2019 Terrapin general discussion

[wcheng]

Thank you to all of you for playing 2019's iteration of Terrapin! We hope that you enjoyed the efforts of our writers and editors during this weekend and the weekends that follow. We hope that our decision to write for medium, as opposed to "regular" difficulty, made our set more accessible while still retaining the creativity and craftsmanship that has characterized past iterations of Terrapin like 2016's.

Before anything else, I want to take the time to thank all of our writers! Twelve past and present Terps wrote questions for this set: Graham Reid, Caleb Kendrick, Alex Echikson, Ani Perumalla, Weijia Cheng, Justin Hawkins, Jason Shi, Vishwa Shanmugam, Jack Lewis, Jack Nolan, Jordan Brownstein, Sarang Yeola, and Naveed Chowdhury. I especially want to thank Alex and Ani who made huge contributions to the set (over 90 questions each!) despite not being editors—Alex was instrumental in getting the history done early and providing me a lot of good material to edit off of, and Ani contributed great questions throughout the set in nearly every category, while also doing a lot of work on feng shui.

The five editors on this set were me, Jordan, Graham, Caleb, and Ophir. I edited history, religion, other, and about half of social science (the rest I deferred to Caleb). Jordan edited literature, visual arts, and mythology. Graham edited science with some oversight and assistance from Andrew Wang, Caleb edited philosophy and some social science, and Ophir edited auditory arts. I'll let each editor comment on their own particular goals and approaches for this set, but as an editing team we were strongly focused on accessibility while still trying to introduce some new ideas and novel perspectives on old answerlines. I'll let you, the players, be the judge of how successfully we accomplished our vision.

Some other people I would like to thank for their help on the set are:

• JinAh Kim, who helped proofread and provided comments on the set in the final week before the Discord playtest;
• Will Alston, who graciously offered to playtest the history and provided useful comments on the category;
• Jason Golfinos, Victor Prieto, Mike Bentley, and Chris Borglum, who provided detailed notes and useful feedback after our Discord playtest;
• and all of the other Discord playtesters, since every team submitted feedback after the tournament.

Feel free to leave any general comments you have about the set in this thread. Comments on/requests to see specific questions will be posted in the specific question discussion thread.

[wcheng]

Separate from the opening post (to avoid conflating my personal approach with the approach of the editors as a whole), I would like to explain my philosophy of writing and editing for this set, or really the relative lack of one. I would say that when it comes to editing questions, I am primarily focused on craftsmanship—that is to say, whether the clues are factually correct, whether the clues are pyramidal, whether the questions are worded in a way to steer players towards a confident buzz, how to remove (or at least properly acknowledge) ambiguity—fundamental things like that. That's not to say that I don't think about other things like "classroom" vs. "non-classroom" content or diversity and representation, because I consider those carefully as well, but I have tried to address these concerns to the extent of my ability without sacrificing any playability in the set.

Terrapin is my first major editing project and will probably be my last. I do not intend to write or edit for any future tournaments, and while I would be very flattered to receive any requests to collaborate, I am in the process of winding down my involvement with quizbowl and 2019 Terrapin is the beginning of my one-and-a-half-month-long swan song. I really hope this tournament meets both my and your expectations, but if it doesn't, please feel free to offer any criticisms that you see fit! I will gladly make any corrections necessary before the remaining mirrors of this set.

[AGoodMan]

I generally enjoyed this set. Thank you to the writers and editors for doing a great job. Some comments from our lit players: a lot of the more “creative” answer lines appeared in the earlier rounds and the more “bread-and-butter” answerlines appeared later.

[Carlos Be]

I thought the set was pretty good overall, particularly the art.

One thing that I noticed that got pretty annoying was that a lot of times, the first part of a bonus would go on about lots of cool and interesting things but then hit you with a giveaway. Easy parts should be short and get to the point as quickly as possible.

