Red & White Bowl results

pakman044 · Post by **pakman044** » Sat Mar 18, 2006 2:13 pm

We're not done yet, but enjoy the first four rounds of stats:
http://www4.ncsu.edu/~paking/ncata/2006 ... whitebowl/

Patrick King

EDIT: Changed link for changed directory structure.

First Chairman · Post by **First Chairman** » Sat Mar 18, 2006 2:17 pm

Who da heck set up these divisions!!!???

rchschem · Post by **rchschem** » Sat Mar 18, 2006 6:55 pm

My thinking exactly.

Though it did lead to a RCHS/Cary Academy double fantasy semifinal. The final was RCHS A 525, Cary B 40.

Eric

ProsperoSMS · Post by **ProsperoSMS** » Sat Mar 18, 2006 7:47 pm

Hey, the stats aren't working anymore!

It seemed to go well overall (at least the post-lunch rounds I got to see did). Here's looking forward to continued years like this one, with several good tournaments for NC.

Hugh

rchschem · Post by **rchschem** » Sat Mar 18, 2006 10:39 pm

Stats are at

http://www4.ncsu.edu/~paking/ncata/2006 ... whitebowl/

Eric

radiantradon · Post by **radiantradon** » Sun Mar 19, 2006 12:18 am

Walter Williams did OK but our team had fun. I can't eat a whole lot during a quiz bowl tournament or it makes me sick, so I enjoyed the bagels after we lost to Cary Academy A ;)

I agree that the tourmanemt was very well-run. I think it's the first tournament I've been to that was AHEAD of schedule!

pakman044 · Post by **pakman044** » Sun Mar 19, 2006 1:10 am

Wow, very impressive Eric on guessing the URL. I rearranged the directory structure for additional stats and then added the index file before a few of my friends decided I had to accompany them to Cary Town Center...while I was very close to falling on my nose...on just about 1 hour asleep the night before. Thank goodness I wasn't the tournament director!

Now for the results....

As Eric mentioned, the stats are now located at http://www4.ncsu.edu/~paking/ncata/2006 ... whitebowl/

Results:

Playoffs
1. Raleigh Charter A (6-0, 600.0ppg) 400, 8. Robinson (4-2, 198.3ppg) 120
4. Cary Academy A (5-1, 359.2ppg) 330, 5. Walter Williams (5-1, 344.2ppg) 205
3. Cary Academy B (6-0, 346.7ppg) 265, 6. St. Mary's (4-2, defeated Richmond Senior 250-200, 240.0ppg) 190
2. Raleigh Charter B (6-0, 359.2ppg) 215, 7. Richmond Senior (4-2, 210.8ppg)

Semifinals
Raleigh Charter A 570, Cary Academy A 180
Cary Academy B 310, Raleigh Charter B 185

Finals
Raleigh Charter A 525, Cary Academy B 65

Top individuals prelims
1. Will Schultz (Raleigh Charter A) 117.50ppg
2. Nick Tarleton (Cary Academy B) 90.00ppg
3. Bryan Brooks (Arendell Parrott) 73.33ppg
4. Nancy Vanderveer (St. Mary's) 72.50ppg
5. Mark Hallen (Cary Academy A) 66.00ppg

Neg champion--Nancy Vanderveer (St. Mary's)--9. She added 2 more in the playoffs.

Thanks to all the teams who came out today and our wonderful staff. We were very excited to see so many teams there.

I'll discuss the issue about the division strength in a forthcoming post.

Patrick King

pakman044 · Post by **pakman044** » Sun Mar 19, 2006 1:56 am

One reminder before I completely fall on my nose: As per NAQT policy, please do NOT discuss the contents of the question sets (IS-57A) publicly until NAQT clears the question sets for discussion. This set will be used through the end of May, so please respect the wishes of NAQT. (Obviously you can discuss them internally within your team!)

Patrick King

DrakeRQB · Post by **DrakeRQB** » Sun Mar 19, 2006 8:55 am

Our kids and I had a wonderful time today - not just because we turned in our best performance of the year thus far, but because the tournament was SO efficient and well-run.

I liked the question set, though some of the questions were pretty much softballs - but I guess that's how the A-sets are supposed to be. The moderators were good, everyone knew the rules, the schedule was adhered to, and as someone said IT FINISHED AHEAD OF SCHEDULE! We have a 2.5-hour drive home and we were back at school by 6:00.

Thanks again to Paul, Patrick, and everyone else who made this happen.

rchschem · Post by **rchschem** » Sun Mar 19, 2006 12:21 pm

This was a well-run tournament with good readers all around who knew the rules. The facility was nice and quiet, and I think the question set was an appropriate level for the field of teams. The scores by and large, I think, reflected a nice balance between letting teams show some knowledge and offering a challenge.

And I've never seen such large cream cheese tubs before. God bless America.

