Looking for practice material?

Find thousands of free archived packets for practice and study from the Quizbowl Packet Archive!

MSHSAA/tossup-bonus format discussion

Discussion of quizbowl topics not related to specific tournaments
Post Reply
FishyFreshman
Posts: 501
Joined: Sun Apr 15, 2007 12:00 am

MSHSAA/tossup-bonus format discussion

Post by FishyFreshman »

CherryCokeStain wrote:This does however point out a flaw in the MSHSAA format that not all tossups are rewarded with bonuses, and it really doesn't make any difference what happens in the 1st and 3rd quarters, because those tossups are meaningless without an attached bonus.

I also don't want to discredit the questions, which were just fine. But I just really hope that Missouri switches to 20-20 like the rest of the quizbowl universe, for this reason, and the fact that the MSHSAA games are endlessly long, and without competent moderators, cannot be done with good questions.
This. For sure.

Charbroil
Posts: 1039
Joined: Fri Nov 16, 2007 12:00 am

MSHSAA/tossup-bonus format discussion

Post by Charbroil »

I would actually argue that the MSHSAA format is at its best in situations like these because it keeps close games from hinging on one or two tossups. At the very least, I wouldn't support going to a 20/20 format with nonrebounding bonuses, because in those kinds of situations, a weaker team that snipes off one or two key tossups can win a game.

In any case, Grant, we only answered 3 fewer tossups than you did, and we won by 75 points, so it's not as if playing in 20/20 format would have ensured a victory for you--other than the fact that we played very badly 1st quarter and those mistakes would have had much greater weight. The only real difference here is that we made our mistakes 1st quarter and you made them 2nd and 4th.

Finally, didn't we have a higher bonus conversion that game? I don't know how bonus conversion works for MSHSAA games in SQBS, but that's what the statistics tell me...

Of course, I do agree that using good questions on MSHSAA's format requires competent moderators--there's no question of that. For that format to be legitimate, MSHSAA has to start finding/training better moderators.

User avatar
Charlie Dees
Posts: 4134
Joined: Wed Jul 26, 2006 12:00 am
Location: Columbia, MO

MSHSAA/tossup-bonus format discussion

Post by Charlie Dees »

The bonus conversion statistics on that page are complete irrelevant nonsense so don't bother trying to argue from it because it counts out of 50 tossups, not the 20 that are available for bonuses.

Charbroil
Posts: 1039
Joined: Fri Nov 16, 2007 12:00 am

MSHSAA/tossup-bonus format discussion

Post by Charbroil »

Charbroil wrote:In any case, Grant, we only answered 3 fewer tossups than you did, and we won by 75 points, so it's not as if playing in 20/20 format would have ensured a victory for you--other than the fact that we played very badly 1st quarter and those mistakes would have had much greater weight. The only real difference here is that we made our mistakes 1st quarter and you made them 2nd and 4th.
Charlie mentioned that this statement also somewhat backs Grant's argument, and I would like to clarify that when I say quarter 1 as opposed to 2 and 4, I meant that quarter 1 falls earlier in the match, while in contrast we wouldn't get to half of the tossups in quarter 2 or any of the ones in quarter 4 in a 20/20 match.

Admittedly, there is an issue of how some tossups are arbitrarily weighted more than others, but I don't really see the problem with that, since it doesn't inhibit a team's ability to test deep knowledge and applies equally to both teams. In contrast, tossups are inherently an issue of speed to a degree (because there are always going to be some buzzer races), but having more of them keeps one randomly faster buzz by an inferior team from skewing the game.

I'm thinking about it from a statistics perspective, where more samples (tossups) mean less variation (upsets) in determining a quantity (knowledge of a team)

Of course, I know that you can also mitigate this issue by simply allowing bonuses to rebound, and I'm not preferring MSHSAA to 20/20. I think that on equally good questions (logistics issues notwithstanding), MSHSAA's 50 tossup format with some tossups having bonuses is equally good at preventing upsets as the 20/20 format with rebounding bonuses (various other issues such as subject distribution notwithstanding).

