Bounceback Theory, NSC and Elsewhere

Dormant threads from the high school sections are preserved here.
Locked
jonah
Auron
Posts: 2303
Joined: Thu Jul 20, 2006 5:51 pm
Location: Chicago

Bounceback Theory, NSC and Elsewhere

Post by jonah » Mon Jun 07, 2010 10:29 pm

I've not previously had an objection to bouncebacks on bonuses, though I didn't particularly feel that they should be present either. I still think that for lower- and perhaps mid-level tournaments (in terms of field competitiveness, especially among the lower end thereof), they are excellent for keeping teams who aren't getting lots of tossups engaged. But this past weekend got me thinking: why do they exist to determine a national championship, when they can and do make it advantageous to not answer questions? I personally read at least two games where this was the case (that after the losing-by-just-10-or-20-points team "negged" the last question, it would have been foolish for the other team to answer the question correctly lest the team that "negged" get enough rebounded bonus points to win), and know that others happened. While I appreciate the spirit in which bouncebacks are often used (whether or not that is the reason for using them at NSC, I have no idea), I think situations like this create a compelling case for eschewing them.

There are other reasons to not use bouncebacks, of course. They slow down rounds—by how much is uncertain and perhaps not huge, but given that Saturday finished quite, quite late, every minute saved is clearly valuable—, and they are prone to being screwed up by moderators who aren't used to them. And of course they allow obviously degenerate but still possible situations in which a team wins despite answering very few tossups, while the other team answers (almost or strictly) all the tossups and loses.

So, what's the rationale for having bouncebacks in NSC? Might that rule be reevaluated? What do people think about this issue more broadly?
Jonah Greenthal
National Academic Quiz Tournaments

Adventure Temple Trail
Auron
Posts: 2613
Joined: Tue Jul 15, 2008 9:52 pm

Re: Bounceback Theory, NSC and Elsewhere

Post by Adventure Temple Trail » Mon Jun 07, 2010 11:11 pm

I'm not entirely decided on bouncebacks. They certainly make close games a lot more exciting and dire at any time, and prioritize listening to the other team's bonuses in a completely different way. The effect of bouncebacks on game scores seems to make teams that are closer in skill have closer games, and teams that are further apart have a wider disparity scores. At PACE, our games with Southside and St. Anselm's were literally decided by whether the last bonus bounced or not, in part because bonuses were bouncing back, getting converted throughout the game and keeping things relatively even, and we couldn't have beaten Dunbar if we hadn't stolen 10 from the majority of the bonuses they had, since they converted more tossups than we did. They are certainly fair - the team that got the tossup always controls the bonus - and certainly reward knowledge by making real the threat of having points stolen from you (or giving the opportunity to steal from teams that don't know the answer).

They have another issue or two, though, besides the one Jonah presented. You really have to control your conferring in a bounceback scenario, since the other team can use anything you say against you, even if they don't have any knowledge of the topic at hand. If your team says things to the effect of "hmm, it's an Irish heroic dude, so it's either Cu Chulainn or Finn MacCool...ANSWER: Finn macCool?" and you're wrong, but they know nothing about Irish myth (or whatever), they can still jack points regularly with your educated guessing. It's also worth noting the possibility of a team which gets 10 points on a tossup, but has up to 30 points stolen from them on the bonus, giving them a net point loss of 20 for answering. (As I said in the main PACE thread, I think that 20-point powers go a long way towards counteracting this in top-flight and 2nd-flight games, making the potential net point loss less of an issue to teams that are likely to get all the easy and middle parts of bonuses.)

It'll be hard to make a more informed judgment of how much bouncebacks changed the course of the tournament until more complete stats are available. As of now, though, they seem to be a fair way to give knowledgeable teams more of a chance to show their knowledge, so I'm in no rush to remove them.
Matt J.
ex-Georgetown Day HS, ex-Yale
member emeritus, ACF

Sailing away on my copper boat

User avatar
Nine-Tenths Ideas
Auron
Posts: 1549
Joined: Sat Feb 03, 2007 10:14 pm
Location: MD

Re: Bounceback Theory, NSC and Elsewhere

Post by Nine-Tenths Ideas » Mon Jun 07, 2010 11:16 pm

I found myself a little frustrated with bouncebacks mainly because I seemed to have a lot more trouble converting bonuses than I did answering tossups. For example, in our game against DuPont Manual A, we had answered 2 more tossups, both powers, than Manual, but they stole 90 bonus points from us while we stole 0. If bonuses didn't bounce, we would have won. It's aggravating to power a tossup and find yourself actually getting farther away in points from the opposing teams as a reward-- in some ways, I think this would make not-so-great teams even unhappier. I realize this isn't the demographic NSC goes for, but it essentially tells mid-level teams who attend the tournament, "Even if you get this tossup, you're still screwed."
Isaac Hirsch
University of Maryland '14
Never Gonna Play Again

