This is a series about how I, personally, like to play roleplaying games; it reflects my personal tastes and the norms I’ve adopted from the roleplaying scene I belong to. Ours is a strangely insular hobby: over the 20 years I’ve played roleplaying games, I’ve done it with fewer than 30 people, with the exception of a few convention games and online experiments. Because of this narrow range of experience, which I believe is quite typical, I’m always interested in reading about how other individuals and groups like to play, either because they’ve chosen to deliberately, or because it’s the norm they’ve grown up in the hobby with. This post is for others like myself, who share that interest.
I’ve run about as many sessions as I’ve played in, and the GM’s seat has always been an easy one for me to sit in. I wrote a little bit recently about my approach to dice as a player, and this post is a follow-up to that one, dealing with how I think about and use dice as a GM.
The dice serve two purposes for me, both as a player and as a GM: they are an element of the interface between the player and the game world, and a catalyst for creativity. My earlier post dealt with this second aspect from the player’s point of view, and the difference is not great from the GM’s perspective: when I’m running a game, I will often roll to see how NPCs behave, what the world offers up for the players to confront, and how events beyond the players’ reach unfold. Often – not always, and, depending on how one would keep count, not even most of the time – but often, in the sense that I feel as I’m running, I keep myself closer to too many than too few rolls.
I could easily run a session from start to finish with no rolls except those the system demands (or even without those, but that’s another matter), and comfortably coast from one scene to the next, building them on spontaneous invention and reasonable induction from previously defined aspects of the world and NPCs – but I prefer to keep myself on my toes. Rolling to see how the guard reacts, or if there’s a surveillance camera watching, is less comfortable than just deciding – my experience of the game is less seamless – but it spurs me to come up with events and situations that are more interesting and more memorable than what I would otherwise produce. And if the result leaves me stumped, or clashes with my implicit understanding of the scene, I make note of that and ignore the result. (I enjoy random tables, although they’re in short supply in the games I’ve run, and use them, when available, in the same way: to encourage creativity, not as a replacement for it.)
The other purpose of dice, to function as part of the interface between player and world, is served differently from the GM’s perspective than from the player’s. It’s assumed in my roleplaying culture, and often explicitly stated in the rule book, that the GM may always fudge the dice – and conversely, the player may not, although the latter corollary of Rule Zero is, apparently, so self-evident that it bears no mention either in print or in person. It’s not a bad axiom, either, but I’ve also encountered an attitude that goes much further than this: that the GM should fudge the dice, or not bother to roll, whenever he or she has a plan that necessitates, or even just implies, a specific result.
This is not something I subscribe to. I find fudging distasteful: acceptable as a last recourse if the system just won’t cooperate with my guiding vision, but a bad habit, if a habit is what one makes it. It troubles me as a player if the GM ignores the rules too readily; I have no interest in riding someone else’s railroad, or listening to a story and occasionally contributing a line. (I am, in fact, deeply suspicious of story as a goal of roleplaying, but that’s a subject for another post.) If I find it necessary to flaunt the system as a GM, it’s always with a sense of failure.
Why take it so seriously? The rules are just tools: they serve the GM’s vision and not the other way around – this is true. But they serve it, for me, in a very specific way: they provide the protocol (along with freeform dialogue, of course) for the player’s interaction with the world. For the most part, this interaction is mediated by the character: the player inhabits the character, with varying degrees of immersion, and interacts with the setting through the character’s actions and reactions. Obvious, right? And just as obviously, this interaction between the character and the world is part of the world. If the system calls for a Perception roll in a given situation, with a bonus for being in an alert state, that is a statement about how the game world works: people in it have a trait, which the system calls Perception, that varies from person to person, and they may or may not be in a state, bearing the label “alert”, that moderates the functioning of this trait. And systems vary enormously in what they do and do not represent. Even closely related systems: for example, Gurps 4th Edition has a universal Perception trait influenced by a more fundamental Intelligence attribute, whereas 3rd edition has the same Intelligence, and an optional advantage called Alertness that gives a bonus to it – a subtle distinction, but one that distinguishes slightly different models of one small part of the world. Wildly different systems, correspondingly, present widely different models: in Dungeons & Dragons, Intelligence has no bearing on perceptiveness, but a contrasting attribute called Wisdom does. In Call of Cthulhu, the closest trait is a skill, Spot Hidden, that is not particularly connected to any fundamental attribute of a person, as skills in that game are not – a larger aspect of that system’s model of the world.