FishyFreshman
Posts: 501
Joined: Sun Apr 15, 2007 12:00 am

MSHSAA/tossup-bonus format discussion

Post by FishyFreshman »

What if all of the science questions fall into the bonus quarters? Or math, or lit, or anything?

Charbroil
Posts: 1039
Joined: Fri Nov 16, 2007 12:00 am

MSHSAA/tossup-bonus format discussion

Post by Charbroil »

Well that's the point of a random distribution! :) (Actually, I think MSHSAA has restrictions on how much of each topic falls into each quarter)

In any case, it's the same way with any other format--what if all of the history tossups happen to be about military history? Neither of those situations is particularly plausible...

And again, I'm not saying that MSHSAA's format is superior in any way, nor that everyone should start flocking to it. What I'm saying is that the arbitrarily weighted tossups aren't an issue--and nothing more.

User avatar
Jeffrey Hill
Posts: 6653
Joined: Fri Apr 30, 2004 12:00 am
Location: In between the bright lights and the far unlit unknown (aka Johnson County, KS)
Contact:

MSHSAA/tossup-bonus format discussion

Post by Jeffrey Hill »

Charbroil wrote:I would actually argue that the MSHSAA format is at its best in situations like these because it keeps close games from hinging on one or two tossups. At the very least, I wouldn't support going to a 20/20 format with nonrebounding bonuses, because in those kinds of situations, a weaker team that snipes off one or two key tossups can win a game.
It is true that MSHSAA games are less likely to have an upset based on only one or two questions; admittedly this is a slight disadvantage of the 20/20 format. However, this is more likely a function of the length of the game and not the presence of tossup-only quarters (as you said, more samples leading to less variation). In the 20/20 format, two questions make much more of a difference (80 of 800 points) than it would in MSHSAA (60 of 900 points if in the 2nd/4th quarter). If you lose momentum in a 20/20 game, you don't have nearly as much time to recover as you do in a MSHSAA game.

On the other hand, the shorter format allows for round robin formats that give approximately equal weight to every game. Because MSHSAA games are so long, round robins for fields of larger than even 6 teams are pretty much impossible, and thus they must resort to formats utilizing incomplete preliminaries and single-elimination brackets that put way too much emphasis on certain games. For instance, Fort Zumwalt West went 1-3 in the morning because they played 4 of the top 5 playoff seeds, while Francis Howell Central got to play half of its morning games against the two teams that didn't make the playoffs. While this example doesn't quite show it, in a format with fewer games, a decent team can be eliminated just because it has to play a tougher set of teams than another. It's not the length of MSHSAA games that is the problem; its the resulting suboptimal scheduling that long games require due to the infeasibility of round robins.

Additionally, changing to a shorter game length really only changes the outcome in games between teams of similar strength. A team that is definitively better than another will win a 20/20 game just as easily as a MSHSAA-style game. Why spend an hour on games like this when the same outcome can result from a half-hour game? Shorten the game and let the teams move on to different opponents, and structure the schedule so that teams that lose a close game can possibly have a chance to make up for it in a later match. While the game length makes it less probable, when a MSHSAA game hinges on one question, it can have a much more dramatic effect on the outcome of the tournament (a team gets a significantly lower seed just because of one loss, or a team gets eliminated from the tournament because of one question).

So basically, to summarize, the benefit of the shorter 20/20 format is that its ability to produce more games allows each game to have roughly equal weight and teams to have equal schedules, whereas the longer MSHSAA game, while resulting in fewer random upsets, essentially gives tournaments no choice but to give teams inconsistent opponent strength and allows the outcome of a single-elimination game to basically make all previous games for a team insignificant.
Of course, I do agree that using good questions on MSHSAA's format requires competent moderators--there's no question of that. For that format to be legitimate, MSHSAA has to start finding/training better moderators.
Definitely, though this is really the case with any format. Slow moderators in the 20/20 format that take 35-40 minutes will significantly delay a tournament.
Well that's the point of a random distribution! :) (Actually, I think MSHSAA has restrictions on how much of each topic falls into each quarter)
Yes, MSHSAA does specify question distribution by quarter.