Another thing, although this is almost certainly a result of some sort of cognitive bias, was that there seemed to be much more computer science than math. I don't want to derail this discussion with an argument about how much math there should be per packet, but I will say that I don't think combining math and comp sci in the distribution is a good idea. Unless every question is about primality testing or factorization algorithms, math and computer science have very little overlap.

[wcheng]

One thing that I noticed that got pretty annoying was that a lot of times, the first part of a bonus would go on about lots of cool and interesting things but then hit you with a giveaway. Easy parts should be short and get to the point as quickly as possible.

Not sure if I did this myself in the lead-ins but I've definitely done this on some easy parts of bonuses. Honestly, I think this is a great thing to do; it can help keep players who don't know the "hard" content engaged and provides people opportunities to discuss and learn new things.

EDIT: Come to think of it, I have definitely done this in questions I've written or edited. Example:

During this war, an Irish immigrant named Jennie Hodgers took the name of Albert D. Cashier and served as an infantryman, continuing to live as a man after this war. For 10 points each:
[10] Name this war in which a number of women served in combat roles in disguise as men, such as Loreta Velazquez, who was exposed after being wounded by an exploding shell at the Battle of Shiloh.
ANSWER: American Civil War [or War Between the States; accept War of Northern Aggression]
[10] During the Civil War, this woman earned the nickname “Angel of the Battlefield” for her work in bringing supplies to wounded soldiers at Antietam and elsewhere. This woman founded the American Red Cross in 1881.
ANSWER: Clara Barton [or Clarissa Harlowe Barton]
[10] This woman became the first, and so far, only female Medal of Honor recipient for serving as a Union surgeon. After the war, her preference for men’s clothing led her to be arrested several times for impersonating a man.
ANSWER: Mary Walker [or Mary Edward Walker]
<AE, American History>

I'm sure that the first part could have been much shorter and to-the-point, but I think that would have undermined the theme of the bonus and not provided players as much of an opportunity to learn about some interesting historical figures that they might not otherwise have been exposed to. If the point of quizbowl was just to test people's knowledge as quickly as possible, I'd agree with your suggestion, but I think quizbowl serves a wider purpose than that. Tamara Vardomskaya made a great post about this idea, which I hope I implemented successfully in my questions.

Another example from the same packet:

In his apocryphal Acts, this apostle converts a rich woman named Thecla who spends three days watching him preach from the window before eventually baptizing herself in a pool of killer seals. For 10 points each:
[10] Name this apostle who converted to Christianity after being blinded by a heavenly light on the road to Damascus. Thirteen of the twenty-one New Testament epistles are traditionally attributed to him.
ANSWER: Paul the Apostle [or Paulos; or Saul of Tarsus]
[10] In the Acts of Paul and Thecla, Paul preaches sexual abstinence, similar to chapter 7 of this epistle in which he declares, “It is good for a man not to have sexual relations with a woman,” while giving advice on marriage.
ANSWER: 1 Corinthians [or First Epistle to the Corinthians; prompt on Corinthians]
[10] In both First Corinthians and the first epistle to this person, Paul declares that women should be silent; in that epistle to this person, Paul declares that “I suffer not a woman to teach, nor to usurp authority over the man.”
ANSWER: Timothy [or Timotheos]

Here, the lead-in is unlikely to help you unless you're like me and really into apocryphal Christian texts, but again it provides context to the rest of the question and exposes players to some interesting (and rather funny, in my opinion) content that they might otherwise never have heard of.

[a bird]

Another thing, although this is almost certainly a result of some sort of cognitive bias, was that there seemed to be much more computer science than math. I don't want to derail this discussion with an argument about how much math there should be per packet, but I will say that I don't think combining math and comp sci in the distribution is a good idea. Unless every question is about primality testing or factorization algorithms, math and computer science have very little overlap.

