timer delenda est

Old college threads.
User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

timer delenda est

Post by grapesmoker »

The clock must go from NAQT matches.

I don't think I'm exaggerating when I say that the Northeast SCT was one of the smoothest-running tournaments I've ever attended; kudos to Andy and co. for keeping things moving. However, I would also like to point out that this efficiency was made possible in large part because Brown could afford to give up Aaron, Ian, and Dan (that's half of our DI team and one-third of our DII team, for those keeping track at home) to staffing. The guys did a great job reading, but I have to say that I think it's really kind of ridiculous for a non-staffing team to have to give up half of its players just so a tournament can run on time with competent moderators. We could afford to do this, but that won't be the case every year.

These problems can all be traced directly to the use of the timer. When you have a timer and are forced to have two staffers in each room just so you can get through more than 20 questions in a round, this places a huge burden on hosts. No one wants to bid on an SCT because no single team has the resources to single-handedly staff such an event. We would have loved to host an SCT at Brown, but we can't do it because the most staffers we could field would be about 10 or 12, and for a region the size of the Northeast, which tends to draw many teams (there were 22 teams at Harvard), there's no way we could have two staffers in a room. In fact, even with the setup we had yesterday many rooms were single-staffed (like, every DI room). Sure, those rooms had great, fast readers, but we could have gotten through more questions if we'd had dedicated scorekeepers; however this was impossible because we had already rounded up pretty much anyone around who knows how to help out at a quizbowl tournament.

It's time for the clock to disappear. It adds absolutely nothing to the game experience and its drawbacks are legion. It induces an unacceptable amount of variability into matches with some teams at less fortunate regions hearing as few as 17 or 18 tossups in a game because there are no competent moderators available. It makes it nearly impossible to staff an SCT even (or perhaps especially?) in a region as active as New England. Finally, I think that the clock gives packet compilers an excuse to backload packets with shitty questions in the hopes that most teams will never see that horrendous tossup 25. It is, simply, an anachronism, a terrible gimmick left over from a day when NAQT mistakenly thought it would make the game more "exciting." Get rid of the clock, guarantee every team either 22 or 24 tossups a game, and rework the distribution so that 1/3 of a round isn't decided by trash and geography.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance

User avatar
Important Bird Area
Forums Staff: Administrator
Posts: 5609
Joined: Thu Aug 28, 2003 3:33 pm
Location: San Francisco Bay Area
Contact:

Re: timer delenda est

Post by Important Bird Area »

Not speaking for NAQT, I agree with Jerry and plan to send R. a similar message re: clock policy this week.

The distribution issue is (I think) completely separate.
Jeff Hoppes
President, Northern California Quiz Bowl Alliance
former HSQB Chief Admin (2012-13)
VP for Communication and history subject editor, NAQT
Editor emeritus, ACF

"I wish to make some kind of joke about Jeff's love of birds, but I always fear he'll turn them on me Hitchcock-style." -Fred

User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: timer delenda est

Post by grapesmoker »

bt_green_warbler wrote:The distribution issue is (I think) completely separate.
Not completely. In the actual discussion thread we have instances of people saying that they put questions towards the end of the packet when those questions feel gimmicky or not very good, and I think you and I both know that this actually is a thing that happens. With the clock absent and every team guaranteed X questions, there is no longer an excuse for not having identical per-X-questions distribution in each packet and no excuse for putting awful questions in the back in the hopes that the moderators won't get to them. Distribution reform is not completely orthogonal to this discussion, although I don't want to make this thread about it (I will make a separate thread about that).
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance

User avatar
Important Bird Area
Forums Staff: Administrator
Posts: 5609
Joined: Thu Aug 28, 2003 3:33 pm
Location: San Francisco Bay Area
Contact:

Re: timer delenda est

Post by Important Bird Area »

More to say on that in the specific discussion.
Jeff Hoppes
President, Northern California Quiz Bowl Alliance
former HSQB Chief Admin (2012-13)
VP for Communication and history subject editor, NAQT
Editor emeritus, ACF

"I wish to make some kind of joke about Jeff's love of birds, but I always fear he'll turn them on me Hitchcock-style." -Fred

User avatar
Mechanical Beasts
Banned Cheater
Posts: 5673
Joined: Thu Jun 08, 2006 10:50 pm

Re: timer delenda est

Post by Mechanical Beasts »

grapesmoker wrote:
bt_green_warbler wrote:The distribution issue is (I think) completely separate.
Not completely. In the actual discussion thread we have instances of people saying that they put questions towards the end of the packet when those questions feel gimmicky or not very good, and I think you and I both know that this actually is a thing that happens. With the clock absent and every team guaranteed X questions, there is no longer an excuse for not having identical per-X-questions distribution in each packet and no excuse for putting awful questions in the back in the hopes that the moderators won't get to them. Distribution reform is not completely orthogonal to this discussion, although I don't want to make this thread about it (I will make a separate thread about that).
The per-packet distribution issue is not necessarily something that would be excused by the lack of a clock. Having per-tournament distributions is really important for guaranteeing quotas of things you want 3/3 of (and I think it's intimately tied to NAQT's production model where you have a million people writing questions, and in which no one is per se assigned questions. I do think it's silly to have varying per-packet quantities of what we consider top-level categories like literature, history, science, etc., but I'm not at all uncomfortable with a distribution that provides varying per-packet amounts of, for example, fine arts and RMP (so do most circuit tournaments, anyway; isn't 1/1 usually cut from the 6/6 total of those subjects submitted? I remember some people saying that 3/3 fine arts is almost always retained, but I haven't found that to be true in practice) and certainly not with a distribution that provides varying per-packet quantities of subcategories.

