Blog: Expanded Universe

MIT Mystery Hunt 2025: the unsolicited reviews

preface

I usually thank writing teams in the conclusions of these posts, but now that I'm writing a series of posts before planning where to conclude them, let me say unequivocally: thank you, Death and Mayhem, for a wonderful hunt! Mystery Hunt is a weird event, which I know a ton of effort and sacrifice goes into every time, and it's kind of a miracle that it runs every year. I am excited to see what Cardinality prepares for us.

Where I do have negative things to say below, I hope it's clear they're all meant constructively and as my personal opinion, since puzzle tastes are highly subjective.

metas (the puzzles, finally!)

I actually didn't have a good time with the metapuzzles this hunt. I would attribute this more to me happening to be in the wrong place at the wrong time and experiencing an unusually rough slice of the metas, than to their overall quality; many of the metas I didn't work on much seemed cool. The overarching escape room concept of the Illegal Search round and the "black light" feeder reveal? Unquestionably neat. I have no idea how the second meta, Papa's Stash, works, but the people around me raved about "solving gender" for the rest of the hunt. For the Background Search, I helped with one step of Alias and got to see somebody point out something mindblowing on my own screen. For The Stakeout / Chinatown, I mostly helped out in the spreadsheet and got to use some rarely accessed skills of mine. I think the highlight was the Shell Corporations from the Paper Trail, which had structures that were innovative and story-thematic. Incredible!

The first meta that I didn't enjoy was the actual "metapuzzle" part of Papa's Bookcase (Under Blacklight) (spoilers): I think we got a hint for it, and when we figured out what to do, my reaction was, well, of all the ways one could have chosen for answers to correspond to books, this sure is one of them. You're telling me these are all movies... who have a lead actor... who has narrated a book from the bookshelf... and also many movies have more than one lead actor and you just have to pick the relevant one? It struck me as inelegant and thematically incohesive — tacked on in a "games that think more mechanics equals more fun" way. I would have been perfectly content corresponding the answers to books semantically just like in the first step. If not, I would have preferred something more puzzly, maybe wordplay-based, to contrast with the semantic first step.
The second meta I didn't enjoy was the Killer (spoilers): I found the Morse code in the titles and thought it was a cool step; they felt better camouflaged than I would have expected, and elegantly signposted all the remaining mechanics of the puzzle. The answers went in the blanks, and once we figured out how to group and reorder the pages, the Morse code would provide the answer. How hard could that be?

At first, I hoped there was a tight, puzzly rule for the grouping and reordering, but I didn't find one, and after staring at the pages a lot, I was seeing enough glimpses of continuity, last sentences of one page plausibly going into first sentences of another, that I was starting to believe that the puzzle was actually just about assembling narratives from those glimpses, as in Cain's Jawbone. I was also warming up to the idea. Well, it was wrong. There was a puzzly rule. That's fine with me, but I didn't like that the puzzly order seemed to actively clash with narrative coherence rather than being orthogonal. To continue solving, we had to destroy the coherence, however partial, that we had painstakingly assembled from the fragments in order to solve.

My biggest complaint about the meta is that the six categories are wildly asymmetric. The specificities vary from the extremely general "historical events" category to the two categories each about specific pieces of media. The clue natures felt imbalanced as well: one category with direct quotes, three with mostly semantically "straight" descriptions, and two with heavy reinterpretation of proper nouns as other things with the same name. This meant that after breaking in with one or two of the categories, we were totally misled about the general shape of what the other categories might be and how they might be clued, brainstorming in the wrong direction and putting things in spurious patterns. All this also made red herrings harder to eliminate. Things we had at various points:

Are these less compelling than the puzzle's intended matchups? I could be persuaded. But I want puzzle solutions to be tighter than "things I could be persuaded of", and the checking provided by the Morse code and the frame story are both just so weak and debatable.

I will also say that I had a worse time just because the puzzle's cultural references are almost all simply things I know nothing about, but I don't think this is complaint-worthy. The puzzle isn't for everyone, and that's fine. Unlike me, some people can recognize classical portraits from oblique descriptions and it's awesome that they get a puzzle to do that. However, it feels even a little worse than that description suggests when I can't even figure out what kind of identification help to solicit from the rest of the team.

keys

In this Mystery Hunt, teams unlocked puzzles with a custom currency of "keys" based on short descriptions, which I think is a neat idea. I agree with many other commentators that the descriptions are absolutely key to pulling off the choose-your-own-adventure vibe. And many of the stated goals, e.g., preventing unlocks from overwhelming smaller teams unless they opt in, seem great.

Still, my team is pretty big and top-10 competitive, and purely selfishly, I think I slightly prefer to not have the choice, to just have the puzzles unlock in some arbitrary order chosen by the organizers. Exploration is fun, but gets heavily diluted by the fact that each hunt team can only make decisions as a unit. If I'm playing an open-world game by myself, I get 100% of the agency; in a group with 50 people, I only get 2%. For me, any enjoyment of that 2% agency gets dwarfed by the organizational overhead and communication required for distributed decision-making. I don't feel as bad about using keys suboptimally in a way I regret, as I do about using keys suboptimally in a way that my teammates regret.

The preference is slight, though, so I really wouldn't mind seeing future hunts keep running with this unlock structure. I would also be curious to see if it works well in a smaller hunt, where each team member gets a bigger share of the vote.

the radio and the gala

I watched kickoff from our classroom and took a while to comprehend that D&M was giving every team a physical radio. Is this real? Like, a physical object, not (just) a virtual simulation of one? That's ridiculous! That's the coolest thing ever, the kind of thing you can only do at Mystery Hunt scale. I admired the radio from afar when our team members brought it back from kickoff, and then... proceeded to almost never interact with it during the hunt. I remember hearing a solve sound play through the radio early on, and that might have been it. Oops.

Some possible reasons for this:

My feelings and interactions with the gala / open HQ were also very similar. I thought it was great to have a place to freely interact with the hunt runners and even people from other teams, and then I went a grand total of once over the entire hunt, for some kind of puzzle interaction involving taking photos. Our classrooms weren't even that far from Stata.

I don't think D&M could have done these things any better; this is mostly just a reminder to myself to set some time aside for experiencing the cool innovative parts of puzzlehunts instead of puzzlemaxxing the whole time, especially when we're not trying to be super-competitive.