User avatar
Down and out in Quintana Roo
Auron
Posts: 2907
Joined: Wed Apr 09, 2008 7:25 am
Location: Camden, DE
Contact:

Re: Bounceback Theory, NSC and Elsewhere

Post by Down and out in Quintana Roo » Mon Jun 07, 2010 11:24 pm

Isaacbh wrote:I found myself a little frustrated with bouncebacks mainly because I seemed to have a lot more trouble converting bonuses than I did answering tossups. For example, in our game against DuPont Manual A, we had answered 2 more tossups, both powers, than Manual, but they stole 90 bonus points from us while we stole 0. If bonuses didn't bounce, we would have won. It's aggravating to power a tossup and find yourself actually getting farther away in points from the opposing teams as a reward-- in some ways, I think this would make not-so-great teams even unhappier. I realize this isn't the demographic NSC goes for, but it essentially tells mid-level teams who attend the tournament, "Even if you get this tossup, you're still screwed."
I think what Isaac is trying to say, that i was about to say until he sorta started to, was that bouncebacks made big upsets basically never happen. For example, if HSNCT (somehow) was run on bouncebacks, we never would have had a shot against Auburn because they would have stolen a lot of what we didn't get. But without them, we lost by just 5 because we picked up all five of their negs and they didn't get any of those bonuses, obviously. Also, i'm thinking Adair County probably would never have won all those games in a row with bouncebacks.

So, anyway, if you like big upsets and teams pulling off wins that "statistically" they shouldn't do because they're just a good team versus a very good team, then bouncebacks actually likely prevent those from happening more often.
Mr. Andrew Chrzanowski
Caesar Rodney High School
Camden, Delaware
CRHS '97-'01
University of Delaware '01-'05
CRHS quizbowl coach '06-'12
http://crquizbowl.edublogs.org

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

Re: Bounceback Theory, NSC and Elsewhere

Post by nobthehobbit » Mon Jun 07, 2010 11:45 pm

Apart from anything else, I don't like bouncebacks for the simple reason that I don't think there should ever be a reason not to answer a tossup. Without bouncebacks, the only reason I can think of not to answer a tossup is when ties are broken by PPB and you know that answering a tossup and failing to get a certain amount on a bonus will drop your PPB below that of a team with which you'll be tied--and how often is a team going to know that's the case?
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.

Strongside
Rikku
Posts: 475
Joined: Thu Apr 14, 2005 8:03 pm
Contact:

Re: Bounceback Theory, NSC and Elsewhere

Post by Strongside » Mon Jun 07, 2010 11:47 pm

jonah wrote: make it advantageous to not answer questions? I personally read at least two games where this was the case (that after the losing-by-just-10-or-20-points team "negged" the last question, it would have been foolish for the other team to answer the question correctly lest the team that "negged" get enough rebounded bonus points to win), and know that others happened.
Yeah. This is my main problem with bouncebacks. The primary goal of quiz bowl is to answer tossups and it is somewhat of a problem that bounceback bonuses discourage this. One could argue that it adds more strategy to the game at times, but I don't really like that.

One thing that might be good to try on bounceback bonuses is to make bouncebacks count for 5 instead of 10, or to have a rule where the team that didn't get the tossup can't score more total points on the question than the team that got the tossup. The problem with this is that it could make things more complicated, and mess up statistics.
Brendan Byrne

Drake University, 2006-2008
University of Minnesota, 2008-2010

User avatar
Haaaaaaaarry Whiiiiiiiiiite
Auron
Posts: 1124
Joined: Tue Dec 04, 2007 8:46 pm
Location: Fairfax, VA
Contact:

Re: Bounceback Theory, NSC and Elsewhere

Post by Haaaaaaaarry Whiiiiiiiiiite » Mon Jun 07, 2010 11:49 pm

The only reason to not answer a tossup when there are bouncebacks is when you are ahead by 10 and it is TU 20. If a team gets a tossup and the other team steals all 30, it's not like that bonus won't still be read later on in the round. Plus, you don't know what the bonus will be about anyway.
Harry White
TJHSST '09, Virginia Tech '13, PACE Member
Tournament Database Search
Quizbowl Schedule Generator
Will run stats for food

jonah
Auron
Posts: 2303
Joined: Thu Jul 20, 2006 5:51 pm
Location: Chicago

Re: Bounceback Theory, NSC and Elsewhere

Post by jonah » Mon Jun 07, 2010 11:50 pm

The Gambler, the Nun, and the Radio wrote:The only reason to not answer a tossup when there are bouncebacks is when you are ahead by 10 and it is TU 20. If a team gets a tossup and the other team steals all 30, it's not like that bonus won't still be read later on in the round. Plus, you don't know what the bonus will be about anyway.
Well, or you're ahead by 20—then all 30 being stolen would leave the game tied.
Jonah Greenthal
National Academic Quiz Tournaments