Here's my own position on timers, as articulated somewhat to Jerry and Ted at SCT. There was a time in which NAQT was doing something to make the game exciting by having a timer: when the game could end in the middle of a tossup. Now, that sucked (just as it sucked even more in CBI when it could end in the middle of a bonus). So we got rid of it (yay!). Now there's no excitement argument. Or, if there is one, it's so totally overwhelmed by everything else you can't breathe. ("I'm sitting in the room with the fourteen tossup per game moderator (these beasts are known to exist). I am excited to find out which is the category with two-thirds of its questions in tossups 15-26 (a range large enough to easily hold that many questions of one category. I am excited to find out which category will only be present in one third of its intended strength. Maybe I am playing solo Ted Gioia and will win because that category is lit. Maybe I am playing Nowhere State and will lose because that category is science.")
Andrew Watkins

User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: timer delenda est

Post by grapesmoker »

The restrictions on time that you can take to think after buzzing are another deleterious aspect of the clock. Sometimes it takes more than two seconds to piece together the disparate clues already being spewed at warp speed at you, and more than once I found myself missing key pieces of information (e.g. I didn't even hear the name of Myra St. Clair in that tossup on This Side of Paradise because it was being read so fast, I missed Hrothgar's name in the bonus part on Beowulf, etc.) because moderators were under pressure to read fast. We had great readers, but at some point you do sacrifice a certain amount of intelligibility for speed, and I would much rather we had a few extra minutes per round so that moderators wouldn't have to read themselves hoarse in every packet and so I could understand what's being said in questions.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance

Susan
Forums Staff: Administrator
Posts: 1827
Joined: Fri Aug 15, 2003 12:43 am

Re: timer delenda est

Post by Susan »

If, for whatever reason, losing the timer is not an option, I think that having a 20-question floor for all games (as Jeff has proposed on IRC and possibly on the boards somewhere as well) is an absolute necessity. It's ridiculous to have 12- or 13-tossup games deciding anything.
Susan
UChicago alum (AB 2003, PhD 2009)
Member emerita, ACF

User avatar
Cheynem
Sin
Posts: 6707
Joined: Tue May 11, 2004 11:19 am
Location: Grand Rapids, Michigan

Re: timer delenda est

Post by Cheynem »

Removing the clock would also (hopefully) allow some more room to put information in bonus parts (or, although I don't mind the shorter length, tossups).

I hate clocks. They're stupid. What's the justification here? Aside from pure capriciousness and some compulsion to "move things along" (which, as demonstrated, can be done without clocks), what's the deal here?
Mike Cheyne
Formerly U of Minnesota

"You killed HSAPQ"--Matt Bollinger

User avatar
naturalistic phallacy
Auron
Posts: 1416
Joined: Tue May 01, 2007 12:03 am
Location: Minneapolis, MN
Contact:

Re: timer delenda est

Post by naturalistic phallacy »

Cheynem wrote:Removing the clock would also (hopefully) allow some more room to put information in bonus parts (or, although I don't mind the shorter length, tossups).

I hate clocks. They're stupid. What's the justification here? Aside from pure capriciousness and some compulsion to "move things along" (which, as demonstrated, can be done without clocks), what's the deal here?
I agree with this sentiment. There doesn't seem to be much net benefit from having timers that can't be reproduced with a little prudence on the part of the readers and players.
Bernadette Spencer
University of Minnesota, MCTC
Member, ACF
Event Manager, PACE
Order Support and Administrative Assistant, NAQT

User avatar
marnold
Tidus
Posts: 705
Joined: Wed Jan 17, 2007 12:32 pm
Location: NY

Re: timer delenda est

Post by marnold »

I'm always surprised at the vitriol behind clock-hatred since I actually like the clock quite a bit. It fosters an atmosphere of not-fucking-around that I enjoy. On the clock, players and (not terrible) moderators avoid flapping their jaws between questions, which I like. It's so ridiculous to say that this could be accomplished just by prudence or whatever: I can count on one hand the number of players I know that don't talk between questions (god knows I'm not in that group). Quizbowl is a social activity, the people that play tournaments often know each other and that leads to chit chat, so people always, always, always talk between questions in untimed formats. This isn't bad, it's just a fact, and timing matches makes questions come faster and shit doesn't get bogged down. I also enjoy the gamesmanship aspects the clock introduces and I'm unpersuaded that it's a bad quizbowl gimmick that create fake excitement. Having "play faster to try to get to more questions" as a legitimate strategy adds a playing-style element that doesn't do anything to keep the game from rewarding knowledge and adds some legitimate excitement, at least to me and I'm guessing I'm not alone in that.

It seems that putting in an 18 or 20 tossup minimum would solve a lot of problems, but even more would be solved by not giving SCTs to sites that have no chance at staffing them competently.
Last edited by marnold on Sun Feb 07, 2010 4:34 pm, edited 1 time in total.
Michael Arnold
Chicago 2010
Columbia Law 2013

2009 ACF Nats Champion
2010 ICT Champion
2010 CULT Champion
Member of Mike Cheyne's Quizbowl All-Heel Team

Fundamental Theorem of Quizbowl (Revised): Almost no one is actually good at quizbowl.

User avatar
Mechanical Beasts
Banned Cheater
Posts: 5673
Joined: Thu Jun 08, 2006 10:50 pm

Re: timer delenda est

Post by Mechanical Beasts »

marnold wrote:It seems that putting in an 18 or 20 tossup minimum would solve a lot of problems, but even more would be solved by not giving SCTs to people that have no chance at staffing them competently.
I will stop you here and say that the "no SCT in the northeast ever" solution is probably the worst one; we staffed this SCT by a miracle of contortions. One solution is for NAQT to get realistic and say well, look: Brown and Harvard were in last year's top bracket at ACF Nationals; they'll probably qualify for ICT--give them both an autobid and let them all staff.
Andrew Watkins

User avatar
marnold
Tidus
Posts: 705
Joined: Wed Jan 17, 2007 12:32 pm
Location: NY

Re: timer delenda est

Post by marnold »

The lowest team in TUH at your site's D1 got through over 22 tossups and your tournament was widely and repeatedly lauded for its excellent logistics! I find it remarkably hard to believe that you just barely scraped by here, dude. I'm talking about more grisly sites - Indiana, for example. That said, giving bids to historically successful teams seems like a fine idea too.
Michael Arnold
Chicago 2010
Columbia Law 2013

2009 ACF Nats Champion
2010 ICT Champion
2010 CULT Champion
Member of Mike Cheyne's Quizbowl All-Heel Team

Fundamental Theorem of Quizbowl (Revised): Almost no one is actually good at quizbowl.