For this tournament, we evenly distributed categories by quarter:
1st/3rd: 3 Lit, 3 Social Studies, 3 Science, 3 Math, 1 Fine Arts, 1 General, 1 Pop Culture
2nd/4th: 2 Lit, 2 Social Studies, 2 Science, 2 Math, 1 Fine Arts, 1 General/Pop Culture
(General includes Religion, Mythology, Philosophy, Social Sciences, interdisciplinary common-link tossups, etc.)

It is fairly obvious that random distribution of categories throughout the game as opposed to a quarter-by-quarter basis like MSHSAA would be a huge problem, as it could easily result in some categories being overrepresented in the more valuable tossup/bonus quarters. However, this is actually one area where MSHSAA's mandated category distribution for districts/state has an advantage over NAQT-produced rounds (note that I am NOT referring to the generic 20/20 format). By mandating a specific number of questions in each category in each quarter, each round is much more categorically consistent, whereas NAQT has so much leeway in its distribution that it's easily possible for one round to have a category significantly overrepresented - in actuality, this is probably just as much of a cause for random upsets as the shorter game length is. Another way that NAQT's packets have serious category-skewing issues stem from the fact that they are written to be timed (with 24 tossups), so it's possible that 2 questions in a given category fall in the unread questions 21-24. ACF (and presumably HSAPQ) have a much more consistent distribution so that no category is overrepresented.

Charbroil
Posts: 1039
Joined: Fri Nov 16, 2007 12:00 am

MSHSAA/tossup-bonus format discussion

Post by Charbroil »

About what Jeffrey said, I'll moderate my (limited) support for the MSHSAA format and say that outside of the issue of closely matched teams, 20/20 with rebounding bonuses does have significant advantages.

One key contrast, in my mind, is the fact that whereas MSHSAA's format is good for finding the one top team in a tournament (since it prevents upsets from beating that team), 20/20 is better for finding a good overall ranking of all of the teams (because of greater sampling, as Jeffrey described it).

I would still argue that 20/20 without rebounding bonuses produces an excessive number of upsets, though. However, that's just based on my personal opinion, and since this seems to be an issue which is fairly rare outside of Missouri (according to Charlie, anyway), I wouldn't say that it's a particularly good argument to oppose 20/20 (just like the arbitrarily weighted tossups, as opposed to other issues, isn't a good way of opposing MSHSAA).

Thanks for the discussion, by the way--it's helped refine my views quite a bit.


<div class="editby">Edited by <a href='http://s4.zetaboards.com/Academic_Compe ... arbroil</a>, Mar 9 2009, 10:24:53 PM.</div>

User avatar
Jeffrey Hill
Posts: 6653
Joined: Fri Apr 30, 2004 12:00 am
Location: In between the bright lights and the far unlit unknown (aka Johnson County, KS)
Contact:

MSHSAA/tossup-bonus format discussion

Post by Jeffrey Hill »

Charbroil wrote:I would still argue that 20/20 without rebounding bonuses produces an excessive number of upsets, though.
I agree that non-rebounding bonuses do at least slightly increase the number of upsets.

As a side note, I am fairly indifferent whether or not bonuses rebound and have not yet given too much thought to comparing the two approaches.

User avatar
Charlie Dees
Posts: 4134
Joined: Wed Jul 26, 2006 12:00 am
Location: Columbia, MO

MSHSAA/tossup-bonus format discussion

Post by Charlie Dees »

Yeah, honestly, I think Matt Weiner was right when he called it the most irrelevant aspect of the format wars whether bonuses rebound or not.

User avatar
DeckardCain
Posts: 4472
Joined: Mon Feb 09, 2004 12:00 am
Location: Viburnum, MO
Contact:

MSHSAA/tossup-bonus format discussion

Post by DeckardCain »

It's been my experience that upsets are not substantially less likely in MSHSAA-length games than in 20 tossup games*, and any increase in upset likelihood is more than made up for by the fact that the shorter length of 20 tossup games means more games can be played and a fluky result carries less weight. Your mileage may vary, I suppose.