User avatar
pblessman
Tidus
Posts: 500
Joined: Fri Sep 12, 2003 10:38 am
Location: Culver, Indiana

Re: Bounceback Theory, NSC and Elsewhere

Post by pblessman » Mon Jun 07, 2010 11:50 pm

I have come full-circle on bouncebacks. First having coached in Illinois, I thought bouncebacks were a good thing, and at one point I actually lobbied NAQT to put them in. Now, however, I dislike them for four reasons:

1. They contradict the idea of a BONUS. When you're all excited about getting a TU and then get a "bonus" on Finnish poets that you bagel which the other team ends up thirtying... yeah... that sucks.

2. Even though it would appear to encourage listening to the bonus for the non-controlling team, it creates too many situations where teams will get points NOT because they knew it, but because they worked off the controlling team's answer. You can counteract this by writing only certain kinds of bonuses, but that seems weird as well.

3. It REALLY slows down games. This is especially true when both the teams are weaker, so then the controlling team takes a long time on each question, a lot bounce back, the other team takes a long time on each bonus... you get the idea... And in the end, they call it a "bonus" because it is an addition to the most exciting part: The toss-up. I'd rather have a game with four more TUs than bounce-back bonuses. And you can certainly get in four TUs in the time it takes to do bouncebacks in the average game.

4. It should never be to your disadvantage to answer a question correctly. This is the case in what Jonah described (up by 5 before last question, other team negs... only a fool would buzz in...). I remember losing a game in the Illinois State Series that way... and we lost by one point... on a drivers' ed question... but now I'm piling on...
Phil Blessman- Culver Academies Head Quizbowl Coach

User avatar
Cheynem
Sin
Posts: 6592
Joined: Tue May 11, 2004 11:19 am
Location: Grand Rapids, Michigan

Re: Bounceback Theory, NSC and Elsewhere

Post by Cheynem » Mon Jun 07, 2010 11:55 pm

I actually am not a big fan of bouncebacks, but in regards to the last few points:

I don't think it slows down games. Under the timing rules (at PACE), teams have a maximum of two seconds to confer on a bonus bounceback.That's not a "long time" and it will certainly not allow for four more TUs to be read.
Mike Cheyne
Formerly U of Minnesota

"You killed HSAPQ"--Matt Bollinger

User avatar
pblessman
Tidus
Posts: 500
Joined: Fri Sep 12, 2003 10:38 am
Location: Culver, Indiana

Re: Bounceback Theory, NSC and Elsewhere

Post by pblessman » Tue Jun 08, 2010 12:14 am

OK... Assuming you need 60 second per TU, four TUs is 240 seconds. Assuming all 20 TUs get answered and 1.5 bonus parts are converted, you're bouncing back 30 questions... Even if you limit conferral time to 2 seconds, you need to re-read the question (2-3 sec?), give time to actually say the answer (1-2 sec ?), indicate whether it's correct or not (1 sec?), and then transition to the next part (1 sec?). MAYBE some readers can do this consistently under eight seconds total per part, but it's not a lot less. This would also be indicated by the fact that PACE 20-question games tend to take as long as a 24-question NAQT match. That being said, I am willing to concede that four might be a bit high, but it's certainly three.
Phil Blessman- Culver Academies Head Quizbowl Coach

User avatar
Cheynem
Sin
Posts: 6592
Joined: Tue May 11, 2004 11:19 am
Location: Grand Rapids, Michigan

Re: Bounceback Theory, NSC and Elsewhere

Post by Cheynem » Tue Jun 08, 2010 12:15 am

PACE rules do not allow the re-reading of a question for a bounceback. Also, eliminating bouncebacks from PACE rules would not allow for more tossups anyway, as the format is untimed to begin with and is (as of now) 20/20.

Note, I am speaking from bouncebacks only applied to PACE rules and not in a timed format.
Mike Cheyne
Formerly U of Minnesota

"You killed HSAPQ"--Matt Bollinger

User avatar
dtaylor4
Auron
Posts: 3733
Joined: Tue Nov 16, 2004 11:43 am

Re: Bounceback Theory, NSC and Elsewhere

Post by dtaylor4 » Tue Jun 08, 2010 3:43 am

jonah wrote:There are other reasons to not use bouncebacks, of course. They slow down rounds—by how much is uncertain and perhaps not huge, but given that Saturday finished quite, quite late, every minute saved is clearly valuable—, and they are prone to being screwed up by moderators who aren't used to them. And of course they allow obviously degenerate but still possible situations in which a team wins despite answering very few tossups, while the other team answers (almost or strictly) all the tossups and loses.
I think this point is not exactly valid. Here is what stalled the tournament:

1) One room being locked, then having broken buzzer issues.
2) The bracket affected by the issue in 1) had a circle of death. One of the tiebreaker games was tied, and it took a minute for Sorice to decide how to break the tie. The decision was to use the standard full-game tiebreaker (3/3, then change in score).