User avatar
Cheynem
Sin
Posts: 6707
Joined: Tue May 11, 2004 11:19 am
Location: Grand Rapids, Michigan

Re: timer delenda est

Post by Cheynem »

I think imposing something like "Ten Minutes," but at least 20 questions, would be a good start. I agree with Marnold in that I like how the clock keeps things moving, but I have played at tournaments that did not have this problem without the use of clocks.
Mike Cheyne
Formerly U of Minnesota

"You killed HSAPQ"--Matt Bollinger

User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: timer delenda est

Post by grapesmoker »

marnold wrote:I'm always surprised at the vitriol behind clock-hatred since I actually like the clock quite a bit. It fosters an atmosphere of not-fucking-around that I enjoy. On the clock, players and (not terrible) moderators avoid flapping their jaws between questions, which I like. It's so ridiculous to say that this could be accomplished just by prudence or whatever: I can count on one hand the number of players I know that don't talk between questions (god knows I'm not in that group). Quizbowl is a social activity, the people that play tournaments often know each other and that leads to chit chat, so people always, always, always talk between questions in untimed formats. This isn't bad, it's just a fact, and timing matches makes questions come faster and shit doesn't get bogged down. I also enjoy the gamesmanship aspects the clock introduces and I'm unpersuaded that it's a bad quizbowl gimmick that create fake excitement. Having "play faster to try to get to more questions" as a legitimate strategy adds a playing-style element that doesn't do anything to keep the game from rewarding knowledge and adds some legitimate excitement, at least to me and I'm guessing I'm not alone in that.
Marnold, I'd like to bring this year's Penn Bowl as evidence. If my recollection is correct, this was a 24-team tournament that started sometime around 9:30 and finished up around 7:30; this was including a one-game final game played between us and Harvard. For comparison, this year's SCT at Harvard started at around 9:30 and was done around 6:30, with an advantaged final taking place. I believe the same number of rounds were played in each tournament. So we're looking at a total difference of about an hour once all factors (chit-chat, delays, reader variability) are accounted for. Good moderators (of the variety that would be staffing a clock-free SCT) would keep the crosstalk to a minimum anyway. All the other shortcomings of the clock vastly outweigh, in my opinion, whatever 15 minutes or so you might conceivably save by having timed rounds.
It seems that putting in an 18 or 20 tossup minimum would solve a lot of problems, but even more would be solved by not giving SCTs to sites that have no chance at staffing them competently.
Who's going to host then? NAQT was already struggling to find locations. Again, the northeast is one of the most densely populated qb regions there is and Harvard is a perfect location, and yet no club here has the resources to run a tournament with two staffers in one room.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance

User avatar
Important Bird Area
Forums Staff: Administrator
Posts: 5609
Joined: Thu Aug 28, 2003 3:33 pm
Location: San Francisco Bay Area
Contact:

Re: timer delenda est

Post by Important Bird Area »

grapesmoker wrote:Who's going to host then? NAQT was already struggling to find locations.
This. It will be the foundation of my case to R. that either switching to untimed 20/20 or adopting a minimum-tossups-heard-floor will both 1. make it easier to find SCT hosts, which we recognize is a major ongoing problem and 2. limit the downside risk of novice moderators with limited training.
Jeff Hoppes
President, Northern California Quiz Bowl Alliance
former HSQB Chief Admin (2012-13)
VP for Communication and history subject editor, NAQT
Editor emeritus, ACF

"I wish to make some kind of joke about Jeff's love of birds, but I always fear he'll turn them on me Hitchcock-style." -Fred

User avatar
matt979
Lulu
Posts: 80
Joined: Wed Apr 13, 2005 10:22 pm
Location: Alameda, CA
Contact:

Re: timer delenda est

Post by matt979 »

myamphigory wrote:If, for whatever reason, losing the timer is not an option, I think that having a 20-question floor for all games (as Jeff has proposed on IRC and possibly on the boards somewhere as well) is an absolute necessity. It's ridiculous to have 12- or 13-tossup games deciding anything.
Quoted For Truth.

I'm eager to see the breakdown of tossups per game per location this year, but now also a little frightened. (12?)

Novice/uneven moderators (and their unavoidable use at any tournament below national championship level) strongly inform my opinion of optimal tossup/bonus length, but I have to admit that's a separate issue from game format. "The slow room" is a lesser evil than "the 16-tossup room."

Matt

(n.b. I've been one of NAQT's most vociferous clock apologists, largely for the reasons MArnold already gave.)
Matt Bruce
Harvard '96, Boston University School of Law '99

User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: timer delenda est

Post by grapesmoker »

matt979 wrote:(n.b. I've been one of NAQT's most vociferous clock apologists, largely for the reasons MArnold already gave.)
Given that those reasons are based on very little actual empirical evidence, I hope you're ready to be convinced otherwise.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance

User avatar
Mike Bentley
Auron
Posts: 5945
Joined: Fri Mar 31, 2006 11:03 pm
Location: Bellevue, WA
Contact:

Re: timer delenda est

Post by Mike Bentley »

What are people's thought on the use of clocks at ICT? Here it's very rare that you'll hear less than 20 tossups per round as the moderator quality is a lot better.
Mike Bentley
VP of Editing, Partnership for Academic Competition Excellence
Adviser, Quizbowl Team at University of Washington
University of Maryland, Class of 2008

User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: timer delenda est

Post by grapesmoker »

Bentley Like Beckham wrote:What are people's thought on the use of clocks at ICT? Here it's very rare that you'll hear less than 20 tossups per round as the moderator quality is a lot better.
No clocks! Anywhere! I've had some decent ICT mods and I've had some bad ones too, but the problem of the clock goes deeper than just moderator variance.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance

User avatar
Not That Kind of Christian!!
Yuna
Posts: 847
Joined: Mon Feb 26, 2007 10:36 pm
Location: Manhattan

Re: timer delenda est

Post by Not That Kind of Christian!! »

grapesmoker wrote:
Bentley Like Beckham wrote:What are people's thought on the use of clocks at ICT? Here it's very rare that you'll hear less than 20 tossups per round as the moderator quality is a lot better.
No clocks! Anywhere! I've had some decent ICT mods and I've had some bad ones too, but the problem of the clock goes deeper than just moderator variance.
As Jerry pointed out, even if you assume a tournament with moderators of the decent or better level, the clock still messes with moderator quality. I consider myself a pretty solid moderator, but the pressure to get through more tossups in order to ensure the best possible outcome for the game caused me to do things like, say, read Myra St. Clair's name too quickly (sorry, Jerry) for teams listening to really comprehend the tossup. I heard a couple other known-to-be-good mods at the Harvard SCT site comment about similar effects they were noticing on their own reading, particularly as rounds wore on.
Hannah Kirsch
Brandeis University 2010
NYU School of Medicine 2014

"Wow, those Scandinavians completely thorbjorned my hard-earned political capital."

