Post to be Named Later: Writing the 2024 Mystery Hunt

I’m Brian Widerman and I spent 2023 writing the 2024 MIT Mystery Hunt. I served as the fact checking lead and Printing Czar during the construction effort. Given that I spent a year of my life working on this thing, I wanted to take time and write down what I remember about the pieces of the construction process that I was involved with. It’s taken me until June to get this written because that’s how long it takes to recover from running Mystery Hunt.

A few quick notes: I don’t speak on behalf of the team as a whole. The thoughts and opinions expressed in this post are mine alone. Also, there are major spoilers for the following puzzles in this entry: Charon, Good Company, Luxor, Badges Badges Badges..., Revolting Developments, Oil Paintings, This Space Intentionally Left… Well You Know, Spoiler Alert, and Transylvanian Math.

A quick explanation of Mystery Hunt for my friends and family who might not know: The MIT Mystery Hunt is a puzzle hunt that takes place in Cambridge, Massachusetts on the MIT campus every January on MLK Day weekend. The event starts on Friday at noon and doesn’t stop until a team wins. Throughout the weekend, teams made up of very smart people solve tons of puzzles that are hosted on the event website.

I most commonly describe the Hunt as being kind of like an escape room, but one that lasts for 60+ consecutive hours and where figuring out the combination to locks can require you to be familiar with intermediate concepts in astrophysics.

A Mystery Hunt puzzle can be anything: a crossword where all the clues are just emojis, a video with hidden messages, a physical stack of shredded paintings that need to be reassembled in a very specific way, etc. The answer to a puzzle is almost always a word or phrase. The Hunt as a whole is made up of several rounds, each of which has a “metapuzzle,” which uses the answers from the other puzzles in that round to form its own answer. The goal of the hunt is to solve all of the metapuzzles and then find a coin somewhere on campus. The prize for winning is a souvenir copy of that coin. Oh, and also the responsibility for writing the next year’s Hunt.

I know that this doesn’t sound like fun to most people based on the looks that I get when I talk about it. But we had two thousand people in-person in Cambridge this January and many more playing remotely, so it’s definitely clicking with a lot of people. I think the nonstop nature of the event gives people the same sense of accomplishment as participating in endurance sports, like some kind of mental ultra-marathon. That, plus the feeling of extracting an answer after staring at a cryptic puzzle page for hours makes you feel really smart. Participating in Mystery Hunt is unlike anything else I’ve ever done and while I can’t figure out how to convince my friends that this is a fun event, I understand why people keep putting themselves through this.

This is the largest annual event in the puzzling community and because of that, there’s an expectation that the team who writes each year’s Hunt will put a lot of effort into writing enough high-quality puzzles to keep the participating teams solving for the weekend. Producing a Hunt is an objectively unreasonable amount of work: It requires a team to write, edit, fact check, test, and create website versions of a large number of puzzles, plan 4-5 puzzly events, coordinate logistics (including working with MIT on a ton of things to make sure the event can be run safely and responsibly, negotiating hotel room blocks, booking event spaces, reserving classrooms and assigning them to teams, negotiating A/V for kickoff and wrap-up events, etc…), fundraise around $50,000, produce physical puzzles, design and manufacture a coin and merchandise, develop a website and create custom art for each of over a dozen rounds… and to do all of that for zero compensation, in 12 months, and by the fraction of the team that doesn’t disappear from Discord by February. A Mystery Hunt takes thousands of hours of people’s time to produce a weekend of fun for the community and I can’t think of anything else in the world that is such a lopsided amount of effort for such a relatively small payoff. People do it because they love it.

For me personally, this was a particularly bad year to be on a Hunt writing team. I had just started a company with a friend in November 2022 and when I sheepishly texted him a photo of me holding the coin in January, he responded with “So I guess you’re quitting then?” I knew my primary responsibility was to get the company off the ground, but I knew I’d regret missing the opportunity to be on the running team for Mystery Hunt, so I convinced myself that I could do both, thus setting myself up for the best/worst/most fun/most stressful year of my life.

January

When we were dividing up admin responsibilities, I listed fact checking as my first preference because I previously worked on fact checking for the 2018 Caltech Puzzle Hunt and found it fun then, plus I like to think I’m fairly good at tasks that require attention to detail. I also stupidly assumed that being the fact checking admin would take less time than some of the other admin roles.

January started with a few days of mandatory rest for everyone. Team leadership made it clear that winning didn’t mean an obligation to write and the break gave people time to decide whether they were truly committed to a year of writing. We had almost everyone on our winning team sign up to write and we quickly got into theme selection.

