|Mystery Hunt 2017: What the heck happened?
||[Jan. 15th, 2017|07:11 pm]
This post will cover the hunt and hunt structure a bit but will not include any puzzle spoilers.
What the heck happened? A question a lot of people are asking themselves right now, in at least a couple of contexts.
What happened to Beginner's Luck?
Beginner's Luck fractured after the last hunt construction. I am not going to go into too many details here, but a bunch of people left, going off to Palindrome, Death and Mayhem, Left Out, some reddit-based team, and some people did not tell us which team they were moving to. Some people wanted to be on a team with a stronger setup for remote solvers, some wanted to be with specific other friends or followed their friends out of Luck, and some may have had a "seven-year itch" and just felt it was time to move on. I do not harbor any resentment for those who left; all things must pass. But it was not really until October that we realized the scope of the depatures.
What was left was the tightest core of the group. The original 5 members of the plugh team that I assembled to play in and ultimately win the Sekkrets puzzle hunt on Grey Labyrinth in 2006 were all still here. We had a couple students and a few other members remaining, just 10 people in all. This was not a viable Mystery Hunt team. We considered trying to rebuild, adding Charles Steinhardt (who, for various reasons, was never actually on Luck though he did some other non-MIT hunts with us) and some of his buddies and a couple other people, but we were not even going to reach the 22 members we had for Luck's first hunt in 2009. So we looked around for a team we could join and ultimately decided to join with Death and Mayhem this year.
How did you solve the entire Mystery Hunt in 15 hours and change when it has always taken 35+ hours in the past?
The puzzles themselves were good and clean. I did not see the majority of them, but we did not have a lot of pratically unsolvable puzzles lingering around nor overly long ones. D&M has a good puzzle tracking system and people who were dedicated to keeping new puzzles entered and ensuring our progress was tracked properly. These people had little time to actually look at puzzles during the event because of the speed everything was going at, so I thank them for their work.
This hunt had a parallel structure, where we were getting puzzles in various different rounds at once. This meant we could work on all these rounds at once, and unlock further puzzles in all the rounds at once. This was perhaps not surprising, since the last two Setec hunts (Monopoly in 2002 and Normalville in 2005) had similar structures. More importantly, it let us work on several rounds and metas at once.
Solving quest metas was really powerful. Each of these quests leveled up every character, giving 6 points toward opening other quests. These levels also helped open more puzzles in the character rounds and the other open quests. In addition to this, the Cleric's meta was completely backsolvable, to the extent that we were able to submit the answers to his remaining puzzles immediately after unlocking them, giving more points toward opening quests and more puzzle unlocks elsewhere. Some backsolving occurred for some of the other metas as well.
Death and Mayhem was already a strong team, even without a dozen additional strong solvers from Luck. They have been close in a lot of recent hunts. D&M is a large team, with about 60 or so people on campus, a big remote group in California, a smaller remote group in the UK, and other remote solvers elsewhere. Including all the casual solvers who just drop in for a while (some of whom missed the entire hunt this year!) there are perhaps 150-170 total people. There are 288 accounts on the slack, but this includes some bots, past members who aren't playing this year, multiple accounts for some people, etc.
The result of this was that whenever we solved a quest meta we unlocked a whole bunch of puzzles, but we had enough people to work on all of them at once. (Well, maybe not all of them at once. Apparently we had 51 open puzzles at one point. But many, many of them, though.) And we had Charles, sin_vraal, dalryaug, and to some extent me working on metas for much of the hunt, in addition to some of the D&M folks. We solved metas quickly, sometimes based on few answers. I'm sure the wrapup tomorrow will have something to say about this. The result was just more puzzles open faster than we would ever see in a normal hunt and certainly much more than Setec expected us to get.
In the end, the last area on the map required such a high score that we had to have solved not only almost all the puzzles and all the metas, but also all the events. The intent was that these events would provide slower teams with a boost and a bunch of puzzle unlocks, but when we were able to open and solve everything else so early, the events became a bottleneck. Setec realized this was a problem, and had already considered the possibility of having to run the last event early for a fast team to avoid making them sit around waiting for the last event. When the hunt rolled out at the speed it did, they did these private, early sessions of events for several teams, and they did them for the last 3 events for at least a couple teams.
The endgame was short and simple. We had a brief final confrontation with Mystereo Cantos and then got the runaround instructions that led us to the coin. The entirety of three private versions of events, the final encounter, and the runaround took about two and a half hours.