User avatar
marnold
Tidus
Posts: 705
Joined: Wed Jan 17, 2007 12:32 pm
Location: NY

Re: timer delenda est

Post by marnold »

grapesmoker wrote:
matt979 wrote:(n.b. I've been one of NAQT's most vociferous clock apologists, largely for the reasons MArnold already gave.)
Given that those reasons are based on very little actual empirical evidence, I hope you're ready to be convinced otherwise.
I mean, what would this kind of empirical evidence even look like? A lot of reasons for liking the clock are subjective: I think the atmosphere it fosters is fun and a worthwhile change of pace from the majority untimed formats. You don't share those opinions, obviously, and clearly you hate the clock way more than I like it, but it doesn't change the fact that there are benefits to the clock that aren't going to be captured by empirical evidence. But, even on your terms, objective benefits might come in the form of shaving off an hour from a tournament (see your Penn Bowl example). This seems like a non-trivial amount of time to me. Right now it comes at a trade off, but there are policy choices NAQT could make that would reduce the associated costs: a TUH floor would keep games from being totally unfair, for example, and I'm sure there are ways to increase the number of competent staff (higher staff discounts, having sites planned further in advance, granting autobids from performance during the year or at previous ICTs that would free up experienced players to read, etc.).
Michael Arnold
Chicago 2010
Columbia Law 2013

2009 ACF Nats Champion
2010 ICT Champion
2010 CULT Champion
Member of Mike Cheyne's Quizbowl All-Heel Team

Fundamental Theorem of Quizbowl (Revised): Almost no one is actually good at quizbowl.

User avatar
nobthehobbit
Rikku
Posts: 293
Joined: Sun Feb 19, 2006 1:18 am

Re: timer delenda est

Post by nobthehobbit »

I agree with what's said above. I've elsewhere mentioned that the one time I played SCT, I also scorekept to make rounds go longer than 19 questions. I also staffed two SCTs; both times I didn't have buzzers in my room (slapbowl, yay!) but did have a scorekeeper and was able to go through 23-24 tossups per round.

If the decision is made to keep the buzzer, I hope that the policy of guaranteed minimums for TUH is adopted.
Daniel Pareja, Waterloo, Canadian quizbowl iconoclast

Stats zombie.
William Lyon Mackenzie King wrote:There are few men in this Parliament for whom I have greater respect than the leader of the Co-operative Commonwealth Federation. I admire him in my heart, because time and again he has had the courage to say what lays on his conscience, regardless of what the world might think of him. A man of that calibre is an ornament to any Parliament.

User avatar
Not That Kind of Christian!!
Yuna
Posts: 847
Joined: Mon Feb 26, 2007 10:36 pm
Location: Manhattan

Re: timer delenda est

Post by Not That Kind of Christian!! »

I'm a little confused as to how keeping the timer but creating a x TUs per game minimum is a good idea. Wouldn't this make rounds go at least as long as they would go without a timer, thus, well, making the clock moot anyway?
Hannah Kirsch
Brandeis University 2010
NYU School of Medicine 2014

"Wow, those Scandinavians completely thorbjorned my hard-earned political capital."

User avatar
nobthehobbit
Rikku
Posts: 293
Joined: Sun Feb 19, 2006 1:18 am

Re: timer delenda est

Post by nobthehobbit »

Not That Kind of Christian!! wrote:I'm a little confused as to how keeping the timer but creating a x TUs per game minimum is a good idea. Wouldn't this make rounds go at least as long as they would go without a timer, thus, well, making the clock moot anyway?
Keeping the timer means that teams that play quickly (provided the mod is a sufficiently fast reader) get to hear more questions; I believe this was mentioned as a benefit of the clock. Putting in the TUH minimum keeps there from being too much variability between games (if there's 24 TU per pack and a 20 TU minimum, say), and I imagine that most often low-TUH games are due to readers being slow, or at least that's the impression I get from what I'm reading here (and coincides with what I recall of the SCTs in which I've participated).
Daniel Pareja, Waterloo, Canadian quizbowl iconoclast

Stats zombie.
William Lyon Mackenzie King wrote:There are few men in this Parliament for whom I have greater respect than the leader of the Co-operative Commonwealth Federation. I admire him in my heart, because time and again he has had the courage to say what lays on his conscience, regardless of what the world might think of him. A man of that calibre is an ornament to any Parliament.

User avatar
Not That Kind of Christian!!
Yuna
Posts: 847
Joined: Mon Feb 26, 2007 10:36 pm
Location: Manhattan

Re: timer delenda est

Post by Not That Kind of Christian!! »

nobthehobbit wrote:
Not That Kind of Christian!! wrote:I'm a little confused as to how keeping the timer but creating a x TUs per game minimum is a good idea. Wouldn't this make rounds go at least as long as they would go without a timer, thus, well, making the clock moot anyway?
Keeping the timer means that teams that play quickly (provided the mod is a sufficiently fast reader) get to hear more questions; I believe this was mentioned as a benefit of the clock. Putting in the TUH minimum keeps there from being too much variability between games (if there's 24 TU per pack and a 20 TU minimum, say), and I imagine that most often low-TUH games are due to readers being slow, or at least that's the impression I get from what I'm reading here (and coincides with what I recall of the SCTs in which I've participated).
Yes, so, say there's a 20-TU minimum, and say a reader is slow enough that only 14 tossups get heard in the timed 20-minute period. That means 6 more tossups still have to be read, which means that the timer might as well not have been there.

Basically, in the event of a tournament with all good readers who can get through at least 20 (or 22, if that's the limit) tossups per round, fine, no big. But since a bracket can only go as fast as its slowest reader, in the event that a tournament has even one reader who cannot do this, the timer-plus-rule configuration makes no sense.

