A Really Absurdly Long Post About The MIT Mystery Hunt
Part 1: With This Puzzle I Thee Wed
When my Mystery Hunt team received the wedding invitation shown above, speculation immediately ran rampant as to who M & G might be. There had been a Hunt years ago that also began with a wedding, of the video-game characters Mario & Peach — maybe Mario was now leaving Peach for… uh, GLaDOS, from Portal?
And then somebody suggested that the wedding might be for our friends Mark Gottlieb and Gaby Weidling. They are both game and puzzle designers, and years ago Mark attended MIT in large part because of his fascination with the Mystery Hunt; he actually created one of them mostly (and maybe entirely) on his own, back in an era where this was possible. During my team’s discussions of whether or not Mark and Gaby would prove to be the weekend’s M & G, no one expressed the slightest objection to the idea that they would want to get married in an MIT lecture center in front of hundreds of puzzlers, some dressed in costume. Of course they would. But that didn’t mean that was the plan.
The MIT Mystery Hunt is a weekend-long event featuring some of the toughest, trickiest puzzles you could ever hope to see. Each Hunt is centered around a different theme, and that theme is usually a closely kept secret — although sometimes the organizing team allows a hint or two to spill out, as with the above wedding invitation. I figured we would still need to wait for the event itself for confirmation.
But then I ran into Mark while walking with a friend to a pre-Hunt escape room. (The local escape room companies were pretty happy that thousands of puzzle nuts had rolled into town.) Mark was with his sister-in-law and two nieces. After a brief conversation, he said with deadpan dryness, “Pay no attention to the fact that I’m here with my extended family.”
So, wow. This was a genuine turn of events. Not only would the Mystery Hunt begin with the wedding of two great people, but we knew what the Hunt theme would be before the fact. Unprecedented!
How, we then wondered, would the matrimonial theme be tied to the many dozens of puzzles we would solve over the weekend? Here’s how:
At the Hunt’s kickoff, an officiant — himself a puzzler — assured us that we were in fact about to see an actual wedding. The groom was introduced. Mark came out to applause and said with deadpan dryness, “Thank you. I hope nothing goes wrong.”
The wedding procession began. The bride was beautiful. The whole thing was deeply moving. The officiant opened his mouth to begin the ceremony, and suddenly a young girl dressed as Cupid ran out on stage yelling “Stop the wedding!”
It seemed that evil ninjas had stolen all of the love in the world, had condensed it to a single dense core of love, and had hidden it somewhere on the MIT campus. For some reason, they had left behind a number of difficult puzzles that would lead to this hiding place. If only there were hundreds of puzzle solvers available to help crack these devious enigmas! Unfortunately, tragically, the wedding would need to be postponed until—
“Found it!” said Gaby, who had run off stage while all this was being explained. She was now back, waving a heart-shaped object that was, presumably, a dense core made up of all the love in the world. The Hunt was over! The wedding was back on!
And so Mark and Gaby were married, and an audience full of puzzlers applauded and wiped their eyes a whole lot.
The newlyweds then gave us an unexpected gift: They invited us along on their honeymoon.
Mark and Gaby planned to go that most magnificent of amusement parks, Penny Park. Unfortunately, the park had fallen on hard times; it just wasn’t the same as when Penny’s grandmother used to run the show. In order to return the park to its former glory, what it needed was buzz, lots of buzz. The kind of buzz generated (for reasons that remain a bit obscure) by solving puzzles! If only there were hundreds of puzzle solvers available to crack some devious enigmas!
And so the Hunt organizers had kept the theme under wraps after all — the wedding invitation was merely a spectacular fake-out. Hunters spent the weekend exploring the rides and attractions at Penny Park, with a puzzle for each and every one.
Part 2: Excelerating the Difficulty
My Mystery Hunt team, Palindrome, is one of the event’s larger teams — the organizers had suggested that teams cap themselves at 75 or so, but we wound up somewhere just north of 100, when you count both our on-campus solvers and our remote solvers working from the comfort of their living rooms.