David Riley
Auron
Posts: 1428
Joined: Tue Apr 29, 2003 8:27 am
Location: Morton Grove, IL

Re: Bounceback Theory, NSC and Elsewhere

Post by David Riley » Tue Jun 08, 2010 9:53 am

Tratidtionally, the only thing in quiz bowl that I've detested more than bouncebacks (see Blessman points #1 and #3) is the practice known as subapalooza (which is rampant among Illinois middle school teams and some high school teams).

Yet, I have never minded bouncebacks at PACE, and I think it's because even with varying abilities among teams, there is a more competitive spirit, and that gap is not nearly as wide as in your standard IHSA formatted Illinois tournament. Would I feel similiarly if NAQT started using bouncebacks? I'm not sure....the larger numer of teams at HSNCT would certainly slow down the tournament.,
David Riley
Coach Emeritus, Loyola Academy, Wilmette, Illinois, 1993-2010
Steering Committee, IHSSBCA, 1996 -
Member, PACE, 2012 -

"This is 1183, of course we're barbarians" -- Eleanor of Aquitaine in "The Lion in Winter"

User avatar
Cody
2008-09 Male Athlete of the Year
Posts: 2304
Joined: Sun Nov 15, 2009 12:57 am
Location: Richmond

Re: Bounceback Theory, NSC and Elsewhere

Post by Cody » Tue Jun 08, 2010 10:24 am

From my experience staffing the NSC this year, I do not feel that bouncebacks slow games down. From a hypothetical standpoint, the maximum time bouncebacks can add to a game is something like 3 minutes (<1s for redirect, ~2s for answer), which only happens if one team gets every tossup and knows none of the 60 bonus parts, and I think everyone knows how improbable that is. I think you will find in reality that it slows down games negligibly, especially with quality readers.
Cody Voight, VCU ‘14. I wrote lots of science and am an electrical engineer.
VCU Tournament Director ‘13-‘17. HSAPQ President ‘15-16.
Hero of Socialist Quizbowl Labor (NSC ‘14). “esteemed colleague” of Snap Wexley, ca. 2016. Stats Hero (Nats ‘16).
Quizbowl at VCU

User avatar
TheKingInYellow
Rikku
Posts: 310
Joined: Sun Sep 14, 2008 5:13 pm

Re: Bounceback Theory, NSC and Elsewhere

Post by TheKingInYellow » Tue Jun 08, 2010 4:54 pm

Personally, I'm a big fan of bouncebacks-- I can't see them slowing down the game more than a few minutes, as Cody said. As far as Mr Blessman's first point, the point of quizbowl is to reward knowledge, in my mind anyway, and bouncebacks provide more of an opportunity to do so.

As far as his second point; answer manipulation is an issue, but from my experience a comparatively very minor one, which I don't think should determine the fate of bouncebacks
Graham Moyer
State College 2011
Harvard 2015

User avatar
Kwang the Ninja
Rikku
Posts: 480
Joined: Thu Apr 23, 2009 3:25 pm

Re: Bounceback Theory, NSC and Elsewhere

Post by Kwang the Ninja » Tue Jun 08, 2010 6:24 pm

Yeah, pretty much all of the points above against bouncebacks (e.g. they can make you lose the game even if you answer a tossup, etc.) can be avoided by knowing things, which I thought was a major strategy in quizbowl anyway.
Dallin Kelson
Chipola '11, UF '13

User avatar
Captain Sinico
Auron
Posts: 2840
Joined: Sun Sep 21, 2003 1:46 pm
Location: Champaign, Illinois

Re: Bounceback Theory, NSC and Elsewhere

Post by Captain Sinico » Tue Jun 08, 2010 6:31 pm

The thesis that the point of the game is to get the most tossups is not one that I can agree with. In fact, even if everyone did agree with that, eliminating bouncebacks would be pretty low on the priority list for rules we should change to facilitate it. It seems to me that the first rule we'd throw out is the rule that says players can only give one answer to a question and the second we'd can is the prohibition on conferring on tossups. The point of that is that any reasonable person would agree that the goal of the game is to win given the rules of the game, and those rules dictate the optimal strategy, not vise versa.