And now that I think about it, if your entire staff can get through at least 20 TU in the 20 minutes... yeah, there's still no need for the rule and the timer.
Hannah Kirsch
Brandeis University 2010
NYU School of Medicine 2014

"Wow, those Scandinavians completely thorbjorned my hard-earned political capital."

User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: timer delenda est

Post by grapesmoker »

marnold wrote:I mean, what would this kind of empirical evidence even look like?
Well, it would look a lot like Harvard being unable to staff its SCT without outside help, for one.
A lot of reasons for liking the clock are subjective: I think the atmosphere it fosters is fun and a worthwhile change of pace from the majority untimed formats. You don't share those opinions, obviously, and clearly you hate the clock way more than I like it, but it doesn't change the fact that there are benefits to the clock that aren't going to be captured by empirical evidence.
A large part of why I hate the clock is subjective too. The clock puts extra pressure on readers to get through more questions, leading to a less-comprehensible reading style in which even the best moderators sacrifice intelligibility for speed. It also shortens the time you have to answer after buzzing and usually ends up shortening the time you have on bonuses as well because again, moderators feel rushed (and moderator timing judgments are not protestable).

I'm a purist, I guess; I want the knowledge content of the question to be the only determining factor in a game and I hate anything that interferes with that. The clock does, so I'd rather it weren't there.
But, even on your terms, objective benefits might come in the form of shaving off an hour from a tournament (see your Penn Bowl example). This seems like a non-trivial amount of time to me.
What my Penn Bowl example shows is that the upper bound on the time you save from the clock is one hour and that's if you attribute all time saved solely to the clock. Consider that Penn Bowl had rebracketing and SCT did not, since we played a single RR in DII and a double RR in DI. Consider also that moderators were, as a rule, more experienced at SCT. Once you start adding those things up I think you'll find that the time saved by the clock is very minimal. It's not a trivial amount of time and might be close to half an hour, but it's really not a whole lot in the grand scheme of things.
Right now it comes at a trade off, but there are policy choices NAQT could make that would reduce the associated costs: a TUH floor would keep games from being totally unfair, for example, and I'm sure there are ways to increase the number of competent staff (higher staff discounts, having sites planned further in advance, granting autobids from performance during the year or at previous ICTs that would free up experienced players to read, etc.).
Hannah already covered the issues with the tossup floor proposal. All I have to add is that I view the clock as unnecessarily accelerating an already fast-paced format, which interferes with my performance and ability to think. Obviously I have a vested interest in this, but I don't think I'm nearly the only one who feels this way.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance

ScoBo
Wakka
Posts: 239
Joined: Wed Jan 10, 2007 5:05 pm
Location: Kansas City area
Contact:

Re: timer delenda est

Post by ScoBo »

The clock can lead to moderator mistakes as well. I'm not sure how much my quick reading affected teams' understanding of the questions (I didn't hear any complaints, at least), but the two times I accidentally revealed the answer early were likely caused by me trying to get through questions faster because of the clock.

At Missouri State, DI teams heard an average of 21.64 tossups per game and DII teams heard 22.95 tossups per game. Only one DI game and two DII games (coincidentally between the same two teams) had less than 20 tossups read (all 3 had 19), so in our case the length of the tournament probably wouldn't have been affected much if there were no clock. Of course, we only had 6 rooms, everyone had a scorekeeper, and all of the moderators were quite experienced.

I personally make it a priority to keep chatter to a minimum, particularly by announcing "tossup" as soon as possible (while I'm switching from the bonuses to the tossups), and for the most part teams will immediately stop talking with or without a clock. As long as moderators are able to keep the natural inter-question chatter to a minimum, games can move quickly enough without a clock with the added benefit of keeping the number of tossups heard consistent between games.
Jeffrey Hill • Missouri Quizbowl Alliance president • UMR/Missouri S&T 2009 • Liberty (MO) 2005
Post your tournaments, SQBS reports, and question sets to the Quizbowl Resource Center Database!

User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: timer delenda est

Post by grapesmoker »

When I read, I tell teams to shut up all the time. And lo and behold, I usually get through a round pretty quickly, clock or no.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance

User avatar
The Friar
Wakka
Posts: 158
Joined: Fri Jul 10, 2009 2:39 pm

Re: timer delenda est

Post by The Friar »

grapesmoker wrote:If my recollection is correct, this was a 24-team tournament that started sometime around 9:30 and finished up around 7:30; this was including a one-game final game played between us and Harvard. For comparison, this year's SCT at Harvard started at around 9:30 and was done around 6:30, with an advantaged final taking place. I believe the same number of rounds were played in each tournament. So we're looking at a total difference of about an hour once all factors (chit-chat, delays, reader variability) are accounted for.
Don't leave out the big one, which is tossup length. NAQT tossups average, what, 75% the length of ACF tossups, and maybe even a smaller percentage of the length of tossups in wild and wooly circuit tournaments with no national editor cracking the "six lines of 10-point TNR" whip? As
cheynem wrote:Removing the clock would also (hopefully) allow some more room to put information in bonus parts (or, although I don't mind the shorter length, tossups).
This is, at first glance, a reason to support the clock-and-floor rule: if tossups are going to be this short I would like to hear more than 20 of them per game, as long as that isn't gonna take any longer than reading 20 ACF or circuit questions. However, the same end could well be accomplished by just declaring that the NAQT game is, say, 22 cycles long, without a clock.
grapesmoker wrote:With the clock absent and every team guaranteed X questions, there is no longer an excuse for not having identical per-X-questions distribution in each packet and no excuse for putting awful questions in the back in the hopes that the moderators won't get to them.
Actually, the improvement would be more fundamental than that: in Twenty20 NAQT (or 22/22 or whatever), there is no excuse for the most awful questions getting produced at all. This SCT is a great example of where questions were included that their authors, let alone the editor and players, described as "filler". These questions were not at all limited to the ends of packets; they were throughout. Well, those marginal questions don't have to be included in packets at all if the packets themselves are simply less long.
grapesmoker wrote:When I read, I tell teams to shut up all the time. And lo and behold, I usually get through a round pretty quickly, clock or no.
I have had less success with doing this. You're a more important quizbowl personality than I am, and I think that makes a difference. I think not just everybody gets to tell people to shut up.
Gordon Arsenoff
Rochester '06
WUStL '14 (really)

Developer of WUStL Updates Statistics Live!