If you are unfamiliar with the Hunt, you might be saying to yourself: 100 people solving puzzles together? Did you complete the event in fifteen minutes? Not hardly. In the end, only five teams out of 150 completed the Mystery Hunt in its entirety. My army of puzzlers was the second to do so. When I say that the Hunt is made up of the toughest, trickiest puzzles you’ve ever seen, I am not exaggerating for effect. And there were nearly 200 of them. We started solving on Friday afternoon and crossed the finish line about fifty hours later — and only because we worked around the clock, making sure we had solvers covering the “graveyard shift.” Without that, we would have had no chance at all of finishing.
How hard are these puzzles? Well, let’s take a look at one example, shall we? I present to you: The Excelerator.
A lot of Mystery Hunt puzzles present an initial obstacle, one I have a hard time surmounting even after twenty years. Call it the intimidation factor. There are a lot of puzzles over Hunt weekend that I cannot hope to solve, as I simply don’t have the necessary background — Hunters will find many puzzles about science, or deep math, or programming. It’s easy to glance at a puzzle and say: “Nope! This is clearly a puzzle for somebody who is not me.” That’s the intimidation factor, and Excelerator had it in spades.
Solvers who are used to puzzles looking a certain way — a grid of letters, perhaps some artful clues, and so forth — will be forgiven for reeling backwards when they open up Excelerator and come face to face with this (click to enlarge):
Yes, the puzzle exists entirely within a massive Excel spreadsheet — 700,000 cells, each containing either a random-seeming trio of letters or an equally random-seeming number. I looked at this puzzle for ten seconds and gingerly shut it back down again.
But in fact, getting started on this puzzle is not that difficult. You see those cells containing red letters? There are eleven of them, and those are your starting points. From each one you can trace out a clue. For example, starting at the MOU (in cell C6), you can move cell to cell and spell out MOUNTAIN FAMOUS FOR BEING INCREDIBLY HIGH. You might even notice that those cells, shaded in, form a letter A.
So what mountain is famous for being incredibly high? Everest? Sure, but there are other very tall mountains — like, for example, K2, which is the second-tallest mountain and also a cell in this spreadsheet. Sure enough, if we go to cell K2, we can trace out another clue (in the shape of another letter), the answer to which is that legendary cruise ship, the QE2. And if you go to cell QE2… okay, you’re getting it.
Ultimately, after locating and deciphering a LOT of clues, you wind up with a list of words like SUMAC — words that begin with Excel commands, followed by a column or cell in the spreadsheet. The word SUMAC, for example, instructs you to take the SUM of column AC. Do that and you get a total of 12. The 12th letter of the alphabet is L. Figure out what letter you get from each of these final “command” words, and you wind up with your answer: LINE NUMBERS.
I guess that’s still pretty intimidating.
But if you can imagine 200 puzzles at this level of extreme craftiness, maybe you can grasp why even a team of 100 die-hard solvers might struggle to reach the end of the Mystery Hunt.
Part 3: Puzzles To Keep You Up At Night
If I didn’t wade into the Excelerator spreadsheet up to my neck, what did I spend my time on? A whole lot. For example:
Storybook Pals: As a Mystery Hunt team solves puzzles, the answers they collect come together to form a “metapuzzle.” Solving a metapuzzle is a big deal — it’s a giant step toward the event’s finish line. We have a lot of superb solvers on my team — people who can crack a much-needed metapuzzle long before all of the necessary answers that feed into it have been collected. The moments where I have been the person to break into a meta have been among the headiest of my puzzling career.
On one metapuzzle last year or maybe two years ago, I had a strong notion of what the solution path might be. I chased my idea down that path a short way, didn’t see anything promising, and gave up… and then, hours later, learned that I had been correct, after somebody else found that same solution path and took it to the winner’s circle. If cracking a meta is the pinnacle of puzzling, then having the right idea on a meta but screwing it up has got to be one of the lows.
Storybook Pals was another low. I thought early on that the answers in the round should criss-cross together, but at that time I didn’t have enough answers to get anywhere — and by the time we did, somebody else had already solved the puzzle. Whee!
But in truth this was a good year for me and metapuzzles. I never had the explosive aha moment, the one that causes the room to burst into cheers and applause, but I made suggestions along the way that proved helpful and accurate, and that’s at least something. I’ll take it.
