The Quizbowl Resource Center

Good morning,

I'm pleased to announce that Eric Mukherjee of Brown, George Stevens of Clemson, and I are collaborating on a tournament set, THUNDER (acronym to be changed/expanded/abandoned later) which will be ready for play in the fall. We will be writing the entire tournament set between the three of us this summer, so there will be no need to submit packets.

What kind of questions can I expect to play on?

The tournament will be of regular season difficulty, so you might look at Penn Bowl 2007, Penn Bowl 2008, or the upcoming ACF Winter set to get an idea of what to expect. We'll be writing 15 total packets utilizing the official ACF distribution, so in a final packet, you can expect to see, with minor variation:

4/4 Literature
4/4 History
4/4 Science
3/3 Religion, Mythology, and Philosophy
3/3 Fine Arts
1/1 Social Science
1/1 Geography
1/1 Trash

Average question length will fall between 6 and 8 lines, and we plan to powermark the questions for great glory.

Where will this tournament be run?

Good question! We're not sure yet. We're officially soliciting bids from all regions, but will of course give higher priority to schools that we attend. For example, while we're happy to accept bids from anywhere in the Midwest, there is a good chance that this tournament will be held at Illinois if it is feasible to do so. The same applies to Clemson University, and wherever Eric Mukherjee ends up attending medical school.

What if my program wants to run this tournament?

You should e-mail me at [email protected], and we'll make arrangements. I'd like to have the hosts picked by the end of the summer so that we can have an official weekend across regions, although we'll be able to accommodate minor variation if necessary. You can expect to pay $30 per team that plays your mirror site.

This should be a terrific set, and I hope you all choose to host or play. We're looking forward to producing a great, controlled difficulty set that everyone should enjoy.

Now that ACF Nationals is over, it's time for us to confirm that this tournament is indeed happening and to try to line up regional hosts. I'm pleased to announce that Chris Carter has joined our team in the capacity of copy-editor and packet creator. Anyone who has seen me try to construct a packet from a file full of questions will know that this is very good news indeed. Eric, George, Chris, and I would like to claim the weekend of October 24 as the date for this tournament. We hope all hosts will be able to accommodate our wish to have all mirrors happen on that weekend, but we understand if that's not possible. If you are able to host on October 24 or a nearby date, please contact me at [email protected]!

The rest of this post will serve as a list of regional hosts, updated as it goes:

Northeast:

Brandeis University

Mid-Atlantic:

University of Pennsylvania

Southeast:

University of Tennessee at Chattanooga
and
University of South Florida

Midwest (We are amenable to having upper and lower Midwest hosts, depending on the geographical situation.):

West:

University of California at Irvine

Southwest:

University of Tulsa

North:

Carleton College

Northwest:

University of Washington

Carleton, huh? Awesome!

I believe the lack of good tournaments in the Southeast is a mediocre or better reason to have it at Clemson, or at least a mirror.

I agree, Amit. Hopefully I'll host the southeastern mirror, and we may try to get another mirror in MS, LA or TX. If Clemson is able to host, it would have to be on the 24th because of the football schedule.

I know USC put in a bid for EFT, but they haven't heard anything from anyone one way or the other. If they don't get the bid for that and if Clemson isn't able to host Thunder, I'm sure they would be willing to host Thunder.

A little off topic, but the Southeast actually had a pretty good lineup of college tournaments in 08-09. Reports of there not being good quizbowl tournaments in the South have been exaggerated.

The Midnight Rider wrote:I know USC put in a bid for EFT, but they haven't heard anything from anyone one way or the other. If they don't get the bid for that and if Clemson isn't able to host Thunder, I'm sure they would be willing to host Thunder.

A little off topic, but the Southeast actually had a pretty good lineup of college tournaments in 08-09. Reports of there not being good quizbowl tournaments in the South have been exaggerated.

Oh yeah, sorry about this. I thought we had replied to you guys but I guess not. I think we decided USC would be a good host for EFT.

Sweet.

grapesmoker wrote:
Oh yeah, sorry about this. I thought we had replied to you guys but I guess not. I think we decided USC would be a good host for EFT.