User avatar
The Friar
Wakka
Posts: 158
Joined: Fri Jul 10, 2009 2:39 pm

Re: timer delenda est

Post by The Friar »

Also, D-value type stats that use a version of points per tossup heard, adjusted only for opponent strength but not individual question difficulty, are biased if teams do not all hear exactly the same tossups. Systems using question-level data, such as FRIAR, would not be (asymptotically) biased in this way, but would be inefficient, because data would be more sparse on some tossups than others.
Gordon Arsenoff
Rochester '06
WUStL '14 (really)

Developer of WUStL Updates Statistics Live!

OntarioQuizzer
Lulu
Posts: 79
Joined: Thu Nov 06, 2008 3:29 pm
Contact:

Re: timer delenda est

Post by OntarioQuizzer »

I bet that teams would shut up quickly if you started applying illegal conferral penalties, no?
Andy Saunders
formerly Brock University
(former) general helper, University of Guelph quiz bowl

Susan
Forums Staff: Administrator
Posts: 1827
Joined: Fri Aug 15, 2003 12:43 am

Re: timer delenda est

Post by Susan »

The Friar wrote:
grapesmoker wrote:When I read, I tell teams to shut up all the time. And lo and behold, I usually get through a round pretty quickly, clock or no.
I have had less success with doing this. You're a more important quizbowl personality than I am, and I think that makes a difference. I think not just everybody gets to tell people to shut up.
If you start reading the next tossup very loudly (loudly enough to cut through background noise), people usually shut up pretty quickly. If they don't, pause the clock (if applicable) and tell them to shut up or you'll neg them for conferring. This always works for me (and I very rarely have had to go to step 2).
Susan
UChicago alum (AB 2003, PhD 2009)
Member emerita, ACF

User avatar
Mechanical Beasts
Banned Cheater
Posts: 5673
Joined: Thu Jun 08, 2006 10:50 pm

Re: timer delenda est

Post by Mechanical Beasts »

OntarioQuizzer wrote:I bet that teams would shut up quickly if you started applying illegal conferral penalties, no?
"Start potentially altering match outcomes if you have no other way to keep control of your room" is something I don't want to say to moderators.
Andrew Watkins

User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: timer delenda est

Post by grapesmoker »

It doesn't take much to get people to be quiet besides starting to read the next question. Bark at them a few times and they'll figure it out, really. It's not that hard and you don't need to be a "quizbowl personality" to do it, you just need to be moderately assertive and just a little bit loud.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance

User avatar
Mechanical Beasts
Banned Cheater
Posts: 5673
Joined: Thu Jun 08, 2006 10:50 pm

Re: timer delenda est

Post by Mechanical Beasts »

So here's where the timer has any benefits: inexperienced mods who can't control a room. Here's where the timer has all of its glaring, horrifying deficiencies: inexperienced mods. The twenty (or twenty-two, or whatever) question floor is a concession to the clock; it's a way to say "here is an unnecessary element that introduces irregularity into the game, but here's one way to control one aspect of that irregularity). Let's not have a floor; let's have no clock.

I consider NAQT's philosophy of having more, shorter questions in packets rather than fewer, longer ones to be "different" enough; a timer doesn't make NAQT too much like the rest of the circuit. (They may well be right that it does a better job predicting who's better; I don't know enough about statistics to tell (or be able to tell how to tell). And that's cool! We don't need gimmickry.)
Andrew Watkins

User avatar
Cheynem
Sin
Posts: 6707
Joined: Tue May 11, 2004 11:19 am
Location: Grand Rapids, Michigan

Re: timer delenda est

Post by Cheynem »

As an educator, I've found that the simplest way to get students to shut up is to attempt to relay important information to them. Invariably, a student will realize what is going on and will begin a chain reaction of telling other students to shut up. Simply beginning to read the next question, in my experience, causes this to happen. Even if talking is still lingering and going on, it will be shushed and if people complain that they didn't hear the lead-in, well that's why they should shut up.

The other thing I've learned as an educator is never give students an opportunity to uncontrollably talk. As a moderator never give them a chance to do so--if you start engaging in banal chatter, even just once, you have the opened the door to disaster. I mean, I'm not saying be a robot, but if you move professionally, I think the game will go professionally.
Mike Cheyne
Formerly U of Minnesota

"You killed HSAPQ"--Matt Bollinger

User avatar
marnold
Tidus
Posts: 705
Joined: Wed Jan 17, 2007 12:32 pm
Location: NY

Re: timer delenda est

Post by marnold »

The twenty (or twenty-two, or whatever) question floor is a concession to the clock; it's a way to say "here is an unnecessary element that introduces irregularity into the game, but here's one way to control one aspect of that irregularity). Let's not have a floor; let's have no clock.
That's not an argument! There's nothing "irregular" about having a clock decide matches. This is the same mindset that thinks that the distribution is somehow sacrosanct and not realizing that there are lots of things that are totally arbitrary about the game we play: delimiting matches by time is exactly as arbitrary as having 20 questions or 22 questions or whatever. The problem of the clock is sometimes there are too-few questions and so the floor is a strategy to address that; it's analogous to yelling at teams to shut up to fix the problem of the untimed format (namely, yammering). Whatever, it's pretty obvious that I'm going to end up on the wrong side of history on this one, but let's stop pretending this is a purging of something that's obviously bad quizbowl.
Michael Arnold
Chicago 2010
Columbia Law 2013

2009 ACF Nats Champion
2010 ICT Champion
2010 CULT Champion
Member of Mike Cheyne's Quizbowl All-Heel Team

Fundamental Theorem of Quizbowl (Revised): Almost no one is actually good at quizbowl.