I intended math and CS to occupy roughly equal parts of the distribution. I wasn't a primary writer for math or CS for this tournament, but I certainly wasn't looking at math and CS as a combined category, and I don't think any other writers were. Here is the breakdown of questions:

Math
First 11 packets: 3/3
Total: 4/5

CS
First 11 packets: 3/3 (11 packets)
Total: 4/4

[Carlos Be]

I'm sure that the first part could have been much shorter and to-the-point, but I think that would have undermined the theme of the bonus and not provided players as much of an opportunity to learn about some interesting historical figures that they might not otherwise have been exposed to.

For the civil war question, all of these historical figures could have been asked about if the Civil War were made the hard part by replacing "Battle of Shiloh" with a clue from one of the later parts or cutting it entirely, and then making one of the later parts a quick easy part. This structure would also reward people for knowing about Jennie Hodgers and Loreta Velazquez.

Also, as a more general point, I find that hard clues in an easy part stick with me much less than clues in a medium or hard part, because once I hear the giveaway I stop thinking about the clues, whereas with a medium or hard part I'll spend some time going over the clues in my head to try to come up with a plausible answer. If the goal of the bonus is to highlight the contributions of women in the civil war, then I think a bonus structure that doesn't make these people footnotes to the Battle of Shiloh would be more effective.

[Cody]

Your personal preferences do not constitute a mandate.

Including harder clues in easy parts makes questions more interesting for players, moderators, and editors. There are legitimate problems with such constructions when they interfere with playability, but you are not critiquing such cases.

Chill out and please stop rehashing this issue in every tournament thread.

Edit: I was a bit too cranky when I posted this, having recently read the EFT general discussion. I could've worded this better.

[Carlos Be]

Your personal preferences do not constitute a mandate.

Including harder clues in easy parts makes questions more interesting for players, moderators, and editors. There are legitimate problems with such constructions when they interfere with playability, but you are not critiquing such cases.

Chill out and please stop rehashing this issue in every tournament thread.

I have no idea who you are or what you have to do with this tournament. Weijia, who was actually involved with the set, responded to my point with actual substantive comments. Obviously, my personal preferences do not constitute a mandate and that's why I included my reasoning in my response. Your personal preference that stuffing easy parts with hard clues is more interesting is not a mandate either, so you should take a cue from Weijia and respond constructively if you disagree.

Also, I don't know who needs to chill out more: the person who recently played a tournament and is now offering (mostly positive) feedback, or the person who had nothing to do with the set that has rudely interrupted a constructive conversation.

[wcheng]

I'm sure that the first part could have been much shorter and to-the-point, but I think that would have undermined the theme of the bonus and not provided players as much of an opportunity to learn about some interesting historical figures that they might not otherwise have been exposed to.

For the civil war question, all of these historical figures could have been asked about if the Civil War were made the hard part by replacing "Battle of Shiloh" with a clue from one of the later parts or cutting it entirely, and then making one of the later parts a quick easy part. This structure would also reward people for knowing about Jennie Hodgers and Loreta Velazquez.

Also, as a more general point, I find that hard clues in an easy part stick with me much less than clues in a medium or hard part, because once I hear the giveaway I stop thinking about the clues, whereas with a medium or hard part I'll spend some time going over the clues in my head to try to come up with a plausible answer. If the goal of the bonus is to highlight the contributions of women in the civil war, then I think a bonus structure that doesn't make these people footnotes to the Battle of Shiloh would be more effective.

To bring this discussion back on topic, I think that this just indicates a difference in people engage in the question. In my experience, people tend to pause to acknowledge novel information has been introduced in the lead-in, because it "primes" them for the rest of the question, but everyone engages with questions differently. Obviously, different bonus constructions work better for different bonuses, and I wouldn't force this construction if I thought it was unnecessary. I'm sure you can find many examples of straightforward easy parts throughout my categories.

[Auks Ran Ova]

Your personal preferences do not constitute a mandate.

Including harder clues in easy parts makes questions more interesting for players, moderators, and editors. There are legitimate problems with such constructions when they interfere with playability, but you are not critiquing such cases.

