Watercooler: Endgames

Well, so much for posting a watercooler every Thursday. It’s been an insanely busy few weeks; on the puzzle front, BAPHL 16 was quite fun, Puzzle Boat 4 was fantastic (track down a team if you haven’t, you can still register and solve), and I’d like to congratulate the United States team for earning silver at the World Puzzle Championships. I qualified for the World Sudoku Championship for the first time this year (held right beforehand in the same location), but unfortunately a week in India in the middle of the semester wasn’t feasible, and the chaos of the last few weeks convinces me this was the right decision, if a sad one.

Thanks for those who shared ideas/experience about puzzle testing on my last discussion post; quantity wasn’t what I hoped for, but quality was top notch. It’s definitely interesting for me to hear about practices for events I haven’t written for, like MSPH or Puzzled Pint.

Today I’d like to talk (and encourage you to talk) about endgames; by this I mean the last portion of a puzzlehunt that wraps up the story and usually has some amount of puzzle content. By nature these tend to occur in large-scale hunts– I wouldn’t call the final metapuzzle of a P&A issue an endgame, per se– though I’ve definitely done some BAPHLs where the last meta or metas felt like a separate phase of the event, especially if that phase occurs at a new location.

Naturally, since I have the most history with Mystery Hunt, those are the endgames I have the most opinions about, and there have been some controversial Hunt endgames. In 2015 (20000 Puzzles) after a very slick, clean Hunt, the final endgame was a sequence of time-consuming activities, including a Family Feud game and a task to take something like a hundred selfies. Setec reached the endgame when there was a glut of teams and were thus given a delayed start; I went to sleep instead of participating, and several members of my team wished they did the same.

On the other extreme of time commitment, in 2007 (Hell) we gave teams a construction task that was (in my opinion, and I didn’t write it) one of the most elegant puzzles I’ve seen in the Mystery Hunt. We thought that in itself would be a satisfying conclusion, and since teams had done another runaround earlier (and had “found the coin” five puzzles into the Hunt!) we basically surrendered the coin when teams showed up with a valid solution to the puzzle. The reaction to this was almost universally “That’s it?” I’m not a huge fan of the final runaround, but obviously a lot of solvers look forward to it.

On the other hand, putting too much into the endgame risks a situation where too many solvers don’t see the coolest stuff in your Hunt. 2013 (Coin Heist) was a Hunt with a lot of problems (I think that’s been accepted) but it had a very ambitious heist finish with high production values that even had tasks for large teams that couldn’t send everyone on the runaround. I think it was the most creative element of the event, but because the Hunt ran long, I believe only two teams got to experience it. (On the other hand, some cool elements of endgames can still be shown to people who don’t make it there in person; the GlaDOS interaction at the end of 2011 (Video Games) was a great audiovisual experience, and thankfully it translates somewhat well to video.)

Due to the number of teams participating in the Hunt now, it’s ideal to have an endgame that supports multiple teams at once. I’ve completed several Hunts now where there was a queue and we had to wait for another team to finish before we could start… this is not fun, because you usually have nothing to do during that time (though it’s a good excuse to get people to clean up, or in 2012 (Producers), to get everyone on your giant team to participate in a gigantic award-winning production number from Man of La Mancha). My first Hunt was 1998 (Enigmatology) which ended with two teams looking for the coin in the same weight-limited elevator, requiring the two to make a deal to search in shifts to prevent anyone from getting hurt.

I get super-stressed-out during an endgame if I think there’s any chance of being passed… At the end of 2016 (Huntception), our team was broken into a whole bunch of subteams (which, by the way, would have completely sucked were we a small team) and my subteam got lost; Left Out was beginning the previous task as we completed it, so I was panicking that our group of three people was literally going to cost our team the entire Hunt. In 2009 (Zyzzlvaria), I know that one team (the name-changing team I later joined and constructed with in 2014) had figured out enough about the endgame in advance (based on information we designed not to let teams figure out enough about the endgame in advance) that if they had finished their last meta earlier, I am completely confident that they would have blazed past Luck, which would have been exciting for them and supremely unpleasant for Luck.

For me, I’d like the Mystery Hunt endgame to be a brief brisk victory lap with interesting Hunt content and low pressure.  I was really happy with what we ended up putting together this past year (Monsters et Manus); we knew we wanted to have a significant middlegame accomplishment, and I thought it was a great concept to have our RPG middlegame take place with figures on a tabletop hex grid, while our endgame took place with actual solvers standing on a life-size hex grid. We ended with a short-ish runaround which was a lot more dramatic with musical scoring, and because we had three sets of hex tiles and Mystereo Cantos costumes, we ran more than a dozen teams through it without (as far as I know) requiring any team to wait for a significant amount of time.