That means that bounceback formats are in principle no worse than non-bounceback formats, unless one optimal strategy is in principle better than another, which I don't think anyone has established. In fact, even if the proposition that getting more tossups should always be best appeals to you, the fact is that the presence of bouncebacks does basically nothing one way or another: with no bouncebacks, the optimal strategy is always to get as many tossups as possible, whereas with bouncebacks, the optimal strategy is exactly the same except in the case of a question negged by a behind team that, if unanswered by the ahead team, would establish an insurmountable lead but that, if answered by the ahead team, would not establish such a lead if the behind team steals all the bonus points. In practice, that occurs for much less than 1% of questions (the theoretical limit on such questions in a round is 4 as near as my back-of-envelope calculation can tell; at any rate, it is much less than 10.) Also, on all such questions, the game is effectively decided already for a team pursuing the optimal strategy - it is mathematically decided in a non-bounceback format and decided modulo strategy in a bounceback format.

So, when we look at it in that light, the original argument has become: "Bouncebacks shift the optimal strategy away from what someone arbitrarily says it ought to be, on less than 1% of questions, on all of which the game is already decided anyway." I trust I need say no more about that.


The argument that bouncebacks add a ton to game time doesn't seem to hold much water. Even if we grant that a 2-second bounceback might take up to twice that long to implement, bouncebacks can then add at most 4 seconds*20*3 = 4 minutes to a PACE-format game.


It is definitely the case that bouncebacks can be and are messed up by even experienced moderators and the current rules to rectify such mistakes are imperfect. I don't think that makes a very strong argument against them, but it is something.

A stronger argument against bouncebacks is along the lines of Brendan's: that it's not really fair to give a team the same points for a bounceback that the first-crack team could have earned, given that the bounceback team has had longer to consider the answer and often has more information that the first-crack team in the form of an incorrect answer on the question, which can often be very, very valuable.


On the other hand, I posit that bounceback formats are potentially able to do a better job of rewarding knowledge differences, all else equal, since teams are always given a chance to show that they know something that another team doesn't on a bonus and be rewarded for doing so. That is a strong argument for bouncebacks given that the broader point of the game is to foster learning. Relatedly, bouncebacks allow teams with superior knowledge to overcome runs of tossups that happen to miss their areas and keep teams who aren't getting many tossups engaged better. A still weaker argument for bouncebacks is that (my read of popular opinion at least is that) they are much more liked by the people who like them than they are disliked by the people who dislike them and that most people don't care all that much, and probably shouldn't.


In summary, there are some drawbacks to bouncebacks, but I don't seem them as insurmountable, amounting as they do to entailing further moderator training and a careful evaluation of scoring practices. Conversely, there are definitely some good arguments in their favor.

M
Mike Sorice
Coach, Centennial High School of Champaign, IL (2014-) & Team Illinois (2016-2018)
Alumnus, Illinois ABT (2000-2002; 2003-2009) & Fenwick Scholastic Bowl (1999-2000)
ACF
IHSSBCA
PACE

User avatar
Deviant Insider
Auron
Posts: 4613
Joined: Sun Jun 13, 2004 6:08 am
Location: Chicagoland
Contact:

Re: Bounceback Theory, NSC and Elsewhere

Post by Deviant Insider » Tue Jun 08, 2010 6:46 pm

Technically, there are more times than the ones mentioned above when you don't want to buzz with bouncebacks, such as when you are up by 60 points on the 19th question and your opponent negs and when you are up by 110 points on the 18th question and your opponent negs. However, I don't see this as a major reason to do away with bouncebacks--in general, it's pretty unusual that the strategy comes into play since the score has to be right and the losing team has to neg, and teams that don't follow the strategy almost always win anyways.

I think the main argument comes down to fairness vs timing. I think having bouncebacks is more fair than not having them, since they reward teams that know something their opponent does not know. Even this I don't see as a huge advantage of having them, since it's pretty rare that the existence of bouncebacks affects who wins a match, but I think it's a legitimate reason. As to timing, my experience is that you either save thirty minutes (about three minutes per round) over the course of a tournament by eliminating them or get to add an extra round. It is for this reason that I don't want bouncebacks at the New Trier Varsity and would like to see the IHSA get rid of them.

The time PACE NSC would save might be less than at other tournaments because the moderator quality is so good. Tournaments move at the rate of their slowest moderator, and the slowest moderator at NSC probably is significantly faster than the slowest moderator at any large local competition. While a good moderator can recognize a team, give them two seconds to answer, and react to that answer in four seconds, it takes longer for mediocre moderators, largely because they take more than two seconds to call time.
David Reinstein
PACE VP of Outreach, Head Writer and Editor for Scobol Solo and Masonics (Illinois), TD for New Trier Scobol Solo and New Trier Varsity, Writer for NAQT (2011-2017), IHSSBCA Board Member, IHSSBCA Chair (2004-2014), PACE Member, PACE President (2016-2018), New Trier Coach (1994-2011)