User avatar
Mechanical Beasts
Banned Cheater
Posts: 5673
Joined: Thu Jun 08, 2006 10:50 pm

Re: timer delenda est

Post by Mechanical Beasts »

No, dude, there's plenty irregular about the fact that if a game lasts n questions and the question n+1 might be the fourth lit tossup. Or it lasts n questions and question n was the fifth lit tossup. When the distribution depends in part on who your reader is, see, then you have problems.
Andrew Watkins

User avatar
marnold
Tidus
Posts: 705
Joined: Wed Jan 17, 2007 12:32 pm
Location: NY

Re: timer delenda est

Post by marnold »

But that's just restating the premise of the debate: one notion says that matches should end after some time and you say that it should end when a particular number of questions (within a particular distribution) have been read. You think that the former is bad quizbowl and the latter is good quizbowl and I don't see any reason why that's true. ACF rounds have an R and an M in one round an M and a P in the next - I don't see why the round-to-round "irregularity" introduced by a moderator is different from the round-to-round "irregularity" introduced by an editor.
Michael Arnold
Chicago 2010
Columbia Law 2013

2009 ACF Nats Champion
2010 ICT Champion
2010 CULT Champion
Member of Mike Cheyne's Quizbowl All-Heel Team

Fundamental Theorem of Quizbowl (Revised): Almost no one is actually good at quizbowl.

User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: timer delenda est

Post by grapesmoker »

marnold wrote:But that's just restating the premise of the debate: one notion says that matches should end after some time and you say that it should end when a particular number of questions (within a particular distribution) have been read. You think that the former is bad quizbowl and the latter is good quizbowl and I don't see any reason why that's true. ACF rounds have an R and an M in one round an M and a P in the next - I don't see why the round-to-round "irregularity" introduced by a moderator is different from the round-to-round "irregularity" introduced by an editor.
The RMP thing you're talking about is mostly, in my view, an artifact of doing things like trying to make space for geography and "your choice" in the distribution. If RMP is considered a logical grouping (which I'm skeptical about) and there's only space for 2/2 from that category, that's what will happen. I don't think it's as egregious as having a round run out with only 3 literature and 2 science tossups being heard if your team is aces at literature and science. This certainly won't happen in any ACF tournament because 4/4 of the big 3 is guaranteed during every match. For what it's worth, I'm a strong advocate of a full 3/3 RMP distro and the abolition of geography or its absorption into social science (as I've argued elsewhere). The inconsistency being tolerated in ACF packets is much smaller and more fine-grained than what's going on in the timed rounds.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance

User avatar
nobthehobbit
Rikku
Posts: 293
Joined: Sun Feb 19, 2006 1:18 am

Re: timer delenda est

Post by nobthehobbit »

marnold wrote:But that's just restating the premise of the debate: one notion says that matches should end after some time and you say that it should end when a particular number of questions (within a particular distribution) have been read. You think that the former is bad quizbowl and the latter is good quizbowl and I don't see any reason why that's true. ACF rounds have an R and an M in one round an M and a P in the next - I don't see why the round-to-round "irregularity" introduced by a moderator is different from the round-to-round "irregularity" introduced by an editor.
Because there's two sources of irregularity in the SCT set: that introduced by an editor (which packets will have an Egyptian mythology question?) and that introduced by the moderator (who can read every tossup in the pack with 2 minutes left at the end, and who can only get through 14?) An untimed set will have only the one source of irregularity, that introduced by the editor.

EDIT: Jerry says very similar things to this, and more specific things.
Daniel Pareja, Waterloo, Canadian quizbowl iconoclast

Stats zombie.
William Lyon Mackenzie King wrote:There are few men in this Parliament for whom I have greater respect than the leader of the Co-operative Commonwealth Federation. I admire him in my heart, because time and again he has had the courage to say what lays on his conscience, regardless of what the world might think of him. A man of that calibre is an ornament to any Parliament.

User avatar
Mechanical Beasts
Banned Cheater
Posts: 5673
Joined: Thu Jun 08, 2006 10:50 pm

Re: timer delenda est

Post by Mechanical Beasts »

marnold wrote:But that's just restating the premise of the debate: one notion says that matches should end after some time and you say that it should end when a particular number of questions (within a particular distribution) have been read. You think that the former is bad quizbowl and the latter is good quizbowl and I don't see any reason why that's true. ACF rounds have an R and an M in one round an M and a P in the next - I don't see why the round-to-round "irregularity" introduced by a moderator is different from the round-to-round "irregularity" introduced by an editor.
The one can be controlled by a sensible editor attempting to provide some measure of balance and fairness in the face of packet constraints. The other is totally at the whim of the packet randomizer.

Besides, this isn't a referendum on the ACF distribution; I could well prefer a consistent round to round ACF distribution, to whatever extent possible; I could also recognize that no distribution except one of infinite size is truly consistent round to round because it's a coincidence that "Science" and "Geography" are a top-level categories instead of "Organic Chemistry," "Computational Chemistry," and "Dumb Shit" (and rounds that are consistent in the first system can be inconsistent in the latter).
Andrew Watkins

User avatar
cvdwightw
Auron
Posts: 3446
Joined: Tue May 13, 2003 12:46 am
Location: Southern CA
Contact:

Re: timer delenda est

Post by cvdwightw »

Crazy Andy Watkins wrote:No, dude, there's plenty irregular about the fact that if a game lasts n questions and the question n+1 might be the fourth lit tossup. Or it lasts n questions and question n was the fifth lit tossup. When the distribution depends in part on who your reader is, see, then you have problems.
I just want to point out that it's not just the distribution, but the very outcome of the game, that can potentially affect who wins a game. In the game we beat UCSD by 5 points on, we were down 50 with less than a minute left. We grabbed the next tossup, went into speed-up mode on the bonus, and with 15-20 seconds left we passed in the middle of Mik reading the third part of the bonus so that we could ensure that we could get to tossup 26 and go from a sure 20-point loss to a chance at winning. Clock strategy, a clutch power by Carl, and a fortunate biology bonus netted us the win. Put that match in 19-questions-per-game Steve Katz's room and UCSD wins by about 100.

I understand and enjoy the idea of clock strategy playing a minor but occasionally significant role in who wins the match, but the idea that the best team after {20, 21, 22, 23, 24, 25, all 26} questions does not win because the moderator only gets through 18 (or 12 in some cases!) is why I'm not a fan of the clock.