One funny note about theme selection: TTBNL has tried to put together two follow-ups to the Caltech Puzzle Hunt that part of our team wrote in 2018. We chose a Pandemic board game theme in late 2019 and had to stop writing around March 2020 when our theme got a little too real. We later started a hunt with the theme of “it's an unfinished hunt” and for various reasons, it became an unfinished hunt. With the realization that whatever theme we chose was going to manifest itself as reality, there was a lot of pressure to pick something with manageable consequences, like the gods coming to our world and sending all of humanity to the underworld in one go.

I made the mistake of getting really invested in my own theme submission, which made it past the first round of voting but just barely missed the cut in round 2. Round 3 was between two very popular theme choices and the vote could not have been closer. If I ever get to write again, I hope I’ll get another shot at my theme idea.

Charon

The first writing task was the metapuzzles. I don’t have a lot to say here - my involvement during the meta writing phase was mostly as a fact checker and some idea development, though Bella listened to my idea for a River Lethe meta and very generously brought me on board with something similar she was working on. Bella, RJ, and I worked through a ton of different ideas involving rivers, boats, and wordplay and eventually settled on what became the Charon meta. It was fun brainstorming the mechanics of this puzzle with Bella and RJ and I was happy that my pun answer suggestion made it into the final puzzle. Adalbert got involved to help finalize the structure of the meta and to generate answers to make our puzzle answer work through multiple revisions. I’m proud to be associated with the final product.

I also wrote a Zeus meta that was going to be set at the Rock and Roll Hall of Fame in Cleveland. The puzzle used Springsteen lyrics and I was pretty happy with the answer I wrote - there’s a rock group on stage but the audience can’t hear them. What can be done? COVER BAND WITH WALL OF SOUND AND HIT RECORD. Though I liked the meta that I built and it passed testing, this was one of the rounds that got cut early on. It worked out well in the end, though, because the puzzle mechanics were kind of awkward in a meta but they worked great as a regular puzzle. More on that later.

Metas were mostly finalized in late spring (I think a few got swapped in and out throughout the year), though it took a little longer than expected and we got to a place where we desperately needed to finalize the metas in order to move forward with the rest of the Hunt development. Ursula, our test solving lead who is great at organizing people to do not-fun things in very fun ways, came up with the idea of a Discord event called “10 METAS IN 10 DAYS,” which aimed to graduate the final 10 metas as quickly as possible. Ursula and I had a great time overhyping the event, always obnoxiously written in all-caps and bold, to try to recruit testers. And it worked! We were able to fast track fact checking and test solving sessions to pass most of the final 10 metas and it was one of several badly needed “getting things back on schedule” moments throughout the year.

Fact Checking

I guess this would be a good place to write up my fact checker spiel.

TTBNL’s puzzle writing workflow put a special emphasis on fact checking, which made my job harder but I think led to better puzzles overall. When an author pitched a puzzle, the Editors-in-chief would review the idea and give it a thumbs-up as long as it felt like it could be a satisfying puzzle and it didn’t conflict with an existing puzzle’s subject matter or seem impossible to write. The author would write a first draft with a complete solution document and once the editors had approved the draft, it would proceed to fact checking. At that point, a member of the fact checking team would claim it in our puzzle software and start evaluating the draft. The instructions I gave them were to pull up the puzzle’s solution document and step through everything as if they were a solver. Fact checkers made sure that every component that was supposed to be in the puzzle was actually there, that every clue was accurate and unambiguous, and that there were no major accessibility blockers.

Puzzles could not enter testing without a fully fact checked draft. We did this because we had a limited number of test solvers to work with and we didn’t want any single puzzle to burn through all available testers. With Mystery Hunt puzzles, solvers are often looking for things that are “off” as an initial way to break into a puzzle and so something as minor as an accidental extra period in a clue or a missed font change on part of a word might send solvers off on an unintended tangent and sink a test solve session before it even had a chance to get started. (In fact, the font choices on Luxor caused so many testing rabbit holes that it inspired This Space Intentionally Left... Well, You Know) I think it helped our testers to know that someone had performed a QC check on every puzzle draft before they saw it so they could have some confidence that the puzzle was probably not broken.

Fact checking some puzzles was easier than others. One that sticks out to me is when I lost my mind while checking a puzzle that required me to watch Too Many Cooks several dozen times. Possibly even worse was The Dan Miller’s Sesame Street puzzle that introduced me to Big Bird’s song “I Just Adore 4,” which still pops up randomly in my brain at inconvenient times and for no good reason. Also, the fact checks for each draft of Oil Paintings required insane attention to detail and there were A LOT of drafts. I can’t believe that puzzle worked as well as it did.