User avatar
Captain Sinico
Auron
Posts: 2840
Joined: Sun Sep 21, 2003 1:46 pm
Location: Champaign, Illinois

Re: Bounceback Theory, NSC and Elsewhere

Post by Captain Sinico » Tue Jun 08, 2010 7:01 pm

Westwon wrote:Technically, there are more times than the ones mentioned above when you don't want to buzz with bouncebacks, such as when you are up by 60 points on the 19th question and your opponent negs and when you are up by 110 points on the 18th question and your opponent negs.
I think what I said accounts for those. I was trying to say it only happens on a question where the other team's negged and not answering gives a team an insurmountable lead.
In any event, the point is this: the strategy-shifting event is rare to begin with. If it does happen, if the team affected enacts the optimal strategy once, it then has an insurmountable lead and should go for as many tossups as possible, so the event should really only occur once per round at absolute most.

M
Mike Sorice
Coach, Centennial High School of Champaign, IL (2014-) & Team Illinois (2016-2018)
Alumnus, Illinois ABT (2000-2002; 2003-2009) & Fenwick Scholastic Bowl (1999-2000)
ACF
IHSSBCA
PACE

jonah
Auron
Posts: 2303
Joined: Thu Jul 20, 2006 5:51 pm
Location: Chicago

Re: Bounceback Theory, NSC and Elsewhere

Post by jonah » Tue Jun 08, 2010 7:22 pm

I think Mike's post conflates two different sorts of optimality: the optimal strategy for playing/winning a match, which is exactly what he says it is (to win given the rules); and the optimal way to design the game's rules. The latter is certainly a matter of opinion, and as we continually demonstrate on this board, opinions vary widely. My opinion as it relates to this discussion is that we should not design the rules of a game to be such that, for the in-game strategy, it is anything but always favorable to answer questions correctly.

I don't think that the point of quizbowl is to get the most tossups, since bonuses by and large test deeper knowledge than tossups, and deep knowledge ought to be rewarded—which is precisely the reason we have bonuses that are worth more (in total) than tossups, as well as sometimes powers. I even agree that bouncebacks make sense from the point of view that teams should be rewarded for showing knowledge that their opponent has failed to show. However, the following sum up why I oppose bouncebacks in high-level competition such as the NSC:
*They make it possible for a team to be effectively punished for answering a tossup. Yes, this is unusual, but it certainly happens, and I think it's a big problem when it does.
*Teams can gain, as has been pointed out by several people, a substantial advantage from knowing one thing that the answer isn't, as well as potentially from hearing the opponents' conferral. The former problem can be somewhat alleviated by not writing bonuses with parts that are effectively binary, but I don't think such bonuses are necessarily problematic (that is, if they require some knowledge to become binary; I'm not advocating parts that are literally isomorphic to true/false, of course). I don't think that reducing the value of stolen bonus parts helps this situation all that much, and assuming we would make them 5 points, leaves the previous problem intact (though reducing its incidence) for formats in which it's possible to have a 5-point difference (formats with negs and/or formats with 15-point powers).
Jonah Greenthal
National Academic Quiz Tournaments

User avatar
jdeliverer
Rikku
Posts: 296
Joined: Tue Sep 29, 2009 10:26 pm
Location: Providence

Re: Bounceback Theory, NSC and Elsewhere

Post by jdeliverer » Tue Jun 08, 2010 9:11 pm

I do support bouncebacks, since I think they make the narrow gap between teams at the top level a little bit more likely to show. When the top 10 teams are all playing games that end up within 100 points, that extra bit of differentiation can help the more knowledgeable ("deserving") team win.

Also, I think bouncebacks differentiate between a team being screwed by an overly difficult bonus and a team simply missing important knowledge (one thing I don't like is that a team can 0 a trash bonus that the other team can 30 for whatever reason, but that's a somewhat separate issue).
Captain Sinico wrote: In any event, the point is this: the strategy-shifting event is rare to begin with. If it does happen, if the team affected enacts the optimal strategy once, it then has an insurmountable lead and should go for as many tossups as possible, so the event should really only occur once per round at absolute most.

