You are getting this email if I know you've GMed a game on mud in the past. I was reading through lpsmith's notes for ussura, and it got me thinking. I was wondering if you had any kind of notes - maybe not as detailed - of the campaigns you were running, and if you would mind sharing them with me. I was asking Sargent recently about any "Tips-for-new-GM" things I can read, and it occurred to me that lpsmith's game notes was a good start. So I am looking for any more. Thanks!

Well, I have some scribbled paper notes for Nobilis, and probably somewhere some similar paper for the D&D game, but the only notes I seem to have online are notes for the musketeers game. Which are probably not super-helpful, especially if you didn't play in the musketeers game. But, um, because I like talking about this, let me write up some more general stuff about the theory of GMing. I should start off by saying that some of this is going to be trivially obvious. I am not sure exactly how much you know or are comfortable with so I am going to be as basic as I can.

There are lots of reasons why people play RPGs, and there are lots of RPG styles, and not all of them even require a GM, let alone have the same idea of what the GM's role is. I'm only talking here about a specific style, where the GM's main role is to take the lead in coming up with and working through a story, and the other players' main role is to be active parts of this story via a particular character in it.

So to get more specific, the way a campaign usually starts is with the GM coming up with a basic premise for a setting, and then finding some interested players. So the GM might say "Hey, let's do a tropical-island sort of pirate adventure" or "Hey, let's do a Star Wars adventure" or whatever. Assuming folks are interested, then the other players start creating characters that fit into this setting. Meanwhile, the GM is fleshing out the details of the setting and working out a storyline to take place in this setting, plus designing some characters, places, and events that'll be part of the story. When everyone's done, play begins. The GM gives a starting setup, which, combined with their character writeups, gives the players some goals. They try to achieve those goals, while the GM works on making achieving the goals more difficult. Eventually the story starts moving towards a satisfactory place to end and the GM steps back on the adversity level, and the game comes to a conclusion.

GMing during play

Ok, so, we see from this that the GM's main job during play is adjudicating player's attempts to accomplish goals. When a player tries to accomplish a goal (and for purposes of this we can say that all character actions are the players trying to accomplish a goal), there are three basic responses.

The first response is "yeah, sure, you can just do that." This allows the player to accomplish the goal without argument. This is a good response if accomplishing the goal is necessary for the storyline to proceed — say, the players need to find some secret door to continue tracking the bad guys. It's also a good response if the response is totally irrelevant to the storyline — if nobody cares whether the character is able to whistle, just let them.

For everything in between those two extremes, "roll the dice and let's see" is a good response. The point of rolling dice is to make the story less predictable and more interesting; to make things happen that nobody involved is predicting. Therefore, you should never roll dice if both outcomes aren't interesting. Usually, this is covered by the previous case — if not finding the secret door isn't interesting, don't make people roll to find it. Part of the job of the GM is to set the parameters for these rolls. This means the GM needs to have a good feel for what skills/abilities are available to the characters, and what different difficulty levels mean. If a player doesn't make a roll they were expected to make (or if they make one they were expecting to fail) this should make the storyline more interesting, not bring it to a halt. One way of doing that sometimes is to let the character sacrifice something meaningful to turn failure into success, but the GM has to be prepared for them to not want to make that sacrifice, and stick with the situation.

Finally, the third response is "You can't accomplish all of thing X, but you can accomplish part of it — how about Y? Or is there another part you'd like to work on?" This creates another goal that has to be accomplished before X can be done, which also gets resolved in one of these three ways.

This last is an important technique because it's one of the major ways to introduce new goals for the players. The other way to give the players a new goal is to change the game world somehow. Depending on the change, this probably gives the players a new goal or two to work on. Changes to the game world can occur purely by GM fiat (eg, the GM declares that the king dies in his sleep) or in response to PC action (the player has to roll to see if they can climb the cliff in time to warn the village about the approaching army; whether they succeed or fail, the world changes in a way that's likely to give them new goals). It's best to minimize the number of world changes from pure GM fiat. They can be surprising and put the story on a new and interesting course, but in excess they make the story feel railroaded and like the players aren't making a difference.

So, the summary here is that the job of the GM during play is keep the story going. This doesn't necessarily mean force it to go in a particular direction; just that the players should be given things to want to do, and then the GM should make it hard to do those things. Rolling dice is generally a useful tool in making this all happen: it increases the tension level, can serve as an obstacle to accomplishing goals, and can send the storyline heading off in a new and cool direction. Just don't call for rolls too often, or about things that don't matter.

GMing outside of play

The other part of GMing work is planning outside of actual gameplay. How much planning you do depends pretty heavily on how comfortable you are with ad-libbing/retconning and what kind of story you have in mind. For instance, I plotted out the events of the muskeeter campaign in fairly detailed fashion, and applied a pretty heavy hand to make sure people stayed on course, even to the extent of working out a timeline: "ok, so they'll spend the first day messing around as novice guards, and meanwhile they'll meet and have an argument with this one other group of guards, and on the second and third days they'll leave the city and go to this town, and then they'll come back on the fourth day and find out they've been framed for murder". But the Nobilis campaign was almost entirely on-the-fly. I would usually have a plan for the session but there was no overarching storyline until very late in the campaign when I back-fitted one.

However you do it, the thing you have to plan out is how the players are going to get goals, and where the obstacles for those goals are going to come from. If you want a game focused around character interaction, then you should create a couple people with conflicting desires who all want the PCs to help them out. Goals and obstacles flow naturally from that kind of setup if you just play all the NPCs forcefully enough. If you want a game focused around exploration of an area, then you probably need to design the area in detail and then work out some motivation that will prompt the players to explore it. If you want a mystery game, then you need to work out the overall goal, and then enough subgoals for the players to be able to accomplish something as they're trying to puzzle out the mystery.

The last point about out-of-game planning is that a large chunk of it happens before any play occurs, but then between sessions you'll probably have to do more. The players will throw some curveballs, the story will go unexpected places, and you'll need to swap things around, discard things, or make up things to meet the new needs of the story. As long as you stay reasonably flexible it should all be fine.