That is awesome. Thanks.

FYI, we're probably hosting a delayed mirror of this on 11/21 at the University of Washington. If anyone wants to take a trip to Seattle to play a quizbowl tournament, this would be a good tournament to do so for.

Is there a non-Minnesota Midwest mirror for this yet, or one anticipated? If not UIUC, maybe Northwestern or hopefully Iowa, so we could stay in that sweet motel designed like a castle, complete with moat and drawbridge and coats-of-arms.

marnold wrote:Is there a non-Minnesota Midwest mirror for this yet, or one anticipated? If not UIUC, maybe Northwestern or hopefully Iowa, so we could stay in that sweet motel designed like a castle, complete with moat and drawbridge and coats-of-arms.

We've spoken to Wash U, but we've not heard anything definitive. We'd love to have one at any of those places; I don't think Illinois will be hosting this.

I wouldn't be a big fan of a mirror in Iowa, since it'd probably draw teams away from the Carleton site (teams like, say, Chicago!). Obviously, that's not my decision though.

And who would want to miss out on a trip to scenic Northfield. It's the town of cows, colleges, and contentment.

OctagonJoe wrote:And who would want to miss out on a trip to scenic Northfield. It's the town of cows, colleges, and contentment.

Don't forget the Malt o' Meal factory, home of delicious delicious Marshmallow Mateys!

But yes, a mirror in Iowa wouldn't make too much sense.

Fair point on Iowa - I pretty much just wanted to agitate for my best chance to stay somewhere with a portcullis this quizbowl season and didn't think about how close it is - but a lower Midwest mirror would be somewhat nice. Making that drive two weeks in a row might decrease our likelihood of showing up, and would certainly make us less likely to show up with lots of teams.

There will be a mirror at the University of Ottawa on October 24. Contact the club at uotrivia -AT- yahoo.ca if you are interested.

The google quizbowl map says Clemson is hosting THUNDER on October 24. Is this true?

Duke The Dumpster Droese wrote:The google quizbowl map says Clemson is hosting THUNDER on October 24. Is this true?

We don't know yet. If Clemson can host it, they have first priority; if not, we've had other bids from the southeast. We'll announce as soon as we know for sure!

millionwaves wrote:

Duke The Dumpster Droese wrote:The google quizbowl map says Clemson is hosting THUNDER on October 24. Is this true?

We don't know yet. If Clemson can host it, they have first priority; if not, we've had other bids from the southeast. We'll announce as soon as we know for sure!

Assuming Clemson does host it, will it be the only Southeastern mirror, or might there possibly be another? I really want to go to this but just as with EFT, my funds will prevent me from going all the way to South Carolina.

Hey, everyone, progress on this tournament is very steady and we're excited about how it's turning out.

I've updated the post above with a whole bunch of information about the regional hosts. I can't help but notice that there is not yet a Midwest host; Illinois can't do it. So if you're some place in the Midwest and you'd like what amounts to a whole bunch of money for your club at the mere effort of sending me an e-mail and running a tournament, please sign up to host!

PS: I'm really looking to privilege someone that wouldn't draw away from Carleton's site, so, Iowa, I'm afraid, is out.

millionwaves wrote: Southeast:

University of Tennessee at Chattanooga

Yay I can probably go to this now

Hey, Trygve, did you get my e-mail about NIU offering to host a Midwest site?

Hey, everyone,

With a few days to go until the completion of this set, I wanted to check in with you, the community, on its progress:

If everything goes well, all of the questions for this tournament should be written and edited by very late on Wednesday; it may be Thursday instead.

Unfortunately, Chris is going to be busier than he expected due to having a show to manage this week. It might be really helpful to us to have some assistance on moving questions from files into packet format; that would let us focus on fun things like "play testing" and "making sure power marks aren't crazily uneven." I've already had an offer from Amit Bigli of Southside (thanks, dude!). Is there anyone else who would be willing to help us in this endeavor? Similarly, I've had an offer from Donald Taylor to help us copy-edit and proofread. Maybe some people would like to step up and help out with that, too? If so, please send me an e-mail at [email protected]!