Chill out and please stop rehashing this issue in every tournament thread.

I have no idea who you are or what you have to do with this tournament. Weijia, who was actually involved with the set, responded to my point with actual substantive comments. Obviously, my personal preferences do not constitute a mandate and that's why I included my reasoning in my response. Your personal preference that stuffing easy parts with hard clues is more interesting is not a mandate either, so you should take a cue from Weijia and respond constructively if you disagree.

Also, I don't know who needs to chill out more: the person who recently played a tournament and is now offering (mostly positive) feedback, or the person who had nothing to do with the set that has rudely interrupted a constructive conversation.

You don't get to dictate who is and is not allowed to respond to your posts, nor is it inherently "rude" for someone to join a discussion. "Who Cody is", relevant to the topic at hand, is someone who has written a lot of questions and has as much right to opine about the proper nature of their construction as you do. This is especially true when you take your criticisms from the specific to the general, as in the statement "Easy parts should be short and get to the point as quickly as possible." From both this thread and the EFT discussion, you seem to have the opinion that it's inappropriate for a bonus leadin to be used for an amusing or interesting clue if the first part then turns out to be the easy part; there's nothing especially wrong with that being your personal taste, but it's by no means axiomatic.

To bring this discussion back on topic, I think that this just indicates a difference in people engage in the question. In my experience, people tend to pause to acknowledge novel information has been introduced in the lead-in, because it "primes" them for the rest of the question, but everyone engages with questions differently. Obviously, different bonus constructions work better for different bonuses, and I wouldn't force this construction if I thought it was unnecessary. I'm sure you can find many examples of straightforward easy parts throughout my categories.

This seems like a perfectly reasonable writing philosophy!

[$5 Bits of Broken Chair Trophy]

I thought that the science in this set was excellent, both in terms of cluing and answer selection; and I also greatly appreciated the decision to not clue any named reactions in the chemistry- it can certainly be tempting to fill the early clues with cascades of them, and there are plenty that are useful and important, but using that early clue space to ask about concepts in ways that maybe haven't been explored before is definitely preferable in my opinion to beating players over the head with third and fourth-tier named reactions.

[Gen. Winfield Scott Hancock]

I thought this set did an excellent job of hitting its target goals. For the most part I can only speak to my categories, but I thought the visual arts, pre-20th century American history, British history, and Christianity were well-written. The visual arts particularly did what I am personally a fan of at this difficulty, in that it provided good tests of deep knowledge without trying to expand the canon too much with brand-new answerlines.

Also the science tossup answerlines felt accessible even for someone as incompetent at science as myself, so I believe that to be deserving of praise as well!

[TaylorH]

Congrats to the editors and writers on a solid set. I took notes while playing on my thoughts on specific questions, which I will dump in the specific question thread shortly. Some general thoughts on the set overall:

-This set felt a notch harder than EFT, particularly on some of the TU answerlines. I only mention EFT specifically because this tournament had the same difficulty rating on the dot scale. To me, this set felt more like 2.5 dots rather than 2
-The topic selection in RMPSS, Literature, and Fine Arts was excellent. Lots of interesting things asked about. I felt rewarded for knowing both extra-canonical things and deep clues for super canon stuff. A good balance!
-Science felt a tad easier than the other big categories. Maybe this is good? I certainly appreciated the reduction in orgo and the increased applied/engineering-y things that came up, though this is probably a personnel taste.
-History had lots of questions on "this monarch", enough for people to take note at our site. This felt a little stale after 10 rounds.
-American History was probably my favorite sub category: lots of neat social history clues.
-There were some minor issues with transparency of some TUs, difficulty cliffs, and swings in bonus difficulty. I will make notes of specifics of these in the other thread.

[Zealots of Stockholm]

I forgot to give feedback on this earlier this week, whoops!

I remember liking most of the questions I heard while reading, though don't remember too many specifics since I find it a lot different than the experience of playing. I do remember generally liking the lit, US history, and painting, though the Sioux tu seemed quite fraudable and the lit tu on Parma seemed way harder than any of the other lit I read all day.