My team aimed for a 24 hour turnaround time on each draft’s fact check and while we didn’t achieve that 100% of the time, I’m proud at how often we did hit that target. It’s even more impressive considering that fact checkers often stepped in as secondary editors when a fact check went particularly awry. The people who worked on the fact checking team were extremely dedicated to producing quality puzzles and I think that shows in our final product. I don’t have an actual count, but I feel like the number of errata in the 2024 Mystery Hunt is exceptionally low. Say whatever you want about our Hunt (and you have and I’ve read it all), but I feel like we had a really well-edited and fact checked set of puzzles.

Good Company

I’m going to put a lot of spoilers about Good Company in the upcoming paragraphs. It ended up being placed in the second Nashville round, so hardly anyone got to see it during Hunt, which might be a good thing based on the feedback it got. I spent a lot of time making it, so please check it out! I recommend it to anyone who enjoys dumb puns, which I assume is everyone.

I started working on this puzzle (titled “Local Businesses” until a week before Hunt) in June and it took me two months to get the first draft done. I’ve always liked Bob’s Burgers and I still watch it pretty regularly when I need something light and funny. I also love bad puns, so the Stores Next Door from each episode’s opening sequence seemed like a great data set for my first ever Mystery Hunt puzzle.

Audio puzzles are some of my favorites at Mystery Hunt and I really liked the idea of recording radio ads for the fake stores from the show. In my head, the idea was that the same guy keeps opening an oddly specific store next to the burger shop each week and can’t understand why people aren’t buying the weird products that he’s selling. So I put together a chain of stores that had letters that I wanted to spell out a clue phrase in the right positions, recorded ads where every ad referenced the previous store with a bad pun, and then put the ads in episode order to help with identification.

In testing, I was surprised at how few testers had even heard of Bob’s Burgers, a show that I thought was pretty popular. I was also surprised at how difficult it turned out to be to get people to use the phone numbers to index into the text of the ads themselves. I thought my flavor text of "he needs to dial back his rhetoric" was pretty good, but even if people thought to use the phone numbers to extract from the ad text, they often wanted to pull out letters instead of words. I guess there are certain kinds of indexing that are more intuitive than others and it was difficult to overcome that.

I was frustrated when the first testing session turned out to be a disaster. The way I saw it, I had written an easy puzzle and I was upset that the editors had made me add extra clues that I thought made it even easier. The testing session ended after a few days of hinting the solvers through every step of the puzzle and in the end, I got some very thoughtful, honest feedback.

I learned some valuable lessons from that test solving instance. I was initially hesitant to add more clues that I thought gave away the puzzle too much, but I learned that it’s very easy, as an author, to write an under-clued puzzle while imagining that it’s way too obvious. Writing good flavor text and adding checkpoints to help the solver discover what you want them to find is very important and luckily I had great testers and editors who were willing to work with me to help me understand what changes would achieve that. Thank you to the testers and especially to Paul, who worked closely with me to get some additional pieces in place that got the final draft through testing.

Both test solving sessions on my final draft gave the puzzle high fun scores, but the only feedback I’ve seen on this puzzle from Hunt (both in the survey responses and in a few blog posts) has been pretty negative. I had a lot of fun writing it and I hope there’s someone out there who enjoyed solving it.

As a side note, Jessica Li and I spent all of Hunt weekend talking excitedly about the moment where someone was going to solve this puzzle and bring us a burger. I was pretty sad when Hunt was over and we hadn’t received any food. I do want to thank Cardinality, though, for being the only team I’m aware of who forward solved Good Company during Hunt. It made me very happy to see the correct answer notification come through on Sunday night.

Also, some products that got cut when their stores didn’t make the final draft: The Simone de Beauvoirange Marmalade from “I Think Therefore I Jam” and Rosemary's Baby Powder, Nyquil on Elm Street, Rubber Chucky, and the Exorcistern from “Bloodbath and Beyond.”

Retreat

Puzzles trickled in throughout the summer but we started feeling a little panicked around July when it was becoming very clear that we were very behind schedule. Luckily, we had planned to host in-person retreats across the country the first weekend of August and I think that was the smartest thing we did all year.

