With Kapcon around the corner, I've been pondering what makes a good tabletop roleplaying convention game (from either side of the table). I've had some hits and misses in my time running games at conventions and spent a fair bit of time discussing and de-briefing with others.
In the following section I outline 5 common challenges that often undermine convention games, then propose a solution you can employ to account for such obstacles and allow your convention scenario to achieve its full potential.
1. Insufficient character agency
The time constraint of convention games often leads GMs to construct scenarios where there are limited choices. This can be fine, so long as the characters can genuinely make some important decisions, and those decisions are the focus of the game.
Consider the first 20 minutes of any contemporary horror movie - there will almost always be some constraint that limits the characters options and separates them from wider society when the horror strikes. That's not a problem necessarily, so long as the choices the characters then make have real meaning for them and their story (in game terms this means that the story of the characters should be the meat of the action).
Another common type of 'con game with limited agency is the high-concept scenario, where a GM has a vivid and detailed idea about a setting (usually inspired by fiction) but only a rough idea of how the characters will interact with the setting (beyond exploring it) and no backup plan to adapt the story to the characters if they reject some or all of the premise, or simply find it dull.
2. Too little structure
The flip-side of a low-agency game is a game where there is simply too much scope for unstructured character interaction. Consider most investigative scenarios: usually the characters must follow a sequence of clues to solve the mystery. The problem with such scenarios in a 'con environment is that people (and their characters) are likely to have very different approaches to what may seem (to the GM) like a very linear set of clues, and could easily spend the better part of an hour discussing one particular clue and its implications, or following false lead.
Equally scenarios which invite the characters to form plans can fall into an endless planning cycle as characters consider every possible eventuality. This is often not the intent or focus of the game, but an inevitable outcome of a particular set of elements.
3. Assumed character cooperation
Irrespective of whether the players are strangers or not, if the characters are not known to one another, most players will want to 'ease into' character relationships and shared decision making. Some will not want to participate collectively at all, and prefer a 'lone wolf' approach if they feel this fits their character.
That means that most con games need to allow characters space and time to meet and interact, and not count on character cooperation unless this is clearly and explicitly stated from the outset.
4. Poor system fit
Unless a convention game is intended to teach a system it needs to be very light touch, and focussed on only the core activity that will occur for the characters during the few hours of the scenario. Many traditional systems have rules that cover far more ground than that, and this can confuse and bog-down play.
There are many, many different systems out there, and it's well worth shopping around to find one that is well suited for the needs of the scenario. Ideally a discussion of rules is a very short part of the game establishment phase, and players have quick reference sheets handy (or incorporated into their character sheets) to refresh themselves of the key points.
5. Doesn't cater for player styles
People approach games in a lot of different ways. Some are particularly extraverted and likely to enjoy and proactively engage in almost any style of game. Others are less likely to try and grab the spotlight. For some folks having something structured, like an initiative sequence with clear specified actions that are weighted to be equal to those performed by other characters, provides assurance that they'll get an equal level of involvement to their more glib companions.
Irrespective of the system and setting, some players will try and win, or defy the odds, others will try and participate indirectly, playing their characters more as spectators than protagonists. How a GM responds to such approaches will be noticed and reflected upon by the other players, who may change their own approach if they perceive there is a more favourable option.
Creating a scenario sideboard
All of this may seem a little overwhelming, especially for a GM who feels that the players should simply be grateful that they've offered to run a game at all. All too often the temptation is to blame the players if a game doesn't go well, or live up to the expectations of the GM.
But you can do more to ensure that you run the best version of your game possible. When I played Magic: The Gathering, many years ago, decks had a thing called a sideboard, which was an extension to the deck that had a series of additional cards that you could add to your deck to better optimise it once you knew what you were up against.
I suggest that a good convention scenario can have something similar. A series of ideas and options that you can port into your scenario to better adapt it to your players. What options are there for the characters to interact and get on the same page and how can you facilitate this if it doesn't happen organically? What time pressure can be added if the characters seem to want to spend ages planning something? How can you ensure that everyone is involved at the table, and what option might you employ to park a noisy player for a few minutes, to allow quieter players time to be heard?
Because you won't necessarily know what challenges you'll face, the unsurprising advice behind all of this is to playtest your game if at all possible. If you can playtest it with strangers, or people you don't normally game with, so much the better. But a robust playtest should highlight some of the things you need to include in a scenario sideboard.