I've actually been thinking about doing some statistical analysis on this subject for a while now, but as I loathe statistics don't expect to see this actually come to fruition anytime soon. :P

* Assuming, of course, that we're talking about MSHSAA-length games with good, pyramidal questions. I think most will agree that upsets are in fact much more likely in games with one-line tossups, regardless of game length.

nkc derek
Posts: 12
Joined: Sun Nov 02, 2008 12:00 am

MSHSAA/tossup-bonus format discussion

Post by nkc derek »

U. Lou Sthagaim wrote:It is true that MSHSAA games are less likely to have an upset based on only one or two questions; admittedly this is a slight disadvantage of the 20/20 format. However, this is more likely a function of the length of the game and not the presence of tossup-only quarters (as you said, more samples leading to less variation). In the 20/20 format, two questions make much more of a difference (80 of 800 points) than it would in MSHSAA (60 of 900 points if in the 2nd/4th quarter). If you lose momentum in a 20/20 game, you don't have nearly as much time to recover as you do in a MSHSAA game.
Although what you said about two fluke questions is true in terms of the number of points and the impact it has on the game, I think it is worth noting that there is a greater chance for there to be two fluke toss-ups in a MSHSAA game (out of 50 toss-ups).

Or if you want to focus on just the 2nd and 4th quarter (in which there are 20 toss-ups total, there would be a potential 60 point swing out of a 600 points total for those quarters.

So really, it's the same.

And while topics are obligatorily distributed evenly between the quarters, the ease of toss-ups couldn't possibly be.
So there is always the issue of whether there are more buzzer races in the 1st/3rd v. 2nd/4th.

CherryCokeStain
Posts: 79
Joined: Mon Nov 12, 2007 12:00 am

MSHSAA/tossup-bonus format discussion

Post by CherryCokeStain »

Charbroil wrote:One key contrast, in my mind, is the fact that whereas MSHSAA's format is good for finding the one top team in a tournament (since it prevents upsets from beating that team), 20/20 is better for finding a good overall ranking of all of the teams (because of greater sampling, as Jeffrey described it).
I disagree. MSHSAA's format does not guarantee that you will find the best team in the tournament. If anything, the lack of bonuses, and the length of games makes it less likely to determine the better team, because there is too much time for the game to swing, whereas a 20/20 format with rebounds allows for teams to earn the tossups, and be awarded with bonuses, and if they are unable to answer them, then the other team is allowed to demonstrate their knowledge, and earn the points. Without rebounding bonuses attached to every tossup, you cannot determine which team truly knows more. and of course ppg is always the best determinate of which team knows more. There can be and are always fluke games, especially when there is single elimination, which is a horrendous method of having a playoff. Round robins are always superior, and round robins require shorter questions.

FishyFreshman
Posts: 501
Joined: Sun Apr 15, 2007 12:00 am

MSHSAA/tossup-bonus format discussion

Post by FishyFreshman »

Shorter formats, not really shorter questions.

User avatar
Jeffrey Hill
Posts: 6653
Joined: Fri Apr 30, 2004 12:00 am
Location: In between the bright lights and the far unlit unknown (aka Johnson County, KS)
Contact:

MSHSAA/tossup-bonus format discussion

Post by Jeffrey Hill »

