2022 PACE NSC Discussion Thread

[Mike Bentley]

Thanks everyone, hope you all enjoyed the set! Packets are posted here: https://hsquizbowl.org/db/questionsets/2786/. Will likely post a slightly fixed version later in the week, along with docx files.

Longer post coming soon.

[Mike Bentley]

Thanks to everyone who made the NSC happen this year!

The editors for this year's set were:
Mike Bentley - European and Cross History; Visual Fine Arts; Social Science; Geography; Current Events; Other Academic
Hari Parameswaran - American and World History
Joseph Krol - Physics; British and World Literature
Wonyoung Jang- A lot of the American and British Literature; Audio Fine Arts
Jonathan Magin - Some of the British literature tossups
Taylor Harvey - Much of the American and British literature bonuses
Matt Bollinger - Myth
Jordan Brownstein - Other religion and philosophy (with some help from Caleb Kendrick)
Andrew Wang - Biology and Chemistry
Kevin Wang - Other Science
Jon Suh - Judeo-Christian Religion

These people all wrote at least one question for the set:
Alyssa Jorgensen
Andrew Wang
Anson Berns
Benjamin Chapman
Brian Kalathiveetil
Briana Magin
Caleb Kendrick
Caroline Mao
Charles Yang
Chauncey Lo
Clark Smith
David Reinstein
Elaijah Lapay
Emmett Laurie
Eric Yin
Fredrick Morlan
Gabe Guedes
Hari Parameswaran
Jakob Boeye
Jon Suh
Jonathan Magin
Jonathen Settle
Jordan Brownstein
Joseph Krol
Juliet Mayer
Kelin Carpenter
Kevin Wang
Lalit Maharjan
Matthew Bollinger
Michael Bentley
Noah Sheidlower
Rahul Keyal
Ridge Ren
Sameen Belal
Tony Chen
Victor Prieto
Vikshar Athreya
Vishwa Shanmugam
William Grossman
William Yaeger
Wonyoung Jang

Big thanks too to the proofreaders and playtesters. Matt Weiner, Lia Rathburn and Tracy Mirkin were particularly dedicated in attending almost every session and giving great feedback.

For more specific details on editing, let me first shout out Joseph Krol. He helped ensure a science category (physics) got done months before the deadline, an unheard of feat. He also wrote a ton of questions and edited the Euro and World lit categories. All of this on top of the seeming 10 other sets he worked on this year.

Andrew and Kevin continue to produce consistently great science categories full of interesting answers and clues. And as with last year, a lot of the questions attributed to me in these categories are basically full re-writes from these editors.

Huge thanks to Jordan Brownstein, Jonathan Magin and Taylor Harvey who stepped in near the end and helped edit some categories. Jonathan also wrote most of the music, one of the harder NSC categories to fill out.

It was great having Hari as a first-time editor for the NSC. He had a ton of great ideas (in particular the radiation poisoning tossup) and also did a good job making sure these categories stayed balanced, injecting some last-minute questions in American History to make that holistically better.

Like last year, Wonyoung's music editing consistently finds interesting and important new clues and answer lines. We had a bit more blurring of the line between "fine arts" and "trash" this year, I'll be curious to see how that was received.

Having just edited the 2021 NSC, I mostly came into this year's set with a "more of the same" attitude. We did make a few changes from last year, decreasing the philosophy a bit in favor of other academic. (The basic idea here being is that this is a very difficult subject to keep difficulty appropriate with the number of answer lines needed for the NSC.)

Jon and Jordan collaborated with a lot of great "real" religion content. And Jordan and Caleb really helped keep the philosophy category both interesting and accessible.

Finally, you all already know how great of an editor Matt is. A great blend of core-content and new ideas in this category.

And although he's not an editor, let me recognize Ben Chapman who really helped out with some of the hardest categories to write: music and chemistry.

--

In general, my philosophy with this set was "more of the same." I think last year's set ended up pretty well and didn't seek to change too much. That being said, we had some different editors and writers from last year vs. this year so I'm sure some categories played differently.

