Angels and Operators Post-Game Analysis (Part 1)

Jul 31, 2007 09:34


Thanks to everyone who left kind words in the comments to the final installment of Angels and Operators. Now for the promised post-game analysis, or a start on it, anyhow.

All in all, I found A&O an acceptably successful exercise in entertainment but, more importantly, a very illuminating chance to take an up-close, slow-motion look at roleplaying game decision-making, as performed by a very interesting hive mind.

The first thing I came to regret is forgetfully choosing to name the lead character Steve-which is also the name of one of the leads in The Birds. It’s like the old Canadian Football League, with eight teams, two of them named The Rough Riders. This mistake drove me bananas.

Choosing to make the lead character mentally ill was an almost too perfect melding of form and content, allowing commenters to adopt sub-roles as the various competing voices in Steve’s head.

If the number of votes per poll was any indication, the game started to lose around 15% of its audience members in early winter/late spring of this year. This was about the same time when I myself became a little bit frustrated with the pace of the narrative, feeling that I was having trouble bringing it in for a landing. The tension between my intention to extend genuine freedom of choice to the collective player and the desire to present readers with a satisfying narrative is one we’ve all probably experienced to one degree or another in tabletop. Freedom of choice means extensive opportunities for plot branching, and an ever-branching plot is hard to invest with the sort of circularity that lends a traditional story a satisfying sense of completion.

In literary criticism there is the idea of the unreliable narrator-the first person voice whose word we tend to accept, even though we shouldn’t. In RPGs, we’re often faced with a parallel phenomenon: the uncooperative protagonist.

Characters in other story forms are controlled by a sole author and mostly do what you want them to. Maybe during the writing process they take on a life of their own and reveal new intentions and facets, forcing you to alter your preconceptions as you go. But at the end of the day, protagonists respond to the pressures you establish for them by moving the plot forward to its inevitable conclusion.

In roleplaying, as we all know, an inevitable conclusion is a bad thing. Authorship is shared, and the freedom of players to make decisions for their characters trumps concerns of narrative neatness.

In Angels and Operators I was often surprised by the extreme risk aversion of the group consciousness. Characters in stories can’t be so reckless that we lose sympathy for them, but don’t get anywhere by avoiding potential trouble, either. In ordinary storytelling, it is the author’s job to create pressures on the character so great that his only available choices are dramatic ones. In RPGs, this is considered railroading. (As are a lot of other perfectly legitimate GMing gambits, but that’s an exploration for another day.)

In Angels and Operators, we found that the group would always seek more information if it possibly could. Where this option was not available, it would nearly always seek the second-safest seeming choice, eschewing only the most blatantly cowardly move.

The huge exception to this rule was Steve’s decision to sever his hand, for which I, as GM, was extremely grateful.

The biggest surprise confirmation of it was the decision at the climax to hide down in the basement as one of the bad guys approached.

play by blog, angels and operators

Previous post Next post
Up