Buzzing In Too(?) Early

[gyre and gimble]

I was wondering about a particular situation that can arise when playing a tossup with either 1) ambiguous clues or 2) things with multiple names:

1. Suppose you have a tossup leading in with the clue "this poet wrote about a prickly pear leaf," where the answer is Marianne Moore, and a player buzzed in at "pear" with T. S. Eliot. (This is based on a real example from ACF Nationals 2013.) If the moderator stopped reading immediately, the player would either be right or win the protest. But if the moderator didn't react fast enough, or if even the player didn't buzz fast enough, so that the word "leaf" was read, then T. S. Eliot would no longer be a valid answer and the player would be negged. The "incorrect" answer is not the player's fault, right? I don't know of any theory of quizbowl that requires you to wait to confirm that the answer is not something if all of the clues do in fact apply to that thing.

2. At STIMPY last year, there was a tossup that ended, "Name this grammatical case that is often translated using the word 'of' and indicates possession." I buzzed at "indicates" and said "possessive," but apparently the first syllable of "possession" had also been read as I buzzed (this was Skypebowl so there was a bit of delay as well) so I was prompted, could not come up with "genitive," and was negged. (I don't know if genitive is actually identical to possessive, but the answerline treated them as such.) Aside from the technicalities of the rules, did I deserve that neg? I don't think I did--why should I have to think about what the last word of the tossup will be, before I decide whether or not to buzz?

It's hard to tell exactly how much of the question a player heard before he buzzed, as opposed to before he answered. How do we deal with this sort of thing? Is it just a misfortune that players have to deal with occasionally? That doesn't seem satisfying to me.

[the return of AHAN]

I was wondering about a particular situation that can arise when playing a tossup with either 1) ambiguous clues or 2) things with multiple names:

1. Suppose you have a tossup leading in with the clue "this poet wrote about a prickly pear leaf," where the answer is Marianne Moore, and a player buzzed in at "pear" with T. S. Eliot. (This is based on a real example from ACF Nationals 2013.) If the moderator stopped reading immediately, the player would either be right or win the protest. But if the moderator didn't react fast enough, or if even the player didn't buzz fast enough, so that the word "leaf" was read, then T. S. Eliot would no longer be a valid answer and the player would be negged. The "incorrect" answer is not the player's fault, right? I don't know of any theory of quizbowl that requires you to wait to confirm that the answer is not something if all of the clues do in fact apply to that thing.

2. At STIMPY last year, there was a tossup that ended, "Name this grammatical case that is often translated using the word 'of' and indicates possession." I buzzed at "indicates" and said "possessive," but apparently the first syllable of "possession" had also been read as I buzzed (this was Skypebowl so there was a bit of delay as well) so I was prompted, could not come up with "genitive," and was negged. (I don't know if genitive is actually identical to possessive, but the answerline treated them as such.) Aside from the technicalities of the rules, did I deserve that neg? I don't think I did--why should I have to think about what the last word of the tossup will be, before I decide whether or not to buzz?

It's hard to tell exactly how much of the question a player heard before he buzzed, as opposed to before he answered. How do we deal with this sort of thing? Is it just a misfortune that players have to deal with occasionally? That doesn't seem satisfying to me.

Can't say I have sympathy for #1 as the whole sentence hasn't been read. You cite 'all clues' when all you've heard is poet, and prickly pear. Buzzing THAT early is a gamble. A +EV gamble, but a gamble nonetheless.
Protest on #2.

[gyre and gimble]

I was wondering about a particular situation that can arise when playing a tossup with either 1) ambiguous clues or 2) things with multiple names:

1. Suppose you have a tossup leading in with the clue "this poet wrote about a prickly pear leaf," where the answer is Marianne Moore, and a player buzzed in at "pear" with T. S. Eliot. (This is based on a real example from ACF Nationals 2013.) If the moderator stopped reading immediately, the player would either be right or win the protest. But if the moderator didn't react fast enough, or if even the player didn't buzz fast enough, so that the word "leaf" was read, then T. S. Eliot would no longer be a valid answer and the player would be negged. The "incorrect" answer is not the player's fault, right? I don't know of any theory of quizbowl that requires you to wait to confirm that the answer is not something if all of the clues do in fact apply to that thing.

2. At STIMPY last year, there was a tossup that ended, "Name this grammatical case that is often translated using the word 'of' and indicates possession." I buzzed at "indicates" and said "possessive," but apparently the first syllable of "possession" had also been read as I buzzed (this was Skypebowl so there was a bit of delay as well) so I was prompted, could not come up with "genitive," and was negged. (I don't know if genitive is actually identical to possessive, but the answerline treated them as such.) Aside from the technicalities of the rules, did I deserve that neg? I don't think I did--why should I have to think about what the last word of the tossup will be, before I decide whether or not to buzz?

It's hard to tell exactly how much of the question a player heard before he buzzed, as opposed to before he answered. How do we deal with this sort of thing? Is it just a misfortune that players have to deal with occasionally? That doesn't seem satisfying to me.

