MIT Mystery Hunt 2025 ~ A Noir-ther great hunt

This is the kind of puzzle I look for in the opening set of puzzles. Two lists of clues, matching them up, and a nice thematic flair to boot. Smooth and clean and personally my favorite type of easy puzzle.

Not surprisingly, I jumped next to another word puzzle. This solving path was another time-honored tradition of breaking into the aha by offering a completely wrong idea. I proposed that “win over” was CONVERT and “reformed believer” was also CONVERT, but the other pronunciation. This was not the puzzle, but it did lead immediately to other people finding the true “aha” here, which is equally as good and obviously more thematic. It’s hilarious to me though that I managed to propose a completely wrong solution, but because it also started with “CON” it lead to a real aha.

How can I pass up a Pokemon puzzle! I figured early on it had to do with the name changes during evolution—but it was satisfyingly tricky to nail down exactly how it was working. And to be fair, I figured that had to be the mechanic but it was just too good of an idea once the puzzle put it in my head. That said, I’d love to see a puzzle that actually uses the Battle Factory format—so good!

This was a fun enough idea—turning a connect the dots into two different pictures is not too crazy of an idea, but it’s neat to see happen. But this was a time where we were stuck with the pictures for a while. Luckily it was only 15 minutes, but I question the extraction choice to clue a frankly green-paint answer with two picture clues.

This was fun as a chill parallelizable puzzle: everyone could focus on their own rebuses, and it was nice to evolve from “oh! Weird Al songs!” to “there must be something else…” to really getting the puzzle. I also think Bob’s Burger’s is a great dataset I’m always happy to see used.

This was a fun word puzzle—like a P&A magazine in a good way—but man this made me really want to sell util.in as a great solving tool. Util.in’s word search’s ability to highlight words from a list made it super easy to see the final picture—when I’m sure it would have taken longer to see them in our spreadsheet where we just highlighted both sets of words at once. Util.in supremacy—down with nutrimatic! Of course, the final answer was also a little silly in actually getting. Another downfall of picture-based extractions, although in this case it was not because the picture was uninterpretable.

We had trouble keeping our radio on in our HQ the whole weekend where it would be audible while not drown out the room of conversation. I think we encountered the first ad later than most teams probably, but it was hilarious to start hearing the sponsors and realize “hey…these look like puzzles”. We first categorized them as a separate round, before unlocking them as a Stray Lead, then getting them in the actual Paper Trail round. Regardless, this experience really gave a nice “Mystery” aspect that I look forward to in “Mystery” hunts, where we learn to expect the unexpected and things appear that you can’t just cleanly identify at first. Very fun. The puzzles themselves were also satisfyingly tricky, and getting to go to the Gala and getting the “martini” was hilarious—I inadvertently just asked for a martini at first and got a “wtf are you asking for, this isn’t an actual bar?” response from the bartender, which was hilarious. This puzzle was one of the highlights of my hunt.

I only stepped into this puzzle after the cryptic was solved, to try and save it from extraction hell (this was looking to be the first puzzle we really got stuck on!) We deduced that nobody was the killer, and theorized to all hell what this could imply about the actual puzzle, until we just decided to extract for each person assuming they were the killer. I realize that maybe this is the most indirect justification, but I insist that calling Alec Clifton the “housemate’s killer” and “Pat’s neighbor” implies that Pat isn’t the killer—housemates and neighbors are clearly disjoint properties to me. That said, I think if you just changed them both to “clubmate” the logic would be the same as the solution, and it would work. I was tempted to send in an email post-solve to propose an errata, but didn’t end up bothering.

Anyway I still think this puzzle is cool—the idea of hiding information in cryptic clues came up multiple times during hunt and this was a neat execution of it

I was excited to jump on this puzzle after hearing that it was about Adam Ruins Everything, but I’m sorry to say this might of been one of my personal clunkers of the hunt. It felt fairly unmotivated during solving to connect the two disparate things of Adam Ruins Everything and UNESCO heritage sites, and we ended up in spreadsheet hell with columns upon columns of random data that might be useful. We never found the intermediate phrases, but luckily someone sorted by the right column and found the answer in the end. I feel like it would have been cool to have a similar puzzle that just combined different Adam Ruins Everything episodes—that was what was drawing people to this puzzle anyway.

It was very fun to hear Adam come over the radio afterwards though. We, uh, didn’t catch what he said and immediately got worried how we would solve it until we saw the email come through. This was a fun moment though—a properly placed “oh no” moment is very fun!

Hands down one of my favorite puzzles of the hunt. This is such a unique solving path—backsolving an entry through cryptic clues?!? Once we got the first READ PROUST, it was hilarious to see how this was getting used in every clue, and then once we found the New Yorker article, everything made sense in one perfectly silly aha moment. I just can’t get over how unique this type of cryptic solve was, where we slowly get info about an entry but only in a semantic sense or through weirdly specific syntactic constraints. There were a lot of great cryptics in this hunt, but this one was my favorite.