It may have just been the day wearing on me, but it seemed that there were several bonus parts that were three lines which seemed like they could have been shorter, and I found myself complaining about them in the second half of the day more often. I don't remember too many specific parts, but the double helix part Taylor mentioned in the other thread seems like a good example.

Overall, the teams at our site seemed to enjoy the set. Thanks to the writers/editors for producing a good set!

[A Dim-Witted Saboteur]

A few impressions

-science seemed easy in general, especially bio
-This tournament's tossups felt below EFT in difficulty, but bonuses, especially hard parts, did not. Who were parts like "Zazie" (a book that no one in our field had heard of besides Mitch McCullar, who still did not convert it because it only gave plot clues from an already very obscure novel) written for?
-Is there any particular reason why the finals packets seemed to scale up to 3 on the difficulty scale, if not higher, and this fact was not announced in the tournament announcement? I find it difficult to imagine, say, EFT, tossing up the White Tiger, Pierre Corneille, the Shawnee, or the Popish Plot (there are more examples that I will provide once I have the answerlines in front of me)
-The history struck me as well-executed in general, with the exception of a few fairly early clue drops; props to Weijia (the Shawnee tossup especially was one I enjoyed)
-I was not a big fan of the way this tournament's geo was executed; much of it (e.g. Galapagos) seemed rather transparent. You could definitely ascribe this to personal preference, so it's not necessarily something the editors should concern themselves with.

[wcheng]

-Is there any particular reason why the finals packets seemed to scale up to 3 on the difficulty scale, if not higher, and this fact was not announced in the tournament announcement? I find it difficult to imagine, say, EFT, tossing up the White Tiger, Pierre Corneille, the Shawnee, or the Popish Plot (there are more examples that I will provide once I have the answerlines in front of me)

In the interests of transparency, there was never any conscious effort to make the finals packets harder than the rest of the set, but this was the end result of our packetization efforts in which all of the "outlier" questions tagged with "move to finals" were, well, moved to the finals. I apologize if this made playing the finals jarring, though I would hope that it wouldn't be too much of an issue for most teams that make the finals.

[CPiGuy]

-Is there any particular reason why the finals packets seemed to scale up to 3 on the difficulty scale, if not higher, and this fact was not announced in the tournament announcement? I find it difficult to imagine, say, EFT, tossing up the White Tiger, Pierre Corneille, the Shawnee, or the Popish Plot (there are more examples that I will provide once I have the answerlines in front of me)

In the interests of transparency, there was never any conscious effort to make the finals packets harder than the rest of the set, but this was the end result of our packetization efforts in which all of the "outlier" questions tagged with "move to finals" were, well, moved to the finals. I apologize if this made playing the finals jarring, though I would hope that it wouldn't be too much of an issue for most teams that make the finals.

Why would you deliberately move all the "outlier" questions to finals rather than spreading them out? That just seems like a good way to ensure a poorer playing experience for teams that make the finals.

More general comments:

-- This set was a lot of fun to play, and, I thought, didn't overshoot difficulty (an appreciated break after PIANO last weekend, lol.) I agree with Jakob's point about bonus hard parts, but:
-- Overall it actually felt like there were a lot of bonuses with effectively two easy parts. I'm interested to see whether the advanced stats bear this out, but while playing the tournament it felt like teams just got free twenties multiple times per round.
-- It felt like the level of transparency was higher than normal; my impression was that there were a lot more tossups than usual that were not difficult to fraud.

[wcheng]

-Is there any particular reason why the finals packets seemed to scale up to 3 on the difficulty scale, if not higher, and this fact was not announced in the tournament announcement? I find it difficult to imagine, say, EFT, tossing up the White Tiger, Pierre Corneille, the Shawnee, or the Popish Plot (there are more examples that I will provide once I have the answerlines in front of me)

In the interests of transparency, there was never any conscious effort to make the finals packets harder than the rest of the set, but this was the end result of our packetization efforts in which all of the "outlier" questions tagged with "move to finals" were, well, moved to the finals. I apologize if this made playing the finals jarring, though I would hope that it wouldn't be too much of an issue for most teams that make the finals.