I joined the New York retreat, generously hosted by Pacho at his home, and met a few team members for the first time. The first day of the retreat, I happened to be sitting next to Alina and John when Alina mentioned her idea for a Top Chef puzzle. She educated us about a show that neither of us had seen and we quickly started working on what became Luxor. The three of us got along very well and were so excited to support each others’ puzzle ideas that we became frequent collaborators for the rest of the year. Over the retreat weekend, I worked with Alina on Dear Diary and Deep, Deep Trouble and the whole room tried to keep up with John as he worked through Where on Earth Is Carmina Suntigre. (I really only contributed one clue - and my photo - to that last one, but it was a fun collaboration in the room throughout and TTBNL was very generous with author credits)

We held retreats in New York, Austin, and San Francisco, and they generated a ton of ideas that got the Hunt effort back on track. If any future Hunt writing teams are reading this and want advice for the best thing you can do to help write puzzles quickly - get a bunch of people in the same room for a whole weekend and get them talking about their ideas. A lot of the best stuff in our Hunt came out of retreat and I think a few of the retreat attendees set up recurring follow-on retreats for the rest of the year. I know there are groups in a couple cities that are still getting together regularly months after Hunt ended.

Badges Badges Badges Badges Badges Badges Badges Badges Badges

The badge puzzle was the one big idea that I brought to retreat with me. I knew MIT campus security was requiring us to provide name badges for the event and one of my earliest Hunt proposals was to make a puzzle out of them. Before I started participating in Mystery Hunt in 2018, MLK weekend was dedicated to Shmoocon, a DC-area security conference. I had no business attending, but they had a puzzle contest and I went almost exclusively for that. I especially liked how every year, the participant badges were designed for the puzzle contest. For Mystery Hunt, I wanted to do something high-concept and complicated (one of my pitches was something like, "what if the badge was actually programmed like a hotel room key and it could unlock certain geocache-like boxes hidden around campus!? And what if there were several distinct badge styles, but each team only received copies of one of them and they had to trade between themselves to get all the badge types!?"). The editors were very patient in talking me down from all of my bad, complex ideas.

I’m extremely grateful that Alina was there to help me simplify my ideas. At retreat, I showed her a bunch of ideas, one of which was to have 9 different badge designs that formed a 3x3 grid. She was instantly able to see that that was a good concept and she helped to come up with the dwarf planet theming and to distill my insane ideas into the “mini-puzzles that lead to extracting a word from a poster” puzzle that the badges puzzle became. She came up with a bunch of the mini-puzzles and the two of us got to learn Canva to put together the badge image files.

I had never heard of Canva before working on this puzzle. I’m very happy that it exists. I have zero art skills and the thought of trying to pull this puzzle off with zero artistic ability was so depressing. Here’s an example of what the posters might have looked like if I hadn’t been introduced to Canva:

Thanks, Canva!

Originally, the final extraction was going be a straightforward clue phrase revealed by the nine posters (I think the pitch was something like “1998 movie BLANK man in the iron BLANK,” which the editors correctly threw out immediately) before John came up with the genius, very fun idea of mondegreen-style extraction. The pronunciation guides on the back of each badge were intended to hint at that method, but it also let us hide a fun TTBNL/Plugh Easter egg in the staff badges. Our first attempt at that clue phrase was “gym carry move Ian wit cheat earn screen” and then we realized there are, like, three movies where Jim Carrey plays a green character…

I’m very happy with how the badges puzzle came out. Like Good Company, I went into this wanting to make a puzzle difficult enough to be worthy of Mystery Hunt. However, unlike Good Company, I got to work with really smart co-authors who helped me sand down my original idea to make something more accessible. I guess a lesson that I learned - and I didn’t fully understand it until Hunt weekend or maybe even until after I read Jen’s blog post - is that when you write puzzles for Mystery Hunt, you have to remember that not everyone who tries to solve your puzzle is going to be an experienced solver. Especially if you’re writing something for the earlier rounds, it’s not a bad thing to write something that’s accessible and easy to break into. I’m pretty proud that Badges worked as a nice intro puzzle to people who are brand new to puzzling at the 2024 Mystery Hunt because it contains a lot of the basic puzzly ciphers that end up in more complex puzzles. One of the eight posters was right outside TTBNL HQ and, like Jen, I enjoyed spying on teams looking for the poster throughout the weekend and loved seeing that people were still having fun with Badges into Sunday afternoon.

I’m also really happy at how enthusiastically everyone on TTBNL embraced the “game” of people needing to discover our staff badges as the key to solving the puzzle.  It was an unconventional puzzle element that I hadn’t seen done before and there were concerns about whether it would work. I hope we did a good job of getting our badges in front of as many teams as possible in the hallways and at the events. We definitely had fun visiting teams and trying to hint at them to look at our badges. I would like to officially apologize to the team that we visited and had Artemis ask them if they’d seen their lost badger, causing the team in the classroom to urgently ask their teammates on Discord to send multiple photos of badgers.

