2019 MIT Mystery Hunt, Part 1: Structure and Story

(This is a recap/review of the 2019 MIT Mystery Hunt, which happened this month. Puzzles and solutions can currently be found here. These posts will contain many spoilers.)

How did we get to Molasses Awareness Day? As TK Focht noted at wrap-up, some members of our team (okay, mostly Scott Purdy) have been advocating for a molasses-flood-related theme for years, to the point where a lot of Setec was saying we should wait to try to win in 2019. As it turns out, we got the itch earlier than that. Then we forgot to retire. I seriously did not think we were going to win in 2018.

That said, “molasses flood” itself is not the sort of theme that can sustain 100+ puzzles. (We could have woven it into a Willy Wonka theme, but we rejected Willy Wonka as a theme in 2017 due to at least one team member’s fear of Oompa-Loompas, with me warning Setec that if we didn’t use Wonka as a theme for Hunt, I’d use it for something else. And I did!) So we considered a wide array of other themes with varying levels of molasses content. I think we ended up with about 20 options on our first team-wide preference vote, which occurred before we brought a bunch of new members (I’ll say much more about those new members later on; they saved the Hunt this year and they deserve truckloads of gratitude), which narrowed the field to four clear finalists. I’ll call them A, B, C, and D, without revealing their identities, so that we can repurpose one of them when we run Hunt again in 2083.

One thing that became clear early in the process was that the team was interested in messing with a traditional meta-round structure, likely by putting a bunch of answers in one pot and requiring teams to “sort out” which answers fed into which metas; this is a thing that isn’t obviously solvable, but Foggy Brume (not on Setec) has proven in several Puzzle Boats that it can be done, so we were intrigued. All of A, B, C, and D had some degree of this. C and D were themes I was not jazzed about, so I will not expound further on them. B was my idea and had a multi-phase theme that used “sortable” metas in only one phase, which I figured would be easier to build; I really liked the structure we came up with and was endorsing this as my top choice.

Theme A is the theme you saw this year. It evolved from a TNBC holiday misunderstanding theme that Brian Tivol posed in 2005 (not a typo) when Setec wrote Normalville. He threw it in again this year, and I had the idea that if we wanted sortable metas, we could use a graph in which holidays were vertices but metas were edges, so that teams would have to figure out which puzzles were escaping into which other towns. This also seemed to lend itself well to a molasses flood as an inciting incident; a flood would explain why the walls between towns had been newly breached.

After welcoming our new teammates, we spent some time developing all four finalists in Slack channels and then voted on the final proposals (ranking all four). Theme A won according to pretty much any voting system you could choose to apply to the ranked lists, and I understand it was 1st or 2nd on almost every ranking. My first choice of Theme B, also featuring a molasses motivation, was apparently frequently ranked 1st or 4th. Hashtag polarizing.

It was around this time when I realized I’d shot myself in the foot with my meta structure suggestion for Holidays. In 2017*, I had gotten very frustrated with the pace of meta construction, peaking with me flipping out at Tanis O’Connor (who wasn’t even in charge of metas either, sorry, Tanis) in the middle of a trip to Europe. So when we won again, I thought the best way to keep things running at a pace I was comfortable would be to take charge of metapuzzles.

[*I realize that by “in 2017,” I actually mean “in 2016 for the 2017 Hunt,” but if I worry about that level of detail I’m never going to finish these posts, so any years refer to the associated Hunt.]

But in 2017, we generated a whole bunch of character and quest metas, loaded them into an online system where people could test them on their own time, and eventually picked the ones we liked most. We couldn’t just pick the best ones in this case; we had to pick metas that actually fit together into a graph, since if a holiday only showed up once, its sorting would be trivial. We initially intended to have none of these, but we eventually decided it would be good to have one degree-1 holiday as an “orientation” round, and given our source material, Christmas-Halloween seemed like the natural pair to reveal first.