M
Assuming non-optimal strategy, a team can have this situation come up on tossups 11, 13, 15, 17, and 19 (or 12 14 16 18 20). Do you really want 25% of your tossups putting teams in that difficult position?
Robert Volgman
Brown '14
Latin School of Chicago '10

User avatar
TheKingInYellow
Rikku
Posts: 310
Joined: Sun Sep 14, 2008 5:13 pm

Re: Bounceback Theory, NSC and Elsewhere

Post by TheKingInYellow » Tue Jun 08, 2010 9:48 pm

I'd like to point out that bounceback bonuses don't punish a team for answering a tossup, they punish a team for not knowing things
Graham Moyer
State College 2011
Harvard 2015

User avatar
Deviant Insider
Auron
Posts: 4613
Joined: Sun Jun 13, 2004 6:08 am
Location: Chicagoland
Contact:

Re: Bounceback Theory, NSC and Elsewhere

Post by Deviant Insider » Tue Jun 08, 2010 9:49 pm

My previous post was written before I read Mike's, so sorry for the little bit of confusion. Robert is incorrect--the strategy only makes sense when you want to end the possibility of losing a match, so it only can be used once.
David Reinstein
PACE VP of Outreach, Head Writer and Editor for Scobol Solo and Masonics (Illinois), TD for New Trier Scobol Solo and New Trier Varsity, Writer for NAQT (2011-2017), IHSSBCA Board Member, IHSSBCA Chair (2004-2014), PACE Member, PACE President (2016-2018), New Trier Coach (1994-2011)

User avatar
AKKOLADE
Sin
Posts: 15269
Joined: Thu Apr 24, 2003 8:08 am

Re: Bounceback Theory, NSC and Elsewhere

Post by AKKOLADE » Tue Jun 08, 2010 9:52 pm

jdeliverer wrote:Assuming non-optimal strategy, a team can have this situation come up on tossups 11, 13, 15, 17, and 19 (or 12 14 16 18 20). Do you really want 25% of your tossups putting teams in that difficult position?
Teams are seriously going to be saying on tossup 11, "Well, I know this and can buzz in, but if I do then the other team could gain as many as 20 points on us and that will allow them to win the game"? Really? Under any circumstances? Seriously?
Fred Morlan
PACE President, 2018-19
International Quiz Bowl Tournaments, co-owner
University of Kentucky CoP, 2017
hsqbrank manager, NAQT writer (former subject editor), former hsqb Administrator/Chief Administrator, 2012 NASAT TD

User avatar
AKKOLADE
Sin
Posts: 15269
Joined: Thu Apr 24, 2003 8:08 am

Re: Bounceback Theory, NSC and Elsewhere

Post by AKKOLADE » Tue Jun 08, 2010 9:53 pm

TheKingInYellow wrote:I'd like to point out that bounceback bonuses don't punish a team for answering a tossup, they punish a team for not knowing things
This also is a pretty good explanation for my views.
Fred Morlan
PACE President, 2018-19
International Quiz Bowl Tournaments, co-owner
University of Kentucky CoP, 2017
hsqbrank manager, NAQT writer (former subject editor), former hsqb Administrator/Chief Administrator, 2012 NASAT TD

User avatar
jdeliverer
Rikku
Posts: 296
Joined: Tue Sep 29, 2009 10:26 pm
Location: Providence

Re: Bounceback Theory, NSC and Elsewhere

Post by jdeliverer » Tue Jun 08, 2010 10:20 pm

jdeliverer wrote: Assuming non-optimal strategy, a team can have this situation come up on tossups 11, 13, 15, 17, and 19 (or 12 14 16 18 20). Do you really want 25% of your tossups putting teams in that difficult position?
This was meant purely as a theoretical exercise in response to Mike's post. Clearly, even if teams did not play optimally, the situation would almost never arise more than once in one game. The last sentence was just sarcasm -- as I said, I support bouncebacks. The difference they make in terms of teams not answering questions to clinch a win is negligible compared to the difference they make in allowing teams to demonstrate important knowledge.
Robert Volgman
Brown '14
Latin School of Chicago '10

User avatar
Unicolored Jay
Forums Staff: Administrator
Posts: 781
Joined: Mon Jun 15, 2009 3:28 pm
Location: Columbus, Ohio

Re: Bounceback Theory, NSC and Elsewhere

Post by Unicolored Jay » Wed Jun 09, 2010 12:30 am

Food for thought: at the NSC, I wasn't exactly the fastest person on the buzzer, and having bouncebacks helped me earn anywhere from 30 to 70-80 points in matches that would have been much less close otherwise. They really helped in matches like against St. Ignatius, where I know I stole 20 on bonuses several times.
Jasper Lee
University of Tennessee
The Ohio State University '14
Solon High School '10

alexdz
Rikku
Posts: 387
Joined: Wed Sep 03, 2008 7:29 pm
Location: Philadelphia, PA

Re: Bounceback Theory, NSC and Elsewhere

Post by alexdz » Wed Jun 09, 2010 1:24 am

The Granny wrote:
TheKingInYellow wrote:I'd like to point out that bounceback bonuses don't punish a team for answering a tossup, they punish a team for not knowing things
This also is a pretty good explanation for my views.
Ditto here.
Captain Sinico wrote:It is definitely the case that bouncebacks can be and are messed up by even experienced moderators and the current rules to rectify such mistakes are imperfect. I don't think that makes a very strong argument against them, but it is something.
I actually have the opposite problem when moderating. Having played Missouri state format quizbowl all throughout high school (I didn't know any better, I promise!), I am so used to bouncebacks that my impulse is sometimes to give the other team a chance. I'm rarely in that mode anymore but it has happened.

