MIT Mystery Hunt Time!

Do you download the daily sudoku variation from WebSudoku? Are the NY Times Sunday crosswords too easy for you? Did you figure out what the heck my icon means just because you could? When you see a series of dots, do you immediately start interpretting it as Morse, Braille, constellations, or T-stations? If any of these apply to you, then you may be interested in joining our team for the MIT Mystery Hunt.

Over the course of Martin Luther King Jr. Day weekend (Fri Jan 12 - Mon Jan 15), starting Friday Noon (EST), our team of around 30-40 in-person and 10-20 remote people work on solving a set of 100 puzzles of any sort you can think of. Puzzle formats in the past have included sudoku, jig saw, crossword, word searches, physical treasure hunts, "paint by number," chess, and many original types; puzzle topics include words, history, programming, trivia, math, science, music, pop culture, MIT-specific, literature, internet searches, and many more. The winning team has to organize the next year's Hunt. (We will not win, we're not close to good enough yet, but we will have lots of fun.) You don't need to be an expert in anything, whatever anybody knows always gets utilized somehow. You only need to participate for as long or short as you want. Sleep is optional.

My team, Lake Effect Snow, will be participating in our third Hunt this January (2007). We are starting to make a name for ourselves as being more highly cooperative than most teams, and for having a large contingent of remote participants. While we are interested in adding more strong puzzlers to our team (Tortoise has been our only powerhouse for the past two years), and more MIT-natives, the biggest criterion for succeeding on the team is good teamwork.


One of the interesting things to the Hunt is the structure of it. It's not just a set of puzzles and whomever solves the most wins. Instead, each individual puzzle has at least two layers to it, and then combines with other puzzles into metapuzzles, and metapuzzles combine to eventually come up with the winner. When solving puzzles and working towards the goal of winning, you go through stages.

1) Simple (typical) puzzle Answer
A (live) launch event on Friday noonish introduces all teams to the theme of the Hunt. Then every team starts off with something like 3-5 puzzles. Each puzzle you first solve as "normal." For example, finding all the words in a word search.

2) One-word Solution
Somehow the "normal answer" to the puzzle has to be turned into a one- or two-word solution. Maybe the leftover letters in a word search spell something, or identifying locations on the MIT campus from photos will spell out a word if you plot them on a map. Sometimes you'll have to use anagrams or other similar things to go from the "normal answer" to the "solution". Sometimes you can get the Solution without the "normal answer," and if you're progressing towards the goal, the Solution is all you need.

Solutions are usually submitted to the Hunt organizers (Hunt HQ as we call them), via phone, and once you submit them, they will verify the solution and unlock (reveal) some 3-5 additional puzzles that you can then work on. We typically experience exponential growth of the number of puzzles available during Friday afternoon/night, and then taper off on Saturday as we get stuck and can't come up with more Solutions. Puzzles unlocked together are part of a group, called a Meta or Metapuzzle.

3) Metapuzzle
The one-word Solutions for a few "linked" puzzles then combine to create an additional layer of puzzles (metapuzzles), which also have a simple answer and then a one-word solution. Solving one Metapuzzle will usually unlock a few simple puzzles within a new Metapuzzle.

Some Metapuzzles are also typically on timed release, so groups that are struggling with one set will eventually get new puzzles and and can progress from them. Last year there were actually two Metapuzzles for each set of puzzles - we didn't realize this for quite a while, and then when we suddenly realized it we unlocked a crapload more puzzles. :-P (Not telling the puzzlers what's required is typical, figuring out the "rules" to the game/puzzle is part of the fun.)

4) Run-around
Solve enough Metapuzzles, and they further combine into a treasure hunt throughout the MIT campus to discover a coin or token that is hidden. The first team to the Coin wins. Typically 2-3 teams (unlikely to be us, but we can always hope!) make it to this stage. This usually happens between Sunday morning and Sunday night of the weekend.

Once one of the teams completes the Runaround, locates the Coin, and brings it back to Hunt HQ, the game is over. It typically takes another hour for word to trickle down and all the teams to realize this through their sleep-deprived caffeine-fogged brains, but when no one answers phones at Hunt HQ it's usually a good sign. A (live) wrap-up then occurs some 3-8 hours later, where the organizing team summarizes the overall structure of the year's Hunt, describes the Runaround for all the other teams that didn't make it, and then does a Q&A session for anyone who wants to stick around.


If you are interested in participating, please let me know - email me at zandperl-AT-gmail-DOT-com if you want privacy, or comment on this entry if you don't. If I don't know you in person, tell me a little bit about yourself either here or via email - we haven't been particularly screening out people, but since we want to make sure we keep the good teamwork I think it'd important that we know a little about people who want to join.

I hope to hear from people!