Why would you deliberately move all the "outlier" questions to finals rather than spreading them out? That just seems like a good way to ensure a poorer playing experience for teams that make the finals.

This has been done on all the sets I've worked on so far in the past, and I believe this is common practice. My understanding of the issue is that having harder questions in the finals is not a huge issue because 1) teams that get to the finals presumably have more knowledge and 2) harder questions will distinguish between more skilled teams. I definitely would agree that there is an issue if there is too large of a discrepancy, however.

[Cody]

Why would you deliberately move all the "outlier" questions to finals rather than spreading them out? That just seems like a good way to ensure a poorer playing experience for teams that make the finals.

As long as it's at a low level, it "evens" out the set in terms of the subset of teams that are playing on it. Moving a few outliers will, theoretically, not cause a big swing in difficulty. The benefits are that the weakest teams aren't beaten down by them and instead they're played by the strongest teams in the tournament, with the small swing in difficulty having little effect on the finals. Weijia is correct in noting that it is common practice.

Of course, the "why" is different from the practice and my personal opinion is that in practice it generally doesn't achieve its goal and it's better to spread outliers out (to keep better teams honest across the whole tournament) or cut them.

[John Ketzkorn]

Why would you deliberately move all the "outlier" questions to finals rather than spreading them out? That just seems like a good way to ensure a poorer playing experience for teams that make the finals.

As long as it's at a low level, it "evens" out the set in terms of the subset of teams that are playing on it. Moving a few outliers will, theoretically, not cause a big swing in difficulty. The benefits are that the weakest teams aren't beaten down by them and instead they're played by the strongest teams in the tournament, with the small swing in difficulty having little effect on the finals. Weijia is correct in noting that it is common practice.

Of course, the "why" is different from the practice and my personal opinion is that in practice it generally doesn't achieve its goal and it's better to spread outliers out (to keep better teams honest across the whole tournament) or cut them.

Personally, I think I agree with Jakob here. It seemed more than "a few" outliers as we very quickly noticed a jump in difficulty. The frustrating thing about it was that the bonuses remained the same. I would have preferred if the harder tossups were spread out throughout all the packets and not jumbled together (though I guess I can't complain given how we seemed to be favored by the difficulty increase). It's fine to do this as editor's have their own philosophies, but please warn players that the finals will be more difficult.

[John Ketzkorn]

Putting my overall general feelings towards this set in a different post.

I actually loved this set. It did a good job of hitting its target difficulty, and it was well worth the drive up to East Lansing for. In general, I'm not very good at guiding specific praise, so its hard to say what this set did that was different from other sets, but certainly this set did a good job of asking about a lot of important things from a wide variety of disciplines. I remember everything being well within the realm of reason to expect people to know at least something about, and I think this has been my favorite set of the school year. I'll leave some of my gripes of the science for the specific question feedback, but overall the science did a great job of being accessible (watching Jakob Meyers 30 a computer science bonus was legendary).

[Cheynem]

I've started a different thread on the "harder questions in the finals" issue, which I think is very interesting and not just related to this tournament.

[Smuttynose Island]

I may have more to say after looking over the set, but this was my favorite tournament of the year. The set did a great job hitting its target difficulty in the rounds played at UW. With a healthy mix of "interesting" and "well-trodden" clues, the set was rewarding to play, while still remaining very accessible.

[1992 in spaceflight]

I read the set at WUSTL, and I just wanted to say that the set did a great job of hitting its target difficulty as intended. That's really hard even for lots of very-experienced editors to do, so great job to the Terrapin team.

I look forward to reading more sets written and edited by you all in the future.