Can't say I have sympathy for #1 as the whole sentence hasn't been read. You cite 'all clues' when all you've heard is poet, and prickly pear. Buzzing THAT early is a gamble. A +EV gamble, but a gamble nonetheless.
Protest on #2.

It still remains, whether or not your case is sympathetic or not, that if you buzzed at just "prickly pear" and didn't hear the word "leaf," then you'd have a valid protest. It's not like you're buzzing on "This king had a wife named" and saying "Edward II." You've been given a real, substantive clue, and you've been told what the question is looking for (a poet), so you have everything you need to buzz.

And the reason I brought this up is, I'm not sure that #2 is protestable, precisely because there's no real way of determining, conclusively, how many extra words were read after the player buzzed. When we hand out powers, we require that the first word out of power has not begun to be read. The theory behind this is that in order to earn the benefit of a buzz before the powermark, the player must not have been given any information after the powermark. By analogy, in order to earn points for saying "possessive," the player must not have heard any part of the word "possession." The problem, of course, is that the player might have no idea what other names there are for the possessive case, but did not know that they would not be allowed to just say "possessive" at the moment they chose to buzz.

[the return of AHAN]

OK, how often is a protest on toss-up ambiguity after seven words actually won?

[Ike]

I believe NAQT already rules against the player with its first sentence rule.

I think in ACF or other tournaments where I've done protests, it's a case by case basis. Personally in the case of the prickly pear, I'm going to rule against the player. You can't just pick one poet who has written about a prickly pear and expect the entire question to be overturned...players should reasonably know that numerous poets have written about a prickly pear, (For example, Garcia Lorca has a poem about a prickly pear according to my quick googling.) Yes, there is no "universal algorithm" that determines where this point is in a question, but it's certainly not there.

Not entirely sure about what you're saying in case 2, Stephen...are you saying you shouldn't be negged for buzzing in before the tossup was over since you didn't know it was over? If so I certainly think you deserve the neg, at least under the current rules of applying negs uniformly. If you're saying that you didn't know what the last word of the tossup was and you would have changed your answer if you had known, that's pretty silly to complain about, almost everyone changes their answer when they have negged and a tossup is over.

[gyre and gimble]

OK, how often is a protest on toss-up ambiguity after seven words actually won?

I can only assume that I'm not being clear enough here, because I'm making a theoretical point that really can't be resolved by nitpicking about particular examples. To be crystal clear: Let's say that a tossup leads in with Clue A. It's a real, substantive, complete clue. Unfortunately, Clue A applies to both Person X and Person Y. Clue B, the next clue, only applies to Person Y. You buzz in at the very end of Clue A, intending to answer with Person X. But since the moderator didn't react fast enough to your buzz, part of Clue B was read after you buzzed. This extra information now invalidates your answer of Person X. You get negged.

Am I the only person that thinks this is problematic?

Not entirely sure about what you're saying in case 2, Stephen...are you saying you shouldn't be negged for buzzing in before the tossup was over since you didn't know it was over? If so I certainly think you deserve the neg, at least under the current rules of applying negs uniformly. If you're saying that you didn't know what the last word of the tossup was and you would have changed your answer if you had known, that's pretty silly to complain about, almost everyone changes their answer when they have negged and a tossup is over.

I agree with both of those statements, obviously, but neither are the situations I'm talking about.

A similar situation would be a question that asks, "For 10 points, name this early Holy Roman Emperor, the son of Henry the Fowler, who was known as 'the Great' and won at Lechfeld. You buzz at "known as," because you know that the answer is "Otto the Great." But by the time the moderator realizes that you're buzzing, he's already read "the Great." The answerline says to prompt on "Otto the Great" after "the Great" has been read. But you don't know that he's called "Otto I." So you get negged. Should you have predicted that the moderator would say "the Great," then realized that you didn't know his other name, and then decided not to buzz at all? Even though "Otto the Great" was a perfectly valid answer at the point you were buzzing in?

So it's not that I would have changed my answer from a wrong one to a right one if I had heard the last word. My original answer was right, but it was no longer acceptable because it was stated in the question, at some point between when I buzzed and when the moderator stopped reading.

[Irreligion in Bangladesh]

Questions that spot you "the Great" should just be accepting "Otto" once you've said "Great," as they should just be accepting "Richard" after spotting you "the third of his name."

[gyre and gimble]

Questions that spot you "the Great" should just be accepting "Otto" once you've said "Great," as they should just be accepting "Richard" after spotting you "the third of his name."

So should a question that says "name this grammatical case that indicates possession" accept "possessive case" after "possession" is read?

Or a question that says, "name this American writer of The Adventures of Tom Sawyer, also called Samuel Clemens"? If you buzz in at "also," intending to say "Samuel Clemens," and then moderator spillover causes you to hear "called Samuel Cle-" after you buzz, do you have to say "Mark Twain"? Do you deserve to be negged because you can't think of the name "Mark Twain," and only "Samuel Clemens"?