(Again, I am concerned more with the setting and the players’ experience of their characters than I am with the story that unfolds; if I prioritized the coherence of the plot over that of the setting or the player-character relationship, very different views on rules would follow. Some do; I don’t! Likewise, many systems seek to model, not a world, but a particular kind of story – and I have very little to say about that.)
People, to the best of my knowledge, tend to work with the system when it comes to more static aspects, like things written on a character sheet – no GM, as far as I know, looks at a character sheet with a terribly low Perception score and says “Yeah, but you notice things when it counts” – no, you don’t, that’s what Perception is all about! But the dice, I find, are not always afforded the same respect as parts of a model of the world.
In my view, this does players a great disservice. Their understanding of their characters is built on (among other things) those words and numbers on the character sheet. Likewise, their understanding of how the world works is built in part on the rules. If a Perception roll is called for, and a distinction is made between an alert state and either a specifically distracted or an unmarked normal one, those are statements about the world that a player refers to in building an internal understanding of it and planning his or her actions in it. Habitual fudging takes that away: it makes the rules uninformative, and the player dependent on an understanding of the game master’s whims instead – a much more opaque source of information, less likely to be reliably understood. I’m not touching on the issue of fairness here, although it can be a very significant one depending on the social contract in effect, but am only highlighting the fact that when rules are arbitrarily broken, they no longer add to players’ understanding of the game world. And since they keep all or most of their downsides – the cognitive load of remembering and parsing rules text, the bookkeeping, the arithmetic – the sum total tilts towards the negative. I’m also not arguing against consistently ignoring or simplifying any given rule – that’s called a house rule, and it’s the opposite of arbitrary and ad hoc.
These two functions, the interaction protocol and the invention catalyst, support each other more often than not. As an example, a player character in my long-running fantasy campaign Surunkymen portit (The Gates of Mournwater) was killed in a swordfight with an NPC who was not her equal as a fighter. She was a ferocious warrior from a line of distinguished soldiers, and he a treacherous sorcerer; he had tried to murder her and her companions when they were guests at his home, and she had cut off his escape. I fully expected him to be swiftly killed, or else subdued or forced to surrender, with an interrogation and promises of assistance against the main villain of the campaign to follow. But in one turn, all the rolls went in his favor, and -in aggregate, a very unlikely result, but one solidly embedded in the rules – he killed her with one desperate thrust of his enchanted rapier. This was a situation in which many GMs would have fudged. Probably not his initial attack, or her failed defense, but maybe the hit location roll, or the high damage, or, most likely, her final, failed roll against death.
I, though, had chosen a published system and a selection of house rules whose model of the world very clearly included events like this: that swordfights produce corpses, and that luck fails even the very skilled, and as often as not when it matters most. I could have arranged for a much more forgiving set of rules, if that had corresponded to my vision for the campaign, but this one was built for the possibility of tragic happenstance as much as heroic triumph. The rules served the interaction between the player and the world exactly in line with the nature of that world. And what’s more, the surprising result lead me, as the GM, to explore aspects of the world that I had left very loose – in this case, the nature of death and the possibility of travelling to the underworld – and the story took a turn I had not expected, but one that I now see as defining for the whole campaign. A huge win all around, then! (Well, the player was shocked and saddened by the character’s death – I rubbed it in a little by denying her a chance for last words to her father, who ran in a moment too late – and we had an interesting conversation about it afterwards, but that’s a subject for another post.)