nkc derek wrote:And while topics are obligatorily distributed evenly between the quarters, the ease of toss-ups couldn't possibly be.
So there is always the issue of whether there are more buzzer races in the 1st/3rd v. 2nd/4th.
Yes, the value of tossups between quarters is an issue. Why should a tossup on the Battle of Antietam in the 1st quarter only be worth 10 points, while a tossup on Louis XVI in the 2nd quarter be worth those 10 points AND first chance at a 20 point bonus? While this presumably balances throughout the tournament, it does cause a subdistribution variance between games that could cause problems. The all tossup/bonus format ensures that every tossup is equally valuable.
Charbroil wrote:One key contrast, in my mind, is the fact that whereas MSHSAA's format is good for finding the one top team in a tournament (since it prevents upsets from beating that team), 20/20 is better for finding a good overall ranking of all of the teams (because of greater sampling, as Jeffrey described it).
I disagree with this statement as well; nothing can really prevent upsets from beating the best team. While the probability of an upset in an individual game might be marginally less due to the larger number of questions, the reliance on single-elimination still puts way too much weight on specific games. For instance, the scheduling advantages of a round-robin could allow a team that loses in a slightly-more-likely upset to make up for it in a later game, while a team that loses in an upset in a single elimination bracket is always out of luck.
CherryCokeStain wrote:If anything, the lack of bonuses, and the length of games makes it less likely to determine the better team, because there is too much time for the game to swing, whereas a 20/20 format with rebounds allows for teams to earn the tossups, and be awarded with bonuses, and if they are unable to answer them, then the other team is allowed to demonstrate their knowledge, and earn the points.
While it is true that longer games gives a team more time to come back and cause an upset, the effects of the longer game go both ways - what if the better team has a rocky first few questions? The longer game equally gives them a better opportunity to come back and prove that they are the better team.

I would argue that a longer game (in general, not specifically MSHSAA-format) would at least marginally increase the probability of an accurate result, because of the larger number of questions and presumably more consistent subcategory distribution. For instance, in a 20 tossup game with a reasonable distribution, say there are only 4 Science questions, so if you have Bio, Chem, Physics, and Miscellaneous, the category in which that Miscellaneous question falls has a (marginally) greater chance of impacting the game (say, a comp sci question as 5% of one game, then 0% in the next) than it would in a longer game, where there are, say, 2 Miscellaneous Science questions (so you could have a more consistent 2.5% comp sci, 2.5% earth science in each game). So basically, the combination of several minor subcategory fluctuations in a shorter game have a slightly higher chance of a fluke alignment in a team's favor, fluctuations that are not as pronounced in a longer game.

Of course, any statistical improvements that longer games provide are marginal at best, and are greatly outweighed by the advantages round-robin scheduling provides

Awehrman
Posts: 76
Joined: Wed Nov 01, 2006 12:00 am

MSHSAA/tossup-bonus format discussion

Post by Awehrman »

I disagree. MSHSAA's format does not guarantee that you will find the best team in the tournament. If anything, the lack of bonuses, and the length of games makes it less likely to determine the better team, because there is too much time for the game to swing, whereas a 20/20 format with rebounds allows for teams to earn the tossups, and be awarded with bonuses, and if they are unable to answer them, then the other team is allowed to demonstrate their knowledge, and earn the points. Without rebounding bonuses attached to every tossup, you cannot determine which team truly knows more. and of course ppg is always the best determinate of which team knows more. There can be and are always fluke games, especially when there is single elimination, which is a horrendous method of having a playoff. Round robins are always superior, and round robins require shorter questions.
I wonder if you'll show us a year that the Missouri format obviously wronged the best team in the state? Now, nearly all of us agree that the structure of disctricts/state needs to go so that the state tournament features all of the best teams, but other things aside I do think that the question format has, more or less, given teams a fair shot. Second, I'd say that the Missouri format doesn't have a lack of bonuses, but rather an overabundance of tossups. You've got what amounts to a 20/20 round with 30 extra tossups. Now if the questions aren't distributed properly there's a problem (if all the math and science were in rounds 2 and 4, for instance), but with good questions and a good distribution, there's nothing that fundamentally helps bad teams at the expense of good ones. There might be something to the notion that more shorter games produces better overall results than fewer longer games over the course of a tournament, though. Would anyone be in favor of the Kansas format which has just 16 tossup questions per round? You could play a double or triple round-robin on those. Or the Nebraska format which has 15/15 rounds with 1-part bonuses? We used to play tons of rounds in those tournaments (do any Missouri teams ever play in Kansas or Nebraska any more?), and NKC won many of them, but we didn't always go undefeated, even though we were usually the best team, because individual rounds varied so much.