Great tournament. We look forward to coming back next year!

Eric

ProsperoSMS · Post by **ProsperoSMS** » Fri Mar 24, 2006 7:38 pm

So what happened to the discussion of the brackets? Now that Duke Academic Festival is cancelled, we need some kind of way to pass the time.

DrakeRQB · Post by **DrakeRQB** » Sat Mar 25, 2006 6:41 pm

Yeah, Chuck - you're just gonna complain and not give any reasons?

radiantradon · Post by **radiantradon** » Sat Mar 25, 2006 7:41 pm

Apparently the brackets were chosen completely at random. Now, the mathematical definition of "random" and the common definition of "random" aren't exactly the same, which might explain the skewedness of the brackets.

rchschem · Post by **rchschem** » Sat Mar 25, 2006 9:47 pm

SQBS is coldly logical and knows nothing about A and B teams, team strengths, and so on. Should "random" mean "We didn't deliberately screw/reward any team in the pairings" or "We don't care who plays who from round to round?" I personally like the former. When I set up the pairings for the RTO, I used the SQBS "random pairing" function and then looked over the results to eliminate certain types of pairings (not specific pairings). I did not review schedules for any kind of equivalence, but I would really hate to see an undeserving team skate into the finals while good, hard working, red blooded American teams get chainsawed in a Bracket of Death. That's the bottom line.

After looking at the divisions more closely, it seems that they are better balanced, at least on paper, than they seemed at first. Each division did contain what I would consider 3-4 good teams (based on what I've seen). Some appeared in the end to be more competitive than others, but I think in retrospect they were OK. I would have swapped someone from the top 4 in Tompkins with someone from the bottom of Winston, probably. Tompkins was tough. Terry Sanford is a better team than 3-3, and their 3 losses were by 2 questions or so each (with decent bonus coversion--they averaged 19).

Eric

DrakeRQB · Post by **DrakeRQB** » Sun Mar 26, 2006 1:56 am

Tompkins was indeed tough. Our team did really, really well to go 4-2 - especially considering we were missing three of our top five players. I'm pretty sure Saturday was the first time we ever knocked off Terry Sanford's A team.

pakman044 · Post by **pakman044** » Sun Mar 26, 2006 3:04 pm

I guess I should really explain how things were done for the preliminary rounds. I've been a bit on the busy side the last week or so, and it's going to get more so this coming week, so this is the best chance I have to actually get this written....

For the preliminary rounds, I took each of the 21 teams and assigned them a number using the command :randInt(1,24) on my TI-86 (essentially producing a random integer in the interval [1,24]). Thus teams with numbers 1-7 were in Caldwell, 9-15 were in Tompkins, and 17-23 were in Winston. The only proviso on the process was that we decided that two teams from the same school would not be paired in the same division (if the number generated placed two teams in such a situation, I would simply generate another number). The team numbers themselves are mostly transparent to the end user; they were only used to determine a team's division, when their bye was, and when they would play each other team.

Now I presume the next question will be, "why did (we) choose a random system?" The major reason is that the tournament director and I really did not believe it was appropriate to hand select the divisions. We believed that in selecting the divisions in a way to make them as "balanced" as possible, we are in effect creating divisions that produce outcomes that we expect to happen--in essence, the teams that we believed to be stronger would emerge from their divisions. Now this is all well and good, assuming that the judgment we could make as to relative team strengths was accurate. But we did not believe that we could make such a judgment in an accurate manner. Even though we had a number of competition results, the results don't say how good the teams are in general. The results after all reflect which teams show up on a given day, the format, and even the composition of the individual team, which often changes on a weekend-to-weekend basis. In the end, we are left to make a subjective judgment about how good the teams are, and we did not feel comfortable making that kind of judgment.

When I did the randomizations and looked at the divisions that were produced, I had a feeling that the divisions were not evenly "balanced". I have heard from different people observations that were actually different from my own about the source of the imbalances, which indicates that perhaps my initial feelings about how the divisions were imbalanced were wrong (which makes me feel that I likely would not have produced that much "better" of division breakdown by handpicking the divisions or doing some form of classful randomization where the "better" teams are placed in relative identical quantities across the divisions). But to remedy the potential imbalances, there were really only three good options that I could up with:

Handpick the divisions
Rerandomize the divisions until I came up with something that appeared more or less balanced
Let the current division makeup stand

As things turned out, the TD and I settled on that third option. We couldn't find a solution that we felt would make things better while keeping our own feelings out of the situation--and I certainly was very uncomfortable with the situation that I was thinking hard for a solution for several days. But try as I might, there really was nothing else I could come up with.

So there's the methodology and the justification. I'm definitely interested in finding another way to do business, but it has to be a way that keeps my preconceptions of how good everyone is out of the pairings.

Patrick King