Yet another fun word puzzle. This gave the feeling of constructing a puzzle, rather than solving a puzzle, which from experiencing is just a different form of solving. It was really neat to go from feeling like it was impossible to get started, to slowly making progress, to really cruising and sailing into the answer at a nice clip. Solid puzzle all around

Right after hunt, I declared that this was my favorite puzzle of hunt that I did, and I might as well stick to my guns. This was super fun, and a great example of “show don’t tell” in puzzles. This puzzle neatly guided you through its mechanics and got you going with a basic dropquote or two, before revealing itself in a very natural way. And the final two grids felt like a well-made video game boss that challenges you in the skills you’ve learned upon the way. This puzzle just had a natural flow to it, while being a devilishly-tricky-but-satisfying solve and just a fun idea from the start.

In this puzzle, I proudly declared that it would be cool for the puzzle to do X, but that it would be absolutely impossible to construct such a puzzle, so it’s clearly something else. Of course, the puzzle was actually that.

This is a puzzle that seems impossible to construct, and yet it managed to be solvable. Of course, the clues were tricky and hilariously silly because of the constraints, but it still managed to work itself out. And unlike some other puzzles that over-constrained themselves, it was able to show you early on that “no, I really mean this” because we could check the lengths and see what we feared: each answer set was also 100 letters. Impressive stuff

A fun puzzle where guessing puns is the main identification, which manages to preclude most solving tools one would use—you really just have to think about it! I’m terrible at this of course, but I was happy to find the aha after just googling what other people were typing into our sheet. Fun puzzle and leading to an interaction was only natural—it really elevates the puzzle above what it would be otherwise.

Great little cryptic. It was satisfying to throw the full might of our team’s cryptic solvers at this and absolutely annihilate the puzzle, but it was a fun idea and was fun to solve. Sometimes the best compliment is that I don’t have much to say about a puzzle besides I liked it!

This was a good example of a puzzle where you account for many possibilities, but as you solve the puzzle it constantly works out nicer than you imagined. At first I thought it would be like a cryptogram, before realizing that every color was standing for an appropriate letter or two (R for Red), and that it was all Macbeth quotes, etc. The final extraction was a little tricky to read exactly, but it didn’t cause too much of a problem.

It was also funny that you just had to assume this would use the Macbeth Color Checker (which was used in an early 2020 MITMH puzzle), but I’m glad they didn’t feel the need to force it in.

There were a good few puzzles this hunt that basically were of the form: “hey let me show you this cool code/cipher”, of which this was a clear example. It only works in a puzzlehunt-style puzzle if you can execute on it in a cohesive, thematic way, and this puzzle managed to do that. (Another example I saw was A Weathered Note).
That said, the initial step of transcribing numbers from 25 videos was thematic I guess, but a pretty laborious gruntwork step. Hey, at least my teammate did it and not me.

I think the Background Check metas were the first we unlocked in the “main round” of hunt, but they laid unsolved for a while, even after people found the relevant background info. Partially, I think it was because these were quite hard and needed most/all of the feeders. I ended up attaching myself to The Mark (poker chip) meta, which once I was able to wrap my head around, I found extremely difficult and basically needing every answer to do. Even once we had all the answers, it was VERY tricky to actually jigsaw them together into the final grid to solve the puzzle. I was convinced for a long time that it was impossible before my teammate managed to figure it out.

Of course, I would be remiss to point out that we were blocked for a while because we had inadvertently mixed up our metamatching for these metas. I was trying to shortcircuit the poker chip placing with the wrong set of answers and driving myself crazy in the process. It didn’t help that another group was convinced their answers were right because they found ways to change several answers into nice 9-letter answers for The Oversight (foreshadowing…).

Anyway this was actually fine and didn’t really detract from the experience—it made it quite the hard solve though! And then upon getting The Alias and (relatively) quickly realizing what needed to be done, we yet again went back and tried to do a hexagonal jigsaw that we convinced ourselves was impossible, until we managed to do it. We did end up needing all the answers both times though—I’m not sure if this would have been possible to short-circuit. Formula Won ended up being one of the (refreshingly few) puzzles that remained stuck the whole weekend, and we ended up using a Key on it in order to just get all the feeder answers here

The Alias metameta is just very fun though—a real impressive construction from start to finish.

I only really worked on the first Shell Corporation before going to bed on Saturday night, but this was such a cool and thematic way to do a round. When we started looking at the Shell Game, it was a challenge to even wrap our heads around what it was even saying. But we managed to slowly unravel what was going on, and it is just beautifully thematic. Maybe one of the most thematic meta solutions I’ve seen. The crazy shell-corporation graph and having the “hidden company” node is just a great metameta. It was made even more fun by being the last meta we solved before doing the endgame, so a large number of people were just staring at what was an increasingly confusing mess on our whiteboard.

This was my favorite supermeta in the hunt, and one of my favorite of all time. That said, I do think the supermeta structure was the cool, satisfying end of this round, and disentangling the shell corporation was also extremely cool and a unique solving experience, but I think it’s only cool if you also get to do the super meta too. Basically, I slightly wish the shell corporation metas were a bit easier just to prioritize the neat supermeta which uses the shell corporations. I also think there could have had a little more thematic flavor text for some of them, and the cryptic shell corporation was brutally hard. I’m not complaining too much here, but I worry that teams that got started with the shells and didn’t actually get to finish were left a little disappointed. Overall though, this was a clearly highlight of hunt.