In the first few minutes after Hunt “started” and the server wasn’t allowing anyone to access any of the puzzles, people on the team started joking with me that I had sabotaged the server so that every team would be forced to focus entirely on the only puzzle they had available - the badges. No comment.

I selfishly hope that Alina, John, and I have started a trend. I want to see badge puzzles at every Mystery Hunt from now on.

Revolting Developments

As I mentioned earlier, I had written a Cleveland meta that used Springsteen lyrics, but the round got cut. When I was looking for ways to repurpose the meta mechanic (offsets from lyrics within specific songs), I tried to do something with band member birthdays but the editors said that was too close to another meta. But in thinking about using the calendar for offsets, I realized that there was a band I knew a lot about who had interesting lyrics and a name that kind of let the puzzle write itself: The Decemberists.

As I wrote in the author’s note on Revolving Developments, the Decemberists are one of my favorite bands. I’ve seen them live eight times (including just a few weeks ago in DC) and to me, several of their songs are strongly associated with significant moments in my life. Finding an excuse to add them to the Mystery Hunt canon felt pretty good.

In early fall, TTBNL set up a series of virtual retreats over several weekends, asking people to set aside a few hours per weekend to work on each others’ puzzle ideas. People showed up for three hour blocks of time and I showed up at an October retreat with a very rough idea (calendar dates as offsets within Decemberists lyrics) and the group (which included coauthors Alina, John, and Danny) quickly settled on the newspaper headlines and song title depictions concept. The puzzle itself went from concept to final draft in something like 10 hours thanks to a strong and successful collaborative effort. Once again, thanks to Canva for saving the world from my nonexistent art skills.

I was pretty pleasantly surprised that this puzzle was apparently fun enough and memorable enough to get a few mentions in people’s favorites lists after Hunt. I’d like to think this is evidence that I learned my lesson about what didn’t work with Good Company and used my newfound puzzle construction knowledge to build something more straightforward and solver-friendly. Whatever it is that made this puzzle enjoyable, thank you to everyone who said nice things about it. Your comments made me very happy.

And in case you were curious, here’s that Crane Wife 1 & 2 art that I was sad we couldn’t include in the actual puzzle:

Oil Paintings

One of my favorite things about Mystery Hunt is encountering data sets for things I never would have imagined were things and co-author Charles has a seemingly endless list of these data sets to form puzzles around. In this case, I had no idea that different bowling alleys apply oil to the lanes in different patterns that result in the ball rolling down the lane differently. I was fascinated to learn about how oil patterns work in Professional Bowlers Association tournaments and I’ve had a great time explaining it to everyone I know since then.

Because I had pitched the badge puzzle, I got to be the guy who was responsible for actually printing the badges. Through that puzzle, I established a relationship with a local Maryland printer and the team gave me the title of “Printing Czar,” which led to me getting pulled in on every puzzle that required a printed component. That’s how I got involved in Oil Paintings about halfway into its development process.

For a long time, this was an insane puzzle that I didn’t understand and it was a team effort to bring everything together into the Hunt standout that it became. John did an incredible job running with Charles’ initial idea, turning it into the final concept people saw at Hunt, identifying the paintings, and creating the mosaic. Linus managed to perfectly hide the animals in the paintings. And I think Max was the MVP of this one, generating dozens of iterations of word grids as this worked its way through testing, formatting the PDFs for printing, and refactoring everything several times as printing requirements shifted. It’s kind of incredible that it all came together in time for Hunt, especially considering that we mailed out the first printing for testing on December 20.

There was some debate about how “evil” this puzzle should be. For the first test solve, we cut up the paintings into tiles (not the strips that the final hunt version ended up with) and overnighted them to our unfortunate test group. Thankfully I at least separated the tiles out by painting. I want to give a huge shout-out to Michael Andersen (and the rest of the New York holiday test solve crew) who spent the last week of 2023 piecing together a puzzle that was way more difficult than it needed to be. Michael got us our first clean test solve and his feedback was essential to the decision to produce this puzzle in strips instead of cubes.

The contents of the first testing draft of Oil Paintings just before we sent it to our poor victims

Jennifer and her test solving team also deserve a shout out here for providing the critical second clean test solve in dramatic fashion. We weren’t able to send them a physical copy of the puzzle in time, so we sent them the PDFs that we were going to send to the printer. We were prepared to send them the final collage image once they correctly described every detail of the solve up to that point, but Jennifer shocked me by printing every page out on single-sided paper, taping the pages together, cutting everything out, and assembling the final image.