One could argue that single elimination actually works better in the Missouri format than in other formats. With good questions, the longer rounds would make it more likely that the more knowledgeable team would win. Short rounds with single elimination can more often lead to upsets and unwarranted eliminations. This is why tournaments using the 20/20 format should strongly consider bracketed playoffs instead of single elimination.

User avatar
Charlie Dees
Posts: 4134
Joined: Wed Jul 26, 2006 12:00 am
Location: Columbia, MO

MSHSAA/tossup-bonus format discussion

Post by Charlie Dees »

I wonder if you'll show us a year that the Missouri format obviously wronged the best team in the state?
It's impossible to quantify this because of the fact that there has not really been a year where the questions at MSHSAA state were what would be called good quizbowl. Thus if there ever was a year where the best team didn't win state, it would be impossible to figure out if the format caused this when the fact is that the awful questions also played a role. This weekend appears to have had a field with all the state contenders playing (other than the mysterious Kirksville team and maybe Hickman-Columbia), along with for the most part good questions (although there would be an arguable aspect to the 20% of the game that's math), and there appears to be a very good argument on NKC's side that the format did cause them to lost the tournament, so this is probably the only way we will ever be able to assess this data (although even that is problematic considering NKC has lost once to FHC before by 5 on an NAQT packet, and the two teams have roughly comparable statistics other than the unquantifiable bonus conversion).
Would anyone be in favor of the Kansas format which has just 16 tossup questions per round? You could play a double or triple round-robin on those. Or the Nebraska format which has 15/15 rounds with 1-part bonuses? We used to play tons of rounds in those tournaments (do any Missouri teams ever play in Kansas or Nebraska any more?)
Nobody really want to spend money traveling to these tournaments that, to the team that's already motivated enough to travel in the first place, would undoubtedly be a waste of money since again the questions have to be abominable. The only reason NKC would travel all last year was to be able to play sets and fields that would give us a better chance of preparing for nationals. I personally would never advocate the above formats other than for novelty events like Bruce Arthur's history tournaments, because it seems to me with so few opportunities to score points, the likelyhood of upsets happening increases inordinately compared to just playing a 20/20 game of well written pyramidal questions combined with bonuses that adhere well to the easy-medium-hard structure to reward teams with the most depth the best.

Awehrman
Posts: 76
Joined: Wed Nov 01, 2006 12:00 am

MSHSAA/tossup-bonus format discussion

Post by Awehrman »

Right. I just made that argument to show that things like upsets are hard to quantify and that good teams tend to win even on bad questions. Imagine if Questions Galore wrote 20/20 rounds. There would certainly be more outrage and upsets than what you're seeing with 50/20. I think question quality takes much more precedence over the format of games or the schedule of tournaments.

The questions at those Kansas and Nebraska tournaments were actually not that bad. There's a tendency to think that all questions older than a couple years or from non-quizbowl centers have to be horrible. The tournaments we attended were hosted by colleges. Wichita State hosted their tournament on the Kansas format with good questions, much like Rolla does now with the Missouri format. Wichita State's tournament drew over 70 teams each year, and the Nebraska tournament (hosted by Peru State) drew over 100 (in all class divisions, I think there were more than thirty in the largest division, though, and there was a "grand championship". The Wichita State questions were analogous to early NAQT packets (although Kansas did have some funky game play elements), and Peru State's were similar to Bill Luce's (short and to the point, but no hoses). They were far better than the Questions Galore stuff that I've seen and light years ahead of Maple Woods (which never seem to change).