My opinion on this whole issue isn't really informed by any kind of mathematical probabilities nor philosophical point-making, but just as a player. I prefer bouncebacks (/rebounds) myself, simply because they give me a chance to score more points in a game by answering questions that I know the answer to. As most of you probably know, it's certainly no fun to sit and watch the other team bagel a bonus that you could have 30d. I'd much rather have a chance to use my knowledge in the game.

Is it fair that I get more time or more information to determine my answer? Maybe, maybe not, but it's equally as unfair as having more time or information on a rebounding tossup that the first team negged. To me, that part of the argument against bouncebacks would entail a whole bunch of wacky consequences to tossup answering as well.
Alex Dzurick
====
Owner/Editor, SAGES Quizbowl Questions
Coach, Harcum College (PA)
====
Former midwesterner (South Callaway - Mizzou - UIUC) coping with life on the east coast.

User avatar
Captain Sinico
Auron
Posts: 2840
Joined: Sun Sep 21, 2003 1:46 pm
Location: Champaign, Illinois

Re: Bounceback Theory, NSC and Elsewhere

Post by Captain Sinico » Wed Jun 09, 2010 6:00 pm

jdeliverer wrote:
Captain Sinico wrote: In any event, the point is this: the strategy-shifting event is rare to begin with. If it does happen, if the team affected enacts the optimal strategy once, it then has an insurmountable lead and should go for as many tossups as possible, so the event should really only occur once per round at absolute most.

M
Assuming non-optimal strategy, a team can have this situation come up on tossups 11, 13, 15, 17, and 19 (or 12 14 16 18 20). Do you really want 25% of your tossups putting teams in that difficult position?
Okay, so according to your analysis it can happen on at very most 25% of the tossups, if a team is silly and if a hugely unlikely series of events happens. Still, in reality, it happens on less than 1% of questions. So yeah, it doesn't really matter even if we assume that it's bad.

M
Mike Sorice
Coach, Centennial High School of Champaign, IL (2014-) & Team Illinois (2016-2018)
Alumnus, Illinois ABT (2000-2002; 2003-2009) & Fenwick Scholastic Bowl (1999-2000)
ACF
IHSSBCA
PACE

User avatar
Captain Sinico
Auron
Posts: 2840
Joined: Sun Sep 21, 2003 1:46 pm
Location: Champaign, Illinois

Re: Bounceback Theory, NSC and Elsewhere

Post by Captain Sinico » Wed Jun 09, 2010 6:23 pm

The Granny wrote:
jdeliverer wrote:Assuming non-optimal strategy, a team can have this situation come up on tossups 11, 13, 15, 17, and 19 (or 12 14 16 18 20). Do you really want 25% of your tossups putting teams in that difficult position?
Teams are seriously going to be saying on tossup 11, "Well, I know this and can buzz in, but if I do then the other team could gain as many as 20 points on us and that will allow them to win the game"? Really? Under any circumstances? Seriously?
Just to clarify, people are talking about the following situation:
Team A is up by 20 or less on tossup 20, or between 60 and 80 on tossup 19, or between 100 and 120 on tossup 18, etc. Team B gets the tossup wrong. Team A knows the tossup. Team A then faces the following decision:
1. Get the tossup, which gives Team A 10 points and control of a bonus (so a shot at 30 points) but also gives Team B a shot at 30 points.
2. Don't get the tossup, which costs team A 10 points and a shot at 30, but also costs Team B a shot at 30.

Assuming that Team B can't change the score otherwise, e.g. through protest of a previous question; that Team A cares only about winning this single game, so that they care about their stats only negligibly; and that Team A isn't completely sure Team B won't get any rebound points on the bonus, which is vacuous since one can't be sure of such things, the optimal strategy is to not get the tossup, because doing so means that Team A definitely wins. If Team A does get the tossup anyway, it has kept Team B in the game at least until the conclusion of the bonus.
This can and does happen, but not in every match, since it requires the losing team to happen to neg/zero a specific question when the score is in a specific range. Like we've discussed above, it definitely can't happen more than 5 times in a match and also shouldn't happen more than once. However, you can note that this could have happened, for example, in the PACE Final if Southside had tried for and missed tossup 19. Even so, I just don't think it's a big negative against the format.

M
Mike Sorice
Coach, Centennial High School of Champaign, IL (2014-) & Team Illinois (2016-2018)
Alumnus, Illinois ABT (2000-2002; 2003-2009) & Fenwick Scholastic Bowl (1999-2000)
ACF
IHSSBCA
PACE

Locked