Hosts, you should expect to have the set in your e-mail by no later than Friday at noon. I sincerely hope I won't be eating these words in a few days.

List of people who are awesome (comprehensive):

Amit Bigli
Sandy Huang
Donald Taylor
Aidan Mehigan

THUNDER is being constructed as I post this (with a few placeholders.)

Wednesday has turned into Thursday has turned into Friday at noon has turned into... well, I got up at 5:30 and still no packets.

This tournament was announced in January, and it's not like you had to deal with people submitting at a +$50 deadline. This is unacceptable, to put it mildly, and I've got far too many other things to do on the morning of a tournament than sit around and wait for questions.

As far as I know, most (if not all) of the tournament was sent out by 5am eastern time, so maybe they don't have your e-mail?

I'll cc Ben.

Ladies and gentlmen,

I'm at a computer again, so I can post this:

First off, sorry that the questions went out so late!

We're extremely interested in your feedback, but for now, there are two delayed mirrors of this tournament, so we can't start a discussion thread yet. We're interested in solving any errors and etc. before the tournament is sent out to another site, so please send errors according to the following:

Literature, Social Science, Philosophy: Trygve Meade ([email protected])
History, Geography: George Stevens ([email protected])
Science, Visual Art, Mythology: Eric Mukherjee ([email protected])
anything else: all three of us, if possible.

Also, my sincere apologies to Ben from Ottawa. I wasn't able to see your post this morning, so I didn't know until you e-mailed me that you weren't on the CC list to receive packets.

Thanks for playing, and I hope you guys enjoyed!

millionwaves wrote:Average question length will fall between 6 and 8 lines

YOU LIE

Representative Joe Wilson (R-SC) wrote:

millionwaves wrote:Average question length will fall between 6 and 8 lines

YOU LIE

You'll have to do a better job of convincing me that long tossups are better. At the Ottawa site with a mean tossup conversion of 57%:

6 & 7 line tossups: 96 readings, 5/67/7 for 75% conversion, 1 of 24 went dead in all rooms
11 & 12 line tossups: 40 readings, 3/11/3 for 35% conversion, 5 of 10 went dead in all rooms

The math then tells me that 8-10 line tossups (664 readings) went 40/326/85 for 55% conversion, but I don't really have the time to check every tossup to be sure.

The point of having longer questions isn't better conversion, its to have more clues, and thus theoretically better differentiation between teams.

ClemsonQB wrote:The point of having longer questions isn't better conversion, its to have more clues, and thus theoretically better differentiation between teams.

I don't think anyone disputes this. I'm pretty sure Ben's point is the correlation between chosen tossup length and difficulty of the answers to those tossups. Shouldn't the easier answers have longer tossups? If people know more clues about some topic, a longer tossup will differentiate better. If only a few people know about a topic, a longer tossup won't differentiate anything.
For an extreme example, opening up the textbook next to me to a random page, suppose I wanted to write a tossup on Shechtmantite (apparently some quasi-crystalline Al-Mn alloy) and one on a Fourier transform. Would I write a short tossup on Fourier transform and a long one on this random alloy? I would really differentiate knowledge well that way.

(edit to add last sentence to make sure I was clear)

squareroot165 wrote:

ClemsonQB wrote:The point of having longer questions isn't better conversion, its to have more clues, and thus theoretically better differentiation between teams.

I don't think anyone disputes this. I'm pretty sure Ben's point is the correlation between chosen tossup length and difficulty of the answers to those tossups. Shouldn't the easier answers have longer tossups? If people know more clues about some topic, a longer tossup will differentiate better. If only a few people know about a topic, a longer tossup won't differentiate anything.
For an extreme example, opening up the textbook next to me to a random page, suppose I wanted to write a tossup on Shechtmantite (apparently some quasi-crystalline Al-Mn alloy) and one on a Fourier transform. Would I write a short tossup on Fourier transform and the long one on this random alloy?

