I know more about Social Studies than other subjects I can assess them more easily. I should have probably worded my sentence better anyway.

[Scott]

First of all, I would like to congratulate Dunbar and Russell, as well as all of the other teams and individuals who placed at state.

Now...
I would like to take this time for an evaluation.

I really liked the location, probably better than Louisville.

No one likes listening to a five line bonus.
Hopefully this is fixed next year.

The format is very outdated and should be changed again.
I don't really see the point of single elimination in any quizbowl related event...

As far as the quality of the questions, I was actually pleasantly surprised.
Some of the topics were a bit strange to ask about and others were oddly irrelevant, but they weren't that bad overall.
There was at least an effort to make them more pyramidal.
Even thought there was quite a bit of discrepancy on how hard the initial clues were, they were, for the most part, difficult.
The give aways were give aways; that's pretty hard to get wrong.
What needs massive improvement are the middle clues.
While some of the middle clues were things that most good players would know, others were rather outrageous.
Many questions seemed to be very top heavy, skip the clues that good teams would react on, and lead to a buzzer race on the final clues.
This could be due to question writers that are not aware of what is canon as far as intermediate clues, but nonetheless, it is something that needs improvement.

Having said all of this, I do think the questions were comparable to NAQT A set tossups albiet a bit more difficult and scaffolding.
However, that should not really be what a state finals set is aspiring to be...
This is a nice place to be, but there is still room for a lot of improvement.

So, in my opinion, KAAC did better than normal with their state set this year, but still had numerous problems.
Hopefully it will soon be close legitimate quizbowl.

[jackatthedisco]

Shikha and I would like to say that it's called "Quick Recall" for a reason.

[mkanu]

I would like to say in response to Shikha's performance on the written assessments Damn! and congratulations. Congrats to everyone who placed and Johnson and Dunbar for quick recall.

[Rococo A Go Go]

The format is very outdated and should be changed again.
I don't really see the point of single elimination in any quizbowl related event...

I agree with this wholeheartedly.

I'm not going to complain about the pool play (even if it's not really standard pool play) on Sunday, and as long as the pools can somehow become more balanced, then it's an adequate way to determine the 16 teams who should advance to Monday are. However, the Monday single elimination definitely needs to be scrapped. I think 2 balanced pools of 8 would be great, and while I'd prefer a re-bracket after those pools are completed, I understand that time constraints make that improbable so just doing a cross-bracket game between the pools to determine placement would be fine. This would be 8 rounds, but I think a few things could be adjusted (go to 12 minute halves for Monday, start earlier in the morning, shorten time between rounds) that would allow the awards ceremony to take place at a not much (if any) later time than it does now.

Another thing that needs to be paid close attention to is the balancing process for the initial pools, as well as maintaining balance for the Monday pools I would like to see added. KAAC knows how well teams do at District and Region, and presumably can keep up with them outside of Governor's Cup play as well. Some regions are stronger than others, and I hate to say this, but for the purposes of balancing pools Bethlehem winning the fairly weak 5th Region should not be equated with Johnson Central winning the much stronger 15th Region. This is especially important considering that stats and results show that the runner-up in the 15th (Pikeville) would probably easily beat Bethlehem, but was unfortunately placed in a pool with two of the top 8 teams in the state (Dunbar and Simon Kenton) while Bethlehem was placed in a pool that featured no teams who would make the top 8. There has to be some way to recognize this and seed all 32 teams who qualify for State in a more reasonable way.

[Rococo A Go Go]

Kentucky Teams in Fred Morlan's latest rankings:

16. Dunbar A
29. duPont Manual
32. Johnson Central
52. Adair County
70. Danville
71. Dunbar B


I think that at full strength, that ranking is probably about right for Dunbar although maybe they could creep higher. However, I think Manual is being underrated. They were extremely impressive at CHAT, and that was without two people who reguarly start. I know Dunbar wasn't technically at full strength either, but Manual did beat them and I think if both teams were at full strength that Manual would at least be very close (if not victorious) to Dunbar.

As for the other rankings, they mostly seem right. I'm not sure what to think of Danville; they pulled off a win over Dunbar A in the morning at CHAT, but then proceeded to lose to everyone in the top playoff bracket in the afternoon. For the most part though, based on their stats throughout the year at least, I think they are probably ranked at about the right spot. I also think that if the rankings were spread out to 150 (I don't believe they should be BTW) that Simon Kenton, Grayson County, and Russell would all be in contention for a spot.

[Andrew RJ Lyon]

Congrats to Dunbar, Manual A, Adair, and Manual B for placing in Top 4 at KY NAQT State

[AlphaQuizBowler]

If any Kentucky teams are interested in playing BHSAT, I encourage you to attend our mirror at Alpharetta High School.

[FCPanther]

This is really for next year's thread but Fleming County High School has scheduled the 2012 edition of the FCAT for Saturday January 14th (King Weekend). Details on the tournament thread. Basics are really the same: we'll use the new KAAC format for quick recall and plan to use more pyramidal toss ups in keeping with KAAC's changes. We do plan to use one of the national question vendors, and modify their existing sets to KAAC format.

[Rococo A Go Go]

Since we're talking about next year, this is as good a time as any to bring up WKU's fall hosting plans:

WKU JV/Middle School Challenge--Likely to be on October 29, this tournament will have division for JV High School teams and another division for middle school teams. We're hoping to use the SCOP Novice set for this.

WKU Hilltopper Invitational--Likely to be on November 19, this will be our standard fall high school tournament open to all teams. We might order an NAQT set or HSAPQ set, but we're also looking at mirroring a housewrite.

We'll also host CHAT II next spring (we'll announce it sometime this fall) and probably bid to host a collegiate tournament.