<div class="editby">Edited by <a href='http://s4.zetaboards.com/Academic_Compe ... wehrman</a>, Mar 10 2009, 06:12:20 PM.</div>

User avatar
Charlie Dees
Posts: 4134
Joined: Wed Jul 26, 2006 12:00 am
Location: Columbia, MO

MSHSAA/tossup-bonus format discussion

Post by Charlie Dees »

Whether they were bad by 1997 standards or not, they are certainly considered bad by 2009 standards when the major national tournaments are no longer NAC, Panasonic, or ASCN. I've seen questions from the Wichita State tournament, and they're not wholly unacceptable but they wouldn't go over well in D.C., and anything that can be described as "short and to the point" is bad quizbowl as well.

WillHack
Posts: 422
Joined: Mon Nov 10, 2008 12:00 am

MSHSAA/tossup-bonus format discussion

Post by WillHack »

Firstly, to Mr. Dees; why do you disregard the PCH team as a contender so easily? Do you realize that we have twice defeated FHC while missing our top 2 players? Not to be over-proud or anything. : P

I think if I were starting Quizbowl from scratch, with no previous precedents or expectations, I would set up a system with 30 to 40 toss-ups, all accompanied by 2-part bonuses. Any thoughts on this idea?

User avatar
Charlie Dees
Posts: 4134
Joined: Wed Jul 26, 2006 12:00 am
Location: Columbia, MO

MSHSAA/tossup-bonus format discussion

Post by Charlie Dees »

I am going to disagree with the belief that I think a 20/20 game is a worse setup than yours, Will, and here's why. It seems to me that people aren't paying enough attention to the importance of having bonuses that can sufficiently provide distinguishing point values, which is something the current 20/20 format has if written properly. If you have a 3 part bonus where most teams can score 10 points, moderately knowledgeable teams on the subject can score 20, and very well versed teams can score 30, that will go much further towards finding the right team than a format with nothing but 2-part bonuses, because 2 parts is not anywhere near as workable for distinguishing between the 4 types of teams (the fourth type being a team that knows absolutely nothing and is the small fraction of teams who should be zeroing the bonus). I think this aspect of the 20/20 format has not been explored as fully as it should in these discussions, because lots of MSHSAA bonuses suffer from the problem of not sufficiently distinguishing between teams as well (since they are only 20 points and also are often written VERY indistinguishably). However, if every question in your proposed 30 tossup format with 20 point bonuses were written very well, the format would be fine I guess since questions are the most important thing.

WillHack
Posts: 422
Joined: Mon Nov 10, 2008 12:00 am

MSHSAA/tossup-bonus format discussion

Post by WillHack »

Would this make more sense if the bonuses were related to the tossup? In reality, if I were starting Quizbowl from the beginning, I would try closely related bonuses first. In that way, you could get the easy-medium-hard pattern by including the initial toss-up.

User avatar
Jeffrey Hill
Posts: 6653
Joined: Fri Apr 30, 2004 12:00 am
Location: In between the bright lights and the far unlit unknown (aka Johnson County, KS)
Contact:

MSHSAA/tossup-bonus format discussion

Post by Jeffrey Hill »

WillHack wrote:Would this make more sense if the bonuses were related to the tossup? In reality, if I were starting Quizbowl from the beginning, I would try closely related bonuses first. In that way, you could get the easy-medium-hard pattern by including the initial toss-up.
The problem with related bonuses is anything you mention in the tossup itself cannot be used as a bonus answer, so it can become really hard to write related bonuses without degrading the quality of the pyramidal tossup.

WillHack
Posts: 422
Joined: Mon Nov 10, 2008 12:00 am

MSHSAA/tossup-bonus format discussion

Post by WillHack »

I suppose so, I've never written questions. I'd think the 2 bonuses instead of 3 would help. I'll have to try writing this style, to see if it works.