So anyway, finding a way to build a map full of metas was now my problem, and I have lots to say about it, but that probably fits more into a post about metapuzzles (likely coming next after this one). So let me say a little about story.

Those who have written Hunts with me before will not be surprised to hear that I wrote the kickoff script. I’ve done that for seven out of eight Hunts I’ve worked on, and I think I have a good sense of the sort of wacky self-referential humor that functions well in this setting. I think the audience wants larger-than-life characters that help them enter the world of the Hunt, but they also know that they’re looking at a bunch of puzzle constructors who are puzzle hunters just like they are, not professional actors. I usually know which lines are going to get laughs based on Hunt content or nerd content, and to those of you who came to kickoff, you did not disappoint. Particularly when you laughed at the “pigeonhole principle” line, which helped me win an argument with my wife.

We initially had 26-100 for kickoff due to lack of Kresge, but then Kresge became available when another event canceled. I was actually a little bit opposed to moving, since I thought the bigger fancier space might raise people’s expectations in terms of props/costumes. Several teammates reassured me that Setec has a reputation for low production values (uh, do we?) and I think the finished product was quite good for something we didn’t rehearse in person until the day before. The whole cast was phenomenal, but I want to give an extra shout-out to Steve Peters and Jesse Gelles, who played Santa and Jack respectively and arguably had the heaviest lifting to do character-wise. They showed up for rehearsal with their characters pretty much nailed, and it was great to be able to direct them around landing jokes rather than getting their delivery right in the first place.

It was very unusual for me not to act in the kickoff, but I felt like I’d had enough of a turn in recent years, and lots of people wanted to participate (the part I was most likely to cast myself in, believe it or not, was the Valentine emissary played adorably by Margot Pliska; that character used to be very different, and Margot was a major upgrade). I did throw myself in as the organizational person at the end, where my most important task was to tell people the website URL and keep them busy until 1pm… which, of course, I forgot to do. YOU HAD ONE JOB, DAN.

Those who have written Hunts with me before will likely be more surprised to hear that I did not write the scripts for midgame (the Mayor of Your Birthday Town) or endgame. Those were written by Justin Werfel, who also played the April Fool and did an incredible job with both of those tasks. The evolution of the April Fool as a character was really interesting, in that they were initially conceived as a fairly stereotypical villain, but as we wrote the Hunt, a lot of us noted that we sympathized with the Fool, since they were really only doing what they’re supposed to do. One of my favorite ideas to heighten this sympathy got swept slightly under the rug by the art design once we decided to simplify and represent each town by a single building, with the April Fool living in a rundown shack. In an early conception, every town would have been a collection of buildings, but April Fool’s Day Town would still have been that single little shack with no one else living anywhere nearby.

This is actually the second Setec Hunt in a row where I saw a fairly moving emotional arc develop itself under the story elements, and when we noticed it we committed to making it clearer. In 2017, we designed the midgame and endgame around combining the six characters’ skills, since that’s sort of what you do in a “meta-meta”… This encouraged me to heighten the bickering between MIT characters (and between RPG characters) in the kickoff, and what we ended up with was a fairly sweet story about friends putting aside their differences and working together to save another friend. In 2019, we wrote a whole lot of Hunt before realizing that by making the goal to solve these interholiday problems caused by walls breaking down, maybe our Hunt was a bit more pro-wall than we’d intended? (Ggggggghhhhhh.) We had a really good conversation at our east coast retreat where we conceived the idea that having some interaction between towns, and letting the April Fool do what they do, were both good things in moderation. Justin wrote this really nicely into the endgame skit (which you can see in the video shown at wrap-up), and he and Jesse performed this moment really well. I’m not crying, you’re crying.

I think that sums up most of my thoughts on the skeleton of this thing. next up, I’ll reveal more about how we made the metas happen (which can hopefully help future construction teams decide what to do, what not to do, or maybe both).

3 thoughts on “2019 MIT Mystery Hunt, Part 1: Structure and Story

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s