This passed testing on January 3, which was right up against the printing deadline. I placed the order with the printer for 40 copies on January 4. And then on January 6, we discovered that the PBA website had updated the Viper pattern and threw me into the stupidest panic mode I’ve ever been in.

It turns out that the PBA oil pattern names hardly ever change, but the exact patterns change year-to-year. We wrote this puzzle without fully understanding that fact, but when we figured it out deep into the test solving process, we were comforted that the timestamps in the URLs of the 2023 patterns indicated they were updated in March and we figured the patterns were unlikely to change prior to Mystery Hunt. Still, I downloaded all of the 2023 patterns just in case. When the Viper pattern changed on the official website, I started calling every number and emailing every address I could find for the PBA asking if they could put up a note for Hunt weekend with a link to the old Viper pattern. I know this was a ridiculous request, but I was convinced that the success of all of Mystery Hunt depended on it.

I finally got a hold of someone on the Wednesday before Hunt, just as I was about to board my flight to Boston, and he told me the office was cleared out and everyone had headed to Wichita for the first big tournament of the year. Not only were they not going to update their website, there was also a nonzero chance that all the other patterns could change as well. While that was terrible news for me, I think it actually helped because it led to me accepting that the best we could do was to create a shared Google Drive folder with all the 2023 patterns in it, which we could then provide to teams who requested it. It turns out that all of this panicking was unnecessary - fourteen teams solved this during Hunt and only two requested the backup images. I’m very relieved that this puzzle went as smoothly as it did.

I also need to include a huge thank you here to all my of TTBNL teammates who helped to prepare the physical handout. When I placed the final order with the printer, I asked them to leave a tab with an alphanumeric identifier on the edge of each strip so that I could inventory each set of strips before handing them out. That was great for quality control, but awful for everything else because the printer also left the corresponding blank tab at the other end of the strip attached and it meant that each of the 40 copies of this puzzle that we printed had 260 strips of paper that each needed to have two small tabs cut off the end. About a dozen of us spent many hours between Thursday and Saturday cutting those tabs off all 40 copies. Thank you to all the awesome people on the team who volunteered to help me with that.

I loved that several teams sent images of their completed puzzle and Unicode Equivalence made my weekend by sending a video of the moment they flipped their completed grid over for the first time and burst out laughing at the baffling image they found on the other side. They then also sent a team member over with their completed image and agreed to take a photo with a few of the puzzle’s authors. Coins are great and all, but I think the real prize souvenir from the 2024 Mystery Hunt is an assembled copy of Oil Paintings.

I was stranded in Boston due to the snowstorm on Monday night and ended up wandering the Freedom Trail the next day with a few of the other TTBNLers who were still around. Every time we saw the Boston Celtics logo, we took a photo with it. I bought a t-shirt with the logo on it at the airport before my flight home and I’ll definitely be wearing it to Hunt next year. It might be the new team logo. Is it too much to suggest that TTBNL might finally have a team name and maybe we’re The Boston Celtics now?

John and I are tourists

Actual Mystery Hunt

Running Hunt was a ton of fun, but it happened so fast that I know I’ve forgotten half the things I did all weekend. As Printing Czar, I had two heavy boxes of printed materials to transport from Baltimore to Boston and I don’t think it really hit me until I picked up the final order from the printer the night before my flight that I’d have to load a LOT of stuff into my suitcases. I ended up making it work by losing one of the boxes and wrapping a dozen copies of Oil Paintings and half of the badges in a couple bath towels. I had one suitcase that weighed over 80 pounds and my carry-on was probably another 30-40 pounds. The mid-hunt runaround stickers in my bag looked suspicious enough on a TSA scan that I got pulled aside for an inspection. When I went to board the plane, I was denied because of a baggage issue. I never found out what the issue was, but they eventually let me onto the plane, where I got the very last middle seat in the back of the plane. Thanks, Southwest.

 

The contents of my suitcase after arriving at the hotel

The Thursday before Hunt turned out to be way more challenging logistically than I expected. Logan and I spent hours implementing a system to sort the badges to make sure that every team received at least one complete set of badges with badge holders and lanyards for everyone. It was way more work than I expected, but it did result in one of the most satisfying photos from the weekend:

 

All teams’ badges and first aid kits, bagged and sorted by team size

Thank you to Logan, Sophia, and Astroguy for helping me to sort badges and put together the bags for the teams.