So what do you like in an endgame? What don’t you like? And what endgames have you found memorable (for good or bad reasons) in the Mystery Hunt and other puzzlehunts? Have at it.


5 thoughts on “Watercooler: Endgames

  1. I feel a little weird commenting on this one, since I’ve only seen 2 endgames in 15 Mystery Hunts. Alice Shrugged had some really creative bits but was too long, even though it was “only” four challenges. I liked that it used the bits you got from solving earlier metas, but two of the tasks ended up with maybe 6 of us actively doing things and the rest of the team just watching, which is awkward even on a “small” team like II&F. I think I was hoping for something a little more “epic” in M&M, but what you gave us was fine. Again, having most of the team there felt superfluous, but we enjoyed it.

    From what I’ve heard and read about other endgames, I have a few peeves. Answer checks for individual puzzles are unnecessary and should be avoided. At the same time, I want the endgame to relate back to other things teams have done throughout the weekend. I liked the idea from 2005 of using puzzle and meta answers as fodder for challenges along the way. I do like the idea of having jobs for both the “away team” and folks back at HQ. You have to leave someone behind to watch all the computers, so why not give them something to do? But don’t split them up so many ways that they can’t cover all their bases. The requirement of effectively having 30 people available to perform and document the kicks in Huntception would have sunk II&F.

    I’m not sure I agree that the endgame should necessarily be a “victory lap.” If so, then why even have it be a requirement? Again, I’ve never been in a position for this to matter for me, but when I hear that a team made up an hour gap on the front-runner in the endgame, I think that’s dramatic and cool.


    • “I’m not sure I agree that the endgame should necessarily be a “victory lap.” If so, then why even have it be a requirement?”

      Why does the wedding party (often) recess at the end of a wedding ceremony? It’s not going to determine whether the couple is married or not, but it provides a nice sense of closure, and people seem to like it.

      “Again, I’ve never been in a position for this to matter for me, but when I hear that a team made up an hour gap on the front-runner in the endgame, I think that’s dramatic and cool.”

      Depending on the format of the endgame, sometimes two people running the same endgame run it slightly differently (for example, Mystereo had prompts to give if a team didn’t seem to be figuring something out, but there was no formal rule on when to give which hints… the goal was for the team to have fun rather than getting frustrated). If that team that lost because the other team made up an hour found out that the other team had a slightly easier experience due to external factors, I think they’d have a right to be very angry.


  2. I would be okay with an MIT Hunt endgame simply being a runaround with no additional challenges. That would satisfy the “victory lap”/sense of closure requirement and afford the opportunity for a team to actually find a coin hidden on campus. Dan makes a good point about the variable difficulty in endgame experiences. If an organizing team creates an endgame with some possibility of one solving team winning by passing another solving team, the organizers would have difficulty convincing the passed team that the two endgame experiences were equal. I was disappointed during the Normalville endgame when Setec intimated that ACRONYM had a chance to pass another team and win. The suggestion generated false hope and led to some sore feelings afterward. I would have preferred that ACRONYM be told that another team had already started completed the metas and that the endgame was structured so that competent front-runner team with the given time advantage could not be caught, but that we should complete the endgame anyway because it featured fun puzzles and interactions.


  3. I’ve only Hunted remotely and on non-competitive teams so far, so I don’t have anything really to add to this topic, but I’m very curious about what specifically the name-changing team was able to deduce in 2009. Were they able to predict locations on the runaround by analyzing the Collectible Disaster Response Sheets? I would have thought there’d be too many red herrings for that…

    On that topic, a future discussion point could be backsolving and/or leveraging other meta knowledge. I tend to believe that it’s always legitimate and indeed an important part of puzzlehunting (a part that I enjoy, in fact), but I know that some people think of it as “unsporting” or just less fun than forward solving. (This argument might come up more for me specifically because we’re non-competitive; I assume teams that are gunning to win don’t have the luxury of not backsolving.) If that’s not broad enough, throw in how much backsolving should the writing team plan for, and how backsolvable is too backsolvable in terms of a meta?


    • “Were they able to predict locations on the runaround by analyzing the Collectible Disaster Response Sheets? I would have thought there’d be too many red herrings for that…”

      I thought so too, but I talked to them after the coin had been found, and they were able to tell me all the locations without having been led there by the runaround (even the geological sample display on something like the 15th floor of the Green Building…).


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s