Tall Tales: The puzzle was presented as eighteen short videos, each depicting the same sedate white-haired man, the sort of man who seems to have been born a grandfather, sitting on a sofa, reading from a book entitled “Tall Tales.” A sample tale:
There I was, just minding my business, when I don’t know how many thousand caped superheroes—you know who I mean—showed up in town. I remember, because it was the same day I bequeathed my factory to Willy Wonka. I had already invented one thousand items like chocolate bars and gobstoppers and lollipops, and it was time to move on to new challenges.
Here was a rare example of me knowing what to do every step of the way: I recognized that each tall tale contained a reference to “thousands” of something (“thousands of caped superheroes”), and another reference to a specific quantity (“one thousand items like chocolate bars”). I understood that each of these was a clue to something, and soon after saw that these clues must lead to units of measurement both common and obscure. Sure enough, there is an ancient unit called the BATMAN and another unit called the CANDY; these were supposedly used in commerce in Persia. (I said they were obscure.) “There are twenty candys in a batman,” is a thing I actually said this weekend. Making that conversion gave you a number of thousands between 1 and 26. Taking the appropriate letter spelled out a phrase.
We solved this puzzle at 2:00 a.m., long after my brother Dan, one of my teammates on Palindrome, had gone home to get some sleep. The next morning when he came in I said to him, “While you were gone, there was a puzzle about measurements. What was the answer?”
He blinked at me. “What? How should I know?”
“Let me say put it a different way,” I said. “There was a puzzle about measurements at the MIT Mystery Hunt. What was the answer?”
“Oh.” Enlightenment dawned on Dan’s face. “Smoot.”
Of course it was. Oliver Smoot went from ordinary undergrad to MIT legend in the early 1960s, when as a fraternity pledge he was made to lay down repeatedly across the span of the Harvard Bridge, thus measuring it in “smoots,” a unit that has gained enough fame that it’s actually in the American Heritage Dictionary.
And this puzzle went from ordinary MIT Mystery Hunt puzzle to legend because the puzzlewriters actually got Oliver Smoot himself, now a sedate white-haired grandfather-type, to be the narrator of their book of tall tales.
Spaghetti Western: A personal first: A Mystery Hunt puzzle that mentions me by name. Some years ago I invented a silly game where participants try to “solve” a puzzle that is in fact just a bunch of random words. I called it “Spaghetti,” because in an old magazine article about puzzles, I wrote that some of my friends were so smart, they could solve a plate of spaghetti. I usually run a round or two of the game in the days leading up Mystery Hunt — and this year I was specifically asked to do so by Larry Hosken, one of this puzzle’s co-authors. I promised that I would, and then I forgot. Sorry, Larry! Your puzzle was really fun, though!
Concierge Services: A delightful and relatively easy puzzle that brought great glee to my team as they watched me talk to the organizing team on the phone, and then watched my face when the organizing team hung up on me. Eventually we figured out that the questions we were being asked could all be punnily answered with the name of a Tony-winning musical. (What did Tiger Woods say after he missed the putt? “Bye, Bye Birdie.”)
Creative Pictures Studio: The early rounds of the Hunt were fairly standard affairs, as these things go: Great, solid puzzles, and interesting metapuzzles that required just the right amount of aha to crack. But then we ventured into the outer areas of the amusement park. (Remember how this hunt is themed to an amusement park?) The four outer areas cranked up the innovation, on a scale of 1 to 10, to about a 15. In the Creative Pictures Studio round, the answer to each puzzle was not a word or a phrase but an emoji — you literally had to paste a desert island emoji into the answer submission box, and not the words “desert island.” And then all of those emoji were used in a wild braintwister of a meta, in which the growing pool of symbols were used to spell out the plots of ten different movies. This was easily the puzzle that led to the most raucous debates in our HQ, as we tried on and discarded different movies, trying to find the ones that fit.
The Pennies: I actually had nothing to do with solving this, one of the final puzzles of the Hunt, but it was such a beautiful touch that I can’t let it go unmentioned. Each metapuzzle we solved in Penny Park earned us a new pressed penny. You know those machines they have at tourist spots around the world? Put in a penny (along with fifty cents to pay for your souvenir) and you’ll get a flattened oval upon which is etched the Lincoln Memorial or whatever? The Hunt organizers brought in ten such presses, and allowed us puzzlers to press our own pennies, and of course those pennies formed an elegant puzzle of their own. It was a lovely, nostalgic way to make a puzzle leap off the page or the computer screen.