At some point, Alina and I got a message from the editors that they needed us to rewrite a portion of Dear Diary because ongoing test solves had reported a concern. We rewrote most of the clues in about an hour and then put it through another round of live test solving in the room. It was stressful rewriting a puzzle the day before the event, but it was pretty fun to watch it get solved live by a tester who seemed to have a lot of fun with it.

The next 12 hours or so are a blur and I got a little bit of sleep but it was Friday before I knew it. On Friday morning, I helped the team pack up all of the boxes that were going to be going to our on-campus HQ and then Alina and I snuck away to place the posters for the badge puzzle. We joined back up with everyone at Kresge, where I checked to make sure the badges had arrived safely and then watched a few minutes of rehearsal for kick-off.

 

Me hiding the Badges poster just outside Hunt HQ

Eventually teams started to arrive and I enjoyed chatting with people as I guarded the doors to the theater. I was definitely getting anxious when rehearsals kept going past noon, but there was a technical issue with the audio and lighting setup and the team wanted to make sure there wouldn’t be any glitches during the kickoff skit. Thank you to everyone in the lobby for being understanding - I know you all wanted to get into the theater as badly as I wanted to let you in and I appreciate that no one acted even remotely annoyed at me.

After kickoff, we went to launch the Hunt website and it immediately crashed. The whole team was panicking and trying to figure out what to do while the tech team huddled across a couple rows of our lecture hall HQ. A few minutes in, someone realized that we had one round of Underworld puzzles with no significant interactive components and since every puzzle had a draft written in Google Docs, we could put together and release an entire round via Google Docs to buy time for bringing the website back up.

The only issue with that plan was that all of these puzzles had gone through a postprod process and a post-postprod fact check and test solve, meaning that a few of them had changes made after the final Google Docs version. And so we brought together the fact checking team for one last mission: re-fact check the final Google Docs drafts against the latest post-prodded versions and ensure that everything was good to go for release to all teams. We “speed fact checked” and released a 13 puzzle round in under an hour. The fact checking team was incredible.

I think this is a good place to talk about one of the biggest unsung heroes of the 2024 Mystery Hunt: Magii/Nathaniel took on so many roles during both the writing and running of this hunt and he was critical to its success. He was one of the most prolific fact checkers AND one of the top test solve session admins, while also doing a lot of test solving himself. In the last few weeks leading up to the event, he led the postprod fact checking effort and he’s the guy most responsible for getting that round of puzzles out so fast at the start of Hunt. During Hunt, he responded to hint requests, manned the HQ phone, and filled half a dozen other critical roles. We’re all very lucky that magii was on our team this year.

For what it’s worth, there was a brief window where we were considering keeping this fact checking workflow going and releasing the entire hunt that way. The TTBNL tech team deserves a huge amount of credit for getting the servers up and keeping them running for the weekend and Ethan especially deserves everything good in the world for all that he did to make this Hunt happen. For a lot of the year, Ethan WAS the tech team and his accomplishments are unbelievable. Thank you, Ethan, for being amazing all year.

Conclusion

The rest of the Hunt is a blur to me. I remember helping to run the Olympics event and scheming with teams about optimal envelope placement. I remember visiting a team as either Castor or Pollux (even we couldn't tell which one was which) and offering to bestow some “divine favor” if they could convince us two of them were twins, then laughing when the free answer they asked for turned out to be “DIVINE.” I remember setting up the on-campus HQ each morning by carrying boxes of handouts from our Le Meridien offsite HQ and stopping to pick up coffee and pastries at Flour. I remember watching the Discord answer feed and cheering as teams solved metas and puzzles that I had fact checked. It was so much work but I've never had so much fun. I would run Hunt every year if it didn’t mean spending the entire previous year writing it.

I think I’d estimate that I personally put in at least 500 hours of work and I’m just one of dozens of people on TTBNL who dedicated a year of their lives to pulling off an event for one weekend. It was worth it to see people having a good time all weekend and to hear some great feedback afterwards. Thank you to those of you who were generous with your comments - I know I was excited every time I saw a puzzle that I worked on mentioned as someone’s favorite, but I also know most of this team and who wrote what and I was even more excited to see my friends’ work get compliments.