User avatar
Charlie Dees
Posts: 4134
Joined: Wed Jul 26, 2006 12:00 am
Location: Columbia, MO

MSHSAA/tossup-bonus format discussion

Post by Charlie Dees »

No, to me that seems like it's going to inordinately reward teams for being good in one subject. The current system of randomly distributed bonuses does a much better job of rewarding teams that are all-around better and forcing teams to have to be able to perform well in all subjects. If you have nothing but related tossups and bonuses, you end up with situations where teams that are bad in stuff like science or fine arts are not having to care anywhere near as much about those subjects because if they get the tossups in the subjects they do know, their chances of sweeping the bonuses go up as well, whereas a team that is all around knowledgable will get punished because they will lose their chance to hear bonuses in things they might otherwise be able to pull some points from because the other team's specialist got the tossup early. I think that is just too much deviation from rewarding overall knowledge to be comfortable with it, beyond maybe as one segment of a game (for instance, the NSC format opens the game with some questions like that, but the reality is that half of the game's points come from the phase where the bonuses are random so it still takes less of a toll on the outcome.)


<div class="editby">Edited by <a href='http://s4.zetaboards.com/Academic_Compe ... enaziCD</a>, Mar 10 2009, 07:24:03 PM.</div>

WillHack
Posts: 422
Joined: Mon Nov 10, 2008 12:00 am

MSHSAA/tossup-bonus format discussion

Post by WillHack »

Hmm, I don't know. If the distribution was fair, it would still seem to me to reward teams appropriately for knowledge in one subject area. If a team knows a subject area very well, it deserves a full reward for that . . . I have no problem there. The only big difference with current systems that I can see would be higher scores for both winning and losing teams. And it would certainly encourage depth of knowledge, I think more so than most current systems.

rjaguar3
Posts: 14
Joined: Fri Feb 13, 2009 12:00 am
Location: Nashville, TN

MSHSAA/tossup-bonus format discussion

Post by rjaguar3 »

I think 3-part bonuses are the best because they allow for a easy, medium, and hard part, when written correctly, dividing teams into 0, 10, 20, and 30 point score groups.

Unfortunately, many writers fail to heed to this intermediacy criterion, and as such, we still have bonuses like "identify the four nonreal sixth roots of 1," "name 4 of the 6 wives of Henry VIII," and "convert the following angle measures from degrees to radians."

FishyFreshman
Posts: 501
Joined: Sun Apr 15, 2007 12:00 am

MSHSAA/tossup-bonus format discussion

Post by FishyFreshman »

Charles hit the nail right on the head.

Awehrman
Posts: 76
Joined: Wed Nov 01, 2006 12:00 am

MSHSAA/tossup-bonus format discussion

Post by Awehrman »

Whether they were bad by 1997 standards or not, they are certainly considered bad by 2009 standards when the major national tournaments are no longer NAC, Panasonic, or ASCN. I've seen questions from the Wichita State tournament, and they're not wholly unacceptable but they wouldn't go over well in D.C., and anything that can be described as "short and to the point" is bad quizbowl as well.
Oh, Charlie, why must you be so dismissive? I'm not arguing that teams should play on questions from 1997. Certainly questions that old would not hold up terribly well, but on the other hand if you knew everything in those old packets you'd be a hell of a player (I'll feel like I have to parse every point: Players should, of course, study newer material first). I was making the argument that while the college style system is in the vanguard (and I'm not arguing that it shouldn't be), there is nothing inherently wrong with other ways of testing knowledge through the medium of buzzers (or without buzzers as Maple Woods does). Also, I was responding to your point that no one would drive a couple hours for such awful questions by saying that people still do attend tournaments with far worse questions and that those questions weren't all that bad in the first place.

Thousands of people who have participated in a quizbowl-like substance have enjoyed and profited from it. It has made people appreciate the collection of knowledge throughout their lifetimes, and they have made great friends and impressed people. There was no dark age. Ultimately who came in fifth or if which team had the highest bonus conversion at some tournament in 2007 dies away pretty quickly. Ultimately people need to encourage the broader participation in as well as the quality of knowledge distributed in quizbowl rather than more petty concerns about which bonus or tournament format is more in style or popular at the moment. While arguments such as these can be entertaining and occasionally enlightening, they don’t really do much good.

I'm also not convinced that there needs to be one standardized format that all quizbowl-like substances should adhere to. I like a little regional and tournament by tournament variety, so I'm encouraged by some of the other posts in this thread where people offer alternative formats.

Post Reply