Regarding the distribution-timer issue, I for one would like to see NAQT specifically come out and say "we guarantee that 17/17 (or some other sufficiently large percentage) of the first 20/20 will be from these academic categories in this distribution." For instance, if NAQT has 4.2 literature questions per round, NAQT should state that 4 of those questions are guaranteed in the first 20 and the fifth (if it exists) will either be one of the 3/3 "our choice" in the first 20/20 or somewhere between question 21 and 26. I don't have any problem at all with the large amount of non-academic or questionably-academic categories in the NAQT distribution; what I do have a problem with is when one team gets screwed over due to the randomization of the packet and/or the speed of the moderator. Also, this allows NAQT to maintain its per-tournament quotas while ensuring that weird packet issues like "we only heard 2 science questions in 19 tossups" don't occur.
Dwight Wynne
socalquizbowl.org
UC Irvine 2008-2013; UCLA 2004-2007; Capistrano Valley High School 2000-2003

"It's a competition, but it's not a sport. On a scale, if football is a 10, then rowing would be a two. One would be Quiz Bowl." --Matt Birk on rowing, SI On Campus, 10/21/03

"If you were my teammate, I would have tossed your ass out the door so fast you'd be emitting Cerenkov radiation, but I'm not classy like Dwight." --Jerry

Susan
Forums Staff: Administrator
Posts: 1827
Joined: Fri Aug 15, 2003 12:43 am

Re: timer delenda est

Post by Susan »

What would people think about dropping the clock, making games 26/26 (or 24/24, or something), but keeping the question length and the time to answer the same (i.e. both shorter than those of ACF)?
Susan
UChicago alum (AB 2003, PhD 2009)
Member emerita, ACF

User avatar
Mechanical Beasts
Banned Cheater
Posts: 5673
Joined: Thu Jun 08, 2006 10:50 pm

Re: timer delenda est

Post by Mechanical Beasts »

myamphigory wrote:What would people think about dropping the clock, making games 26/26 (or 24/24, or something), but keeping the question length and the time to answer the same (i.e. both shorter than those of ACF)?
I would still prefer some elements of ACF format, but I would quite like this!
Andrew Watkins

User avatar
Cheynem
Sin
Posts: 6707
Joined: Tue May 11, 2004 11:19 am
Location: Grand Rapids, Michigan

Re: timer delenda est

Post by Cheynem »

I actually have no problem with the question length (for tossups) and I really don't have a problem with the time to answer thing.
Mike Cheyne
Formerly U of Minnesota

"You killed HSAPQ"--Matt Bollinger

User avatar
Not That Kind of Christian!!
Yuna
Posts: 847
Joined: Mon Feb 26, 2007 10:36 pm
Location: Manhattan

Re: timer delenda est

Post by Not That Kind of Christian!! »

Crazy Andy Watkins wrote:
myamphigory wrote:What would people think about dropping the clock, making games 26/26 (or 24/24, or something), but keeping the question length and the time to answer the same (i.e. both shorter than those of ACF)?
I would still prefer some elements of ACF format, but I would quite like this!
Agreed; these are the "differences" about NAQT that are not necessarily detrimental and potentially (and increasingly often!) interesting.

Dwight's story perfectly illustrates why the clock is an issue. Ditch it.
Hannah Kirsch
Brandeis University 2010
NYU School of Medicine 2014

"Wow, those Scandinavians completely thorbjorned my hard-earned political capital."

User avatar
grapesmoker
Sin
Posts: 6368
Joined: Sat Oct 25, 2003 5:23 pm
Location: NYC
Contact:

Re: timer delenda est

Post by grapesmoker »

myamphigory wrote:What would people think about dropping the clock, making games 26/26 (or 24/24, or something), but keeping the question length and the time to answer the same (i.e. both shorter than those of ACF)?
This is in essence what I'm advocating. I would be very happy with this.
Jerry Vinokurov
ex-LJHS, ex-Berkeley, ex-Brown, sorta-ex-CMU
code ape, loud voice, general nuissance

User avatar
The Friar
Wakka
Posts: 158
Joined: Fri Jul 10, 2009 2:39 pm

Re: timer delenda est

Post by The Friar »

myamphigory wrote:
The Friar wrote:
grapesmoker wrote:When I read, I tell teams to shut up all the time. And lo and behold, I usually get through a round pretty quickly, clock or no.
I have had less success with doing this. You're a more important quizbowl personality than I am, and I think that makes a difference. I think not just everybody gets to tell people to shut up.
If you start reading the next tossup very loudly (loudly enough to cut through background noise), people usually shut up pretty quickly. If they don't, pause the clock (if applicable) and tell them to shut up or you'll neg them for conferring. This always works for me (and I very rarely have had to go to step 2).
grapesmoker wrote:When I read
ITT I don't do what Jerry writes about doing in games. So, you're talking about moderating, right. This jumped out at me soon as I posted something, walked away, and came back.
Gordon Arsenoff
Rochester '06
WUStL '14 (really)

Developer of WUStL Updates Statistics Live!

User avatar
The Friar
Wakka
Posts: 158
Joined: Fri Jul 10, 2009 2:39 pm

Re: timer delenda est

Post by The Friar »

marnold wrote:There's nothing "irregular" about having a clock decide matches. [...] delimiting matches by time is exactly as arbitrary as having 20 questions or 22 questions or whatever.
marnold wrote:But that's just restating the premise of the debate: one notion says that matches should end after some time and you say that it should end when a particular number of questions (within a particular distribution) have been read. You think that the former is bad quizbowl and the latter is good quizbowl and I don't see any reason why that's true.
The Friar wrote:Also, D-value type stats that use a version of points per tossup heard, adjusted only for opponent strength but not individual question difficulty, are biased if teams do not all hear exactly the same tossups. Systems using question-level data, such as FRIAR, would not be (asymptotically) biased in this way, but would be inefficient, because data would be more sparse on some tossups than others.
Timed matches and PANTS are incompatible.*
myamphigory wrote:What would people think about dropping the clock, making games 26/26 (or 24/24, or something), but keeping the question length and the time to answer the same (i.e. both shorter than those of ACF)?
The Friar wrote:This SCT is a great example of where questions were included that their authors, let alone the editor and players, described as "filler". These questions were not at all limited to the ends of packets; they were throughout. Well, those marginal questions don't have to be included in packets at all if the packets themselves are simply less long.
No clock, same question length, games of 22/22 ± 2/2. Writing 26 good TUs and bonuses per packet hasn't proven replicatable yet.

*EDIT: I will play next year's SCT in a kilt if NAQT is still using both.
Gordon Arsenoff
Rochester '06
WUStL '14 (really)

Developer of WUStL Updates Statistics Live!

Locked