A few weeks ago, Team Snout staged qualifying for their upcoming WarTron game. Held in Portland at the beginning of August, WarTron is another overnight puzzle hunt, or, as it's simply known around Stanford, The Game. Previous incarnations have included WHO, Ghost Patrol, and Doctor When. Though I playtested that last one, I didn't get to play any of these in earnest for various reasons, so I was eager to get a chance here. My usual team decided not to do this one, but I was fortunate enough to be asked to join Do Not Bounce, a team based in Portland and Seattle. We all were at the ready at noon on May 20th, hoping to tear through the qualifying puzzles and gain a spot in the game.
Things were going very well for a while. I got a nice charge out of discovering a message hidden in a quiz using Braille. We were feeling confident as we progressed to a simple WarGames-like interface. We probed around the system, trying to interpret the responses to our various commands. Then disaster struck.
Suddenly, the system was responding very slowly to our input, and then not at all. Game Control had suffered a massive server crash as a result of all the demand. It wasn't as simple as a quick restart; the website was down for hours. GC had no choice but to revert to an alternative system that was far less efficient: answering teams' input through email. This was frustrating, to say the least. We'd been told that qualification should last about three hours; it was now going to be much longer. This was a problem for me in particular, as I had plans at 4 PM. I wasn't willing to be a bad friend by breaking those plans, so I had to abandon my team.
By the time I got back (a bus breakdown delayed me further; thanks, SF Muni!), a scant few of the twenty spots remained. I wasn't clear on what my team was doing, but I got fed a cryptogram, which I solved virtually immediately by plugging a promising-looking word into a grep search and using my Excel/VBA-based cryptogram editing tool to finish it quickly. From that, we learned we had to use an application to read hidden text in some pictures. Since I didn't know what was going on, I waited anxiously while my teammates performed the last task.
We were too late. We finished 22nd. I was, er, not happy. I truly felt we'd been hosed. Yes, the server situation affected everyone similarly, but how many teams were a man short as a result of the lengthened event? I could have saved my team a half-hour if that cryptogram had come up when it should have. Plus, the non-automated email system certainly led to some unpredictability in terms of when teams got their replies and were able to proceed, and we didn't miss the top twenty by much. I wasn't mad at GC; though they perhaps should have tested their system with far more rigor, they did their best in a tough situation. I just found it hard not to be extremely annoyed at our misfortune.
Thankfully, the story has a happy ending, at least for our team. Because the limiting factor in WarTron attendance is number of people and not number of teams, Team Snout was able to admit the first few groups off the waiting list. I do feel for the remaining teams even though their times were a few hours behind the rest of us; I don't know what would have happened if there hadn't been technical frustrations.
It got me thinking about the best way to admit teams to The Game. Even ignoring the technical issues, is a live, timed puzzle hunt the best way to decide who gets in? I can think of a couple flaws. For one, teams who are free on the weekend of the real game but unavailable on the day of the qualification might feel slighted. For another, rewarding only puzzle-solving ability leads to the strongest squads getting in at the expense of less experienced ones, creating a sort of "the rich get richer" situation. Then again, admitting the best puzzle solvers gives you more license to make your game more innovative and difficult.
An alternative, which was used for Ghost Patrol, is a creative challenge that teams must submit by a certain date, and the owners of the best entries are invited to play. This removes the pressure of having to be present for a live qualifying event, but has its own problems. Chief among these is the subjectivity involved; it can be hard or impossible to elucidate why one submission was judged to be more worthy than another. A rejected team might disagree with Game Control's assessment and feel they were excluded for personal reasons, be they game-related or otherwise. That's not a very healthy situation.
Then, there's simply opening sign-ups at a certain time, and players get in on a first-come-first-served basis. In a game with high demand, though, this has an even more extreme version of the problem with the live qualifying. Plus, it's not a satisfying experience for either GC or the players; there's no whetting of the appetite.
Personally, I favor a hybrid. Give teams a short "pre-game" that's puzzle-based and can be completed at any time before the due date. While the bulk of this would have a defined final answer, there could also be a creative challenge used as a sort of tiebreaker. If a team running the game doesn't want that subjectivity, maybe they could include an optimization puzzle, which has many solutions along with a scoring mechanism that makes some answers better than others.
All of these plans, though, make an assumption: Why should the number of teams be limited? Can't a game's organizers resolve to accommodate all the teams who wish to play? It's very easy to feel this way, particularly while feeling the sting of being excluded from a game. It's certainly how I felt when my team wasn't chosen for Ghost Patrol. In an ideal world, of course, every group running a game would welcome everyone. But in a long game with high production values, not putting a cap on the number of participants is asking for trouble. Having to run the game multiple times would likely force sacrifices in it.
What do you think? What's the best way to decide who gets to play in a game with limited room?