Short one this week. Catie is better (she went to school Thursday and had no problems,) but I got cut short by the calendar, I found it's later in the year than I thought.
This week's practice was the make up for last week's. It was also the last one for the calendar year, due to the school's mid-term period being next week, despite it not really being either the mid of -term, or the mid of this particular -term. So I had to put some things straight.
The first item on the agenda was reviewing whether we could use buzzin.live if we ever had the emergency that I couldn't make practice with my buzzer. The answer to that was no, the school's security protocols are probably wired to exclude anything that isn't .com/.net/.edu/.gov/.us, and as an outsider to their system, I can't push them to include a site. But for our purposes, it is pointless to do that since we can just use phones to connect in practice to buzzin.live.
This should allow us to survive snow season with a minimum of disruption.
The second item I needed to cover was a Christmas Break reading list. I pulled together five pages of webpages, infographics, podcasts, youtube videos, and social media threads that could be useful for high school quiz bowl. I tried to cover a variety of subjects and a variety of presentations, so different learners can work from different sections.
As I told the team: "Nothing in this list is required reading. I only ask one rule of people playing quiz bowl: that you seek out one new thing to learn that you find interesting every day. The reason I am packaging this set of links up is that I want you to have something that you can learn from during break. When we're doing classes, that's easy, you can learn from the classes you like. But when you're out of class for a while, you might miss out. So I'm just giving you something you can use as you want. If something doesn't thrill you one day, move on. But you want to develop a habit of learning so you're always growing, because that is how your play gets stronger."
The final copy of this is going to the team after this. I’ll post some of the selections for reading for next week.
The third thing I brought to the practice was the public packet distribution from NAQT. I had mentioned way back in Week 22 that publicly accessible distributions define the difference between the circuit and most other forms of quiz bowl competition, and a lot of what I've written in the book is based on deriving your own distribution from the evidence of previous episodes. What I haven't done here is really explain what you can do once you have a distribution, and how you can use it to plot your strategy. Because without a distribution to quantify what your team will face, you can't create a strategy at all.
So I presented a single-page, abbreviated version of the distribution, and these were the lessons I gave them:
- If you're looking at the numbers and find they don't end up with an integer for a category, that means there's some variability. 1.5 means half the time there's going to be 1 question, and half the time there's going to be two. This is what happens when you have to balance not just the big categories, but the small ones that aren't supposed to appear every time, or you have to balance within a category.
- The variability isn't really something to worry about. If there's 4.0 questions in a round that doesn't mean they'll be 8 in one round and zero in another. If you've ever read the rules of Monopoly you'd be familiar with the build-even principle, where you can't add a second house on Boardwalk before adding a first on Park Place. It has to be balanced as closely packet-to-packet as possible.
- What's more important is the subcategory that is 1.0 or greater. That means that at least one question in the round will always be there. Once you have reliability of 1.0, it ceases to be something you can ignore until later. 1.0 is a guarantee that if you don't know about a category, it will always hamper your team.
- Knowing the relative size of a category is important, because it allows you to both prioritize and emphasize which things to study. It's an obvious point with the data there, but you need the data to make the decision.
- The three big categories are Literature, History, and Science, and they are the three categories that correspond to courses you take in high school. (As I had mentioned before that means) These courses then have value to us, and when you don't have these courses, you're at a slight disadvantage, but we also know you'll go through those courses in the course of your studies.
- The fourth category to worry about is Geography, not because it's large, but because knowing geography helps with answering questions about history, literature, and some sciences. It strongly synergizes with most other categories, and the place of something can always be an answer in any category.
- For those categories that are less than 1.0, we have to employ a different strategy. Because there are subcategories with limited slots over the course of a tournament, there are also going to be a limited list of possible answers that can fit into those slots. And those slots are going to get filled with a few repeats. This means there's a few things you can guess once you understand what the category is, and what sort of information will be at the end of the question. So even if you know just a little bit about the category, you have a guess. In quiz bowl, for most of the time, it's much more valuable to know a little bit about a lot of subjects than to know a lot about a small number of subjects. You can teach that little bit first and come back to more in-depth knowledge later.
- But if you have something that isn't that big a category, there's a good chance that it doesn't have a lot of answers which appear at your level. Take that question on Nietzsche we just asked. (We had just had such a tossup.) There aren't a lot of answers for philosophy at this level. And if you could map out all those answers, and they're a small list, and you can just differentiate between those limited sets of answers, you're going to be able to collect points frequently in that small category. And those small categories add up.
- Most teams don't even look at the distribution, or they kind of take things as an equal set of slices of pie, like Trivial Pursuit wedges. So they don't prioritize what to attack. It's very easy to move from not prioritizing what to attack to not attacking anything at all. There are wide sections of the distribution which against most opponents can be described as undefended territory. It won’t require much effort of us to take that territory, but it will require some effort.
Over the past week, my social media feeds have seen a strong uptick in people discussing ChatGPT and how it applies to education. I've been thinking about this this week, and how it applies to quiz bowl, but I'm not nearly ready to write down all my thoughts. Hopefully next week I'll have something to say on the idea of AI writing, and how it could upset a few quiz bowl production models.