I agree that harder tossups should be shorter than easier ones, but I was just clarifying for Ben because his post didn't make it seem like he understood the reasoning behind having longer questions.

ClemsonQB wrote:The point of having longer questions isn't better conversion, its to have more clues, and thus theoretically better differentiation between teams.

This is a problem when some very well-known clues routinely came up in line 3 or 4 of an 11 line question. If you're going to write those kinds of tossups, you have to make sure your clue selection justifies it, and this didn't.

I agree with Jerry here. There were several history tossups where quite frankly, the length didn't really matter per se because a basic giveaway came very early for those who had even a mild familiarity with the topic. In this case, the question is basically a wash and becomes annoying for novices who don't know what it is until the giveaway and must still sit through clue after clue.

I'd like to know which questions Mike and Jerry are referring to. Email me at [email protected]

ClemsonQB wrote:The point of having longer questions isn't better conversion, its to have more clues, and thus theoretically better differentiation between teams.

Using my data, 65% of the longest tossups had zero differentiation between teams. In fact, the problem compounds itself because lower conversion leads to fewer bonus opportunities, which further diminishes the differentiation.

Michael (Hausinger) is making it clearer that the opposite trend of what I observed should be happening: if a tossup is long, it would surely have sufficient clues to get converted; if a tossup is short, it could be lacking in clues to get that conversion.

ClemsonQB wrote:I agree that harder tossups should be shorter than easier ones

I'm showing evidence that says otherwise.

I'm not making complaints about overall difficulty (all the teams coming to the Ottawa site knew what to expect, and roughly got it); I'm complaining about the effectiveness of these tossups that exceeded the length guidelines you set in the original announcement, which the spoof account pointed out.

So, if you George, a main writer on the set, are perfectly content admitting that you deviated wildly from the announced guidelines, then you need to stop god damn arguing about how it can be a good thing to ask 12 line tossups in a set that announced an 8 line cap and instead just admit you did it wrong and will do better next time staying at the announced cap. Whether or not the questions are good is irrelevant, there is no excuse to deliver something significantly different than what you announced, and from what I'm hearing here, this set's length falls into that category.

Jeremy Gibbs Free Energy wrote:So, if you George, a main writer on the set, are perfectly content admitting that you deviated wildly from the announced guidelines, then you need to stop god damn arguing about how it can be a good thing to ask 12 line tossups in a set that announced an 8 line cap and instead just admit you did it wrong and will do better next time staying at the announced cap. Whether or not the questions are good is irrelevant, there is no excuse to deliver something significantly different than what you announced, and from what I'm hearing here, this set's length falls into that category.

What the fuck are you talking about Charlie, did you even see the set at all? You are really in no place to comment on this.

I wrote no 12 line tossups, I wrote no 11 line tossups! I wrote a handful of 10 line tossups, and sure, that deviates from the guidelines, so I'll apologize for wasting an extra 30 seconds of everyone's time for having to listen to a couple of questions that were longer than advertised.

ClemsonQB wrote:I wrote no 12 line tossups, I wrote no 11 line tossups! I wrote a handful of 10 line tossups, and sure, that deviates from the guidelines, so I'll apologize for wasting an extra 30 seconds of everyone's time for having to listen to a couple of questions that were longer than advertised.

Hey dude, perhaps it's just the margins on the packets I have in my possession, but I definitely counted a few 10 or 11 line monsters in there. Anyway, as I mentioned in the email I sent you, some questions, which may or may not have been written by you (but I think were) had some construction problems that made it weird to have many more clues after a really famous thing was dropped early on.

Eh, I'd be wary of just saying a "handful" of ten line tossups; while I agree that through the five packets I just skimmed, you did in fact write no over-ten line tossups (though two or three were over ten by less than a line), the vast majority of your tossups were over eight lines, assuming that you did, in fact, write the history for this tournament.

I did write all the history, so all the blame should be directed at me. With normal 1x1 margins, I counted no 11 liners. I wrote a good many questions over the 8 line-mark, and realize that I should have cut some clues here and there to make them shorter. I'm sorry that I didn't get around to doing that.