I know the unanimous opinion is that our Hunt was too long and I'm sorry for that. I can’t speak for the rest of the team and I wasn’t on the editorial team, so I wasn’t in the room when the decisions were made, but my personal feeling prior to Hunt was that our Hunt was probably just about right. I know that people fly to Boston from around the world to participate in Mystery Hunt and I believe the running team owes it to those people to provide a weekend of fun puzzles. I know we wrote a shockingly large number of puzzles, but I legitimately believed that we needed the number of puzzles that we wrote to keep teams happy and solving throughout the weekend. I felt that our editors did a great job of trimming down unnecessary steps from lots of puzzles over the course of the year and it resulted in less complex puzzles that test solved cleanly and quickly. I really thought we had a Hunt that would end on late afternoon Sunday, which is where I personally feel an ideal Hunt should end.

I’m sincerely sorry to everyone whose hunting experience was negatively impacted by the length of our Hunt. I hope that our team keeping the site open for submissions for a while helped to soften that a little bit. If nothing else, I took it as a great compliment that teams continued to submit answers for over a week after wrap-up and I was proud of our team's actors for continuing to run the final runaround for teams that crossed the finish line in the week after Hunt.

Despite the (totally fair) criticism, I’m proud of the Hunt that we put on. I really hope that over time, the 2024 Mystery Hunt will be known as “the one that had some tech issues and went long, but had really high-quality, fun, well-edited puzzles and was run by a team that put a lot of effort into making sure everyone had a fun weekend.” 

Shout Outs

I think I’m now realizing that I’m summarizing a Hunt that went too long with a blog post that is way too long, so I’ll stop writing as soon as I can highlight some of the people who made writing Mystery Hunt such a great experience for me:

• Bella: This is just a fact: the 2024 Mystery Hunt would not have happened without Bella. She did an unbelievable job and is an amazing leader. She dedicated more time to this Hunt than anyone else and made so many crucial decisions that led to the event’s success. She oversaw every single puzzle in this Hunt from initial idea to final draft. There were a couple points where Bella made me change one of my puzzles in ways that I disagreed with at the time, but seeing how her changes played out at the actual event, I saw that she was right every time. 

• Craig, Jessica, Nathan, and Adalbert: Please don’t let my praise of Bella take away from the tremendous contributions of the other EICs. I worked with each of the editors at some point throughout the year and I was impressed at how each of them brought their expertise at puzzle construction and unique perspectives to help transform many different puzzle constructors’ ideas into some of the most creative and fun puzzles in the Hunt.

• Ursula: Ursula was one of my first real friends on TTBNL. Besides just enjoying the unconditional optimism she brings to every situation, I think one of the things I appreciate most about her is her dedication to making sure that everything is getting done, but in a way that makes sure that everyone is having fun. Leading the test solving team was not an easy task - I think our team ran over one thousand test solving sessions throughout the year - and Ursula and her team kept detailed spreadsheets about each puzzle’s progress and which team members were unspoiled and available for testing each puzzle. I know that a lot of times people look at the puzzle authors in a hunt and praise the ones they liked, but test solve admins don’t get the respect they deserve. The test solve admin team was critical to the success of this Hunt and Ursula and her team deserve a ton of credit for how well they managed everything.

• Alina: Alina was one of my favorite people to work with all year. It’s scary how smart she is and she has a very good understanding of how puzzles should be constructed. She test solved almost half the Hunt and left some of the most insightful comments I saw, meaning she helped to shape a lot of puzzles’ final drafts. I loved collaborating with her as an author because we helped encourage each other’s puzzle ideas and neither of us was afraid to offer or take constructive criticism. I can’t wait to solve with her next January.

• John: Though he doesn’t want any more recognition, he’ll never get all the recognition he deserves. I could list all the things he did for this Hunt and no one would believe they were all done by a single person. He’s one of the funniest people I’ve ever met and he kept me sane when I thought the Hunt was going to kill me at a few points. I’m also lucky to have been able to collaborate with him on a few puzzles - he’s a puzzle-writing machine and it was amazing to get to see his brain in action.

• As mentioned in the post above, Ethan and magii/Nathaniel deserve so much credit for taking on way more than their share

• Like, literally everyone on TTBNL: I joined as an unattached hunter in 2018 and I flew to Boston that year not knowing a single person and I’ve come back every year since. I’ve been welcomed into this group of people and I had so much fun running this massive, insane event with all of them. Throughout the weekend, every single person was working every minute they were awake to bring the best possible Mystery Hunt to the puzzle community. I’m insanely proud of what we accomplished together.

• My family: I hit a point around Christmas where the time I was spending on this caused some very serious issues within my family. More than anyone, I want to thank my family and friends for putting up with all of this for the last year. You deserve much more than the souvenir Cerberus pins in return for your patience.

   

  Thanks for reading!