I don't think this is a question that can be answered by just saying, "Hey, that question was just written poorly," or "Hey, the answerline should be different." Some questions will inevitably be written in an unreasonable manner that is not necessarily protestable.

[Cody]

I don't think there's any real solution to #1; players have to be smart and editors have to try to know edge cases for lead-ins. Ike is right that such protests have to be evaluated on a case-by-case basis -- I don't know that you can definitively rule anything outside of that.

I would say that #2 is mostly a writer problem. (and not in the Samuel Clemens case -- that's the player's fault and there's no reason to have sympathy in that situation). The key is that the question should not become HARDER as it goes on. Given that the genitive case is commonly called the possessive case, and people are more likely to say "possessive" than "genitive", then you should drop "genitive" in the giveaway instead of "possessive". I don't know of any situations where this occurs where it can't be improved by thinking about how most players are going to answer and then writing the question better. (not that this is always easy -- I've certainly been guilty of it.)

[Skepticism and Animal Feed]

I strongly agree with Cody that problem #2 is being caused entirely by clunky, non-pyramidal giveaways, rather than by a flaw or problem with the rules.

There's no way you could write a rule protecting players from that situation: there's no way to tell that they actually were buzzing with "possessive" or "Otto the Great" before the moderator blurted that out, rather than buzzing with that because they just heard it. This is something where editors and question writers, not protest committees, must protect the player.

If something has multiple names, great: respect the fact that people who interacted with that thing in different contexts may call it different things, and don't randomly make one of them unacceptable by the way you write the question.

[theMoMA]

I don't really think #1 is a conundrum. Here are the rules in question:

4. The following potential errors, and only the following, are protestable: ...

4.2 The question or answer was ambiguous, so the answer given by the player should be accepted since it fits all the clues given in the entire question text, or all the clues up to a significantly deep point in the question.

If the clues of a tossup question (at the point at which a player signaled) do not uniquely specify an answer, then the tournament director should consider when the signaling occurred:

a. If the player signaled prior to the end of the first sentence of the question, the response shall be treated as incorrect. That is, players may not protest that they gave an answer that was “correct when they buzzed” during the first sentence of the tossup.

b. If the player signaled after the end of the first sentence, the response shall be accepted if it is correct (for all the clues that had been read) and precise. If the response was correct but imprecise (and thus should have been prompted), the remedy of Rule J.13.g should be applied.

Under the NAQT rules, this buzz is incorrect. Under ACF rules, there is a judgment call about whether the clues had reached a "significantly deep point." This case doesn't seem to be a seriously troubling one; it appears that many poets have written about prickly pears, so I think a reasonable protest committee would deny the protest and uphold the neg.

Both NAQT and ACF rules require a player to assume some risk when buzzing extremely early. On the NAQT side, the tipping point is set in a concrete place. On the ACF side, it requires a judgment call, and there's obviously a spectrum of reasonableness; a protest after a buzz after the words "this author wrote about a man" is never going to be upheld, but a buzz after a more concretely identifying clue might. As I said, I don't think "prickly pear" rises to the necessary level, but I acknowledge that this is a judgment call that another person might make differently.

(This is for another discussion, which has been ongoing in some form or another for a few years, but I do believe that it makes more sense to throw out questions that have ambiguous lead-ins than it does to simply award points, although perhaps this is only appropriate when the alternate clue-answer pairing is significantly easier than the actual, listed answer. For example, if, at an open-level tournament, a very difficult question on the postmodern novelist Kathy Acker carelessly begins with "in this author's Great Expectations, ..." and someone buzzed with Dickens, I think it would make little sense to treat the question as essentially a five-word speedcheck on Dickens and credit the buzz, but I'm also somewhat uncomfortable saying that a player who buzzes there should be wrong under the ACF rules. On the other hand, if, at an easier event, a tossup on Dostoevsky began "in this author's The Double, ..." and a player buzzed with Saramago, I would be inclined to credit the buzz, because both are plausible clues to plausible answers at that point, and it doesn't seem to implicate any unfairness to award points. I don't know if this makes any sense, and I understand the other side of the argument that we shouldn't be throwing questions out willy-nilly or undoing correct buzzes, but it seems to me that there is room for judgment to govern these situations.)

[Æthelred the Unready Steady Cook]

This is unlikely to help but there are problems with the second question which I think complicate things. The genitive is not simply a possessive case. In both ancient Greek and Arabic it is used prepositionally. I think the Arabic clue points towards possessive because idafa can't replace the prepositional role and thus not the genitive at large, while the Greek -wv ending is definitely genitive plural and not simply possessive plural because people never talk about there being a possessive case in Greek, only a genitive. Really I think you were a little unlucky given the tangle that this question is, but genitive is not an obscure term and possessive is very wooly, given that I'm not sure it explicitly exists in any language.