I've already addressed a few of the points of feedback on the Discord. Some of the general themes that surfaced there were:

-We probably clued too many "adaptation" clues. I'm willing to defend all of these clues individually, but I agree that we probably should have culled a couple of them to have a slightly more balanced set of clues.
-Literature in general may have played hard.
-Mythology perhaps ended up with a greater mix of Greco-Roman than other myth systems, although the overall distribution was the same as last year.
-I personally like other academic and see it as a way to ask about various topics that are hard to ask about in other categories. It also lets us go deeper across 22-24 packets than some other categories without going overboard in difficulty. Some of my favorite questions were in this category (Stephen Hawking and Cantonese for instance). I'm curious to hear more about specific questions in this category.
-Social science perhaps skewed less canonical than other categories. I'm not actually sure this is the case but I'm open to discussion about this. Economics probably had slightly more "real world" clues than typical.
-Perhaps too many "clever" answer lines. Again, I'm willing to defend basically all of these. But I could see that the overall mix might be a little high.

I'm sure there was other stuff (including things not mentioned in the Discord). I encourage you all to post your thoughts here.

Reading the set, I think we could have stood to have made the early parts of tossups a little easier. This is always tricky to get right (and easy to err on the hard side). At some point, I'll take a closer look at conversion rates and power rates compared to last year. A very superficial comparison indicated similar stats for the top 20 or so teams.

[Mike Bentley]

I've updated the posted packets with a few small fixes and also added docx versions.

[ryanrosenberg]

I've compiled stats for NSC from the scoresheets, including category data: tossup stats and bonus stats. Will possibly replicate some of the analysis I did for ACF Nationals on this data, pending time over the next few weeks.

Let me know of any discrepancies between these stats and the scoresheets or via PM or email at [email protected].

[Mike Bentley]

Thanks Ryan!

Regarding "adaptation clues," you do see from this data that some of the literature questions that leaned heavily into that had poor power rates (Native Son, Red Badge and Great Gatsby, all with <= 2 powers). But other more conventionally clued tossups like Alexander Pope, Falstaff and The Dead also had low power rates. The much-discussed Virginia Woolf question was actually in the top third of powered questions. And questions that at least clued some "adaptation" material like All Quiet on the Western Front and Pirandello had some of the highest power rates in the set.

Of course, power rates are an imprecise metric. It's not necessarily a good thing if the tossup was "impossible" for the field and then there were lots of buzzer races on the still-in-power first "conventional" clues.

I do see that Social Science and Other Science played a little harder than I would have liked, with Economics one of the hardest-playing sub-categories in the set. These topics had an above-average number of newish-to-quizbowl answer lines which probably contributed to this. Although on its own that doesn't make a tossup hard to power. Radiation poisoning, Shinto shrines, abolition laws, censuses, tacos, witnesses, and declaring bankruptcy all had reasonably high power rates and are topics I'd put in this category. Which is mostly to say that it can be quite hard to fully predict what the field knows. This can be accounted for somewhat in playtesting, but all of these questions were playtested multiple times and we did still end up with several that had sub-optimal buzz distributions.

[CPiGuy]

I've compiled stats for NSC from the scoresheets, including category data: tossup stats and bonus stats. Will possibly replicate some of the analysis I did for ACF Nationals on this data, pending time over the next few weeks.

Let me know of any discrepancies between these stats and the scoresheets or via PM or email at [email protected].

Do these bonus stats accurately reflect the fact that tossups and bonuses were not paired?

Thanks for compiling them!

[ryanrosenberg]

I've compiled stats for NSC from the scoresheets, including category data: tossup stats and bonus stats. Will possibly replicate some of the analysis I did for ACF Nationals on this data, pending time over the next few weeks.

Let me know of any discrepancies between these stats and the scoresheets or via PM or email at [email protected].

Do these bonus stats accurately reflect the fact that tossups and bonuses were not paired?

Thanks for compiling them!

They should. Let me know if you find anything that seems to the contrary!

[Mike Bentley]

In the interest of getting some more formal discussion, I've created a Discord discussion server: https://discord.gg/eFhK5hYK