Part 4: The End?
Looming over this brilliantly created event was the feeling that the Mystery Hunt might soon need to transform itself or fade away entirely. Every year, more people from outside the school are drawn to the Hunt, and the MIT administration has grown increasingly disgruntled at the idea of thousands of non-students running around campus, unregistered and untended. The Mystery Hunt may be one of the premier puzzle events in the country, with a long and storied history, but the administrators, it is clear, see it primarily as a potential liability nightmare. This year the ruling came down: If your team has no MIT students, you cannot stay in your campus headquarters overnight.
Since overnight solving is one of the great joys of the Hunt, this threw everyone into a tizzy. My team, which in fact has no students (despite years of trying), scrambled like mad trying to find an auxiliary HQ — ultimately we rented a house nearby through VRBO, an Airbnb competitor, and other teammates set up shop in various hotel lobbies. Irony #1: We wound up having our most productive overnight shifts in years. Irony #2: Because the school wanted all of the no-MIT-student teams in one place, our on-campus rooms were switched at the last moment… to by far the nicest HQ we’ve had in decades, including a room with a long conference table that facilitated group solving beautifully.
The new rules imposed this year will likely not be the end of the matter, and right now it feels like the day is looming — though I hope I am wrong — that I learn that because my team has no MIT representation (except for many alumni and one member of the faculty), we are no longer welcome to participate in the Hunt.
I don’t feel “entitled” to access to the school’s resources or goodwill, and I am deeply appreciative of the time I have gotten to spend on campus, solving puzzles with friends and reveling in the school’s bursting-at-the-seams braininess, creativity, and eccentricity. I realize this is a gift that can be taken away from me at any time, and if that happens, what can I do except thank the school for its generosity over the past 23 years.
I sure hope they can find a way to accommodate us, though.
Final note: Palindrome was gunning for a win this year, but despite solving at what I thought was a damn impressive speed, the {plane noise} Galactic Trendsetters {plane noise} outmatched us every step of the way. Those are some bright kids over there. Congratulations to them, and I am excited to see what they create for next year’s Hunt. Hey GT: As much as I enjoyed this year’s event, might I gently suggest a Mystery Hunt that more than five teams can finish?
7 Replies to “A Really Absurdly Long Post About The MIT Mystery Hunt”
I thought the Smoot coup was absolutely brilliant. The reveal during wrap up deserved much more applause than it got. Maybe we were all just too impressed and stunned.
I thought that we (Central Services) were also solving at a damn impressive speed (albeit almost at zero overnight; smaller and older team!). But then new lands kept cropping up. Despite being (as usual) in somewhere around 12th-15th, we saw not much more than half the Hunt, I think.
All of which is to say: yeah, it would be nice to be have the Hunt be one that a student team could make good progress on… and which we might hence get to finish :). Still: it was a *great* weekend!
I was one of those who thought the D&D Hunt was the exact right length. If you’re going to be on a team of 100+ experienced solvers, you SHOULD finish the Hunt in 15 or 17 hours, so that smaller and less experienced teams have a fighting chance of finishing at all.
Constructing teams keep suggesting (though not imposing) a maximum team size. At the end of the day, however, team sizes will be dictated by the difficulty of the event. Give us a few years where giant teams have nothing to do on Saturday night, and team sizes will regulate themselves accordingly.
CServices did indeed finish the D&D Hunt, and it was a moment of GREAT joy for the team. I guess that I should be rooting for Palindrome to win next year :-).
While the coin wasn’t found much later this year than the last few years (with D&D being an obvious exception), it felt too big to me in a way that those didn’t. I think that was a combination of the amount of content and the unlock structure: my team, Up Late, was in 13th place by solve count, and by the end of the event there were dozens of puzzles that we hadn’t even unlocked. I’m especially disappointed because I really liked the interlocking meta structure of Safari, and we only got to see about half of it.
I wonder if the “only 10-15 puzzles open at a time” thing leads to big teams just dedicating more people to the night shift rather than shrinking/splitting the team.
For the penny presses, the mills were bought commercially, but the dies were personally manufactured by someone on LO.
I literally cooed when I read about Mr. Smoot. What a fantastic touch. Thanks for including Nic this year—I continue to look forward to your summaries every year!