How much do you prepare your adventures, and how good is your "improvisation?"

[EP]

So what makes for a good "improvisational" GM? In my experience I've found that it really depends on...well, experience. If you're really into the settings and rules, and you're really warmed up from regular GM-ing, then it's easier to improvise than if you only run a game every once in a while. I think of it this way--GM-ing is a skill which must be practiced regularly, and it comes more naturally to those who do so. Those who don't--or, in my case, can't--practice regularly benefit from lots of preparation.

If you don't GM often, your use of the rules can vary how well you improvise as well. During the game, I don't use heavy doses of rules and just follow a "common sense" approach to the game. Feats and spells must function as they are described and it's up to the players to keep track of that for themselves, but combat is pretty much run as "If you want to leap down from the top of the landing and bring your sword down on their head, make a DC 15 Jump check and make your attack with a +2 bonus for having higher ground. If you fail the Jump check, you can't attack because your jump sucks." I know there is a specific bonus to having higher ground in combat, but I never bother to look it up because I don't run a game very often and turning pages to find the exact number slows the pace.

 

[Dragon Snack]

It's a +1 bonus for attacking from higher ground, PHB pg 151, Table 8-5...

My players can smell preparation and they don't like it. If I have something prepared they always seem to head in the opposite direction, hence I've given up on preparing stuff. I do think I'm better when I'm prepared, I just can't afford to waste my time on things I'm not going to use. About the only thing I prepare is the "boss" monster/BBEG.

I'm pretty good at improvising now, sometimes they can't even tell the encounter is completely random unless they see me looking up the encounter tables. Sometimes they think I planned an encounter even when they've seen me use the encounter tables, since I've gotten pretty good at BSing a "reason" for the random encounter to be in the area.

 

[Peni Griffin]

I absolutely, positively have to overprepare in order to have sufficient confidence to sit in the chair. I know I'm going to overlook something obvious, I know the players are going to surprise me, I know I'm working too hard - but I have to do it. The process of working out where all the NPCs are, what they're doing, what their motivations are, how long it takes to get places, working up tailored random encounter tables for specific areas, figuring out the weather, etc., grounds me sufficiently in the story and setting that, when the play happens, I don't look at the notes often and react properly on the fly. All this means that I have to use modules, as doing all this from scratch would mean about a year between sessions. This is identical to my method when writing a historical or prehistorical story - I read up on relevant information, checking out stacks of books and writing copious illegible, poorly-organized notes that I seldom refer to when writing, because the information is composted in my head and the story grows naturally out of it.

Frustratingly, no matter how hard I overprepare, there's always something I don't put into my notes that obviously needs to be there. When a module refers to a Monster Manual page, for example, in order to keep from having a half-dozen books open, or flipping back and forth in the same book, I type up the stat blocks. Ideally, this fixes the monster's capabilities in my brain so I don't have to refer to the block in order to decide (for example) whether it is reckless, cautious, vicious, or a good team player when it confronts the PCs. But there's always at least one monster that doesn't get all his capacities copied, like the infamous time I forgot to note down that barghests have DR and the party's horses kicked them to death. But if I didn't do the overpreparing, I'd still overlook things like that during the game, would take longer about it, and combats would take even longer than they do now.

A lot of the changes in our DMing styles we see as we get older are due to physical changes in our brains as we age. Our brains are in a state of perpetual growth until our early twenties, when our temporal lobes finally finish growing in. From that point, we spend most of our learning time reinforcing certain connections and neglecting others. The result is that, though we get better and better at recognizing patterns and dealing with familiar types of problems, we get worse and worse at dealing with novel situations. Some DMs get worse at winging it because they get less creative, or because the patterns that are most fixed in their brains don't apply as well in a gaming context (I think this is the biggest problem with railroad DMs - they keep trying to apply the same pattern and can't recognize when it's become counterproductive), while others get better because they have reinforced game-related connections to the point that they can see the underlying structure in every situation. They understand the system well enough and have seen enough that nothing surprises them any more.

You have to understand your own strengths and weaknesses in order to produce a good game. Young DMs, with brains still involved in reinventing the wheel and improving it because they don't know what a wheel is supposed to be like, may be able to wing games entirely because they wing everything in their lives, but as we get older, we lose this capacity outside our areas of expertise. If you have the rules down cold but don't understand story structure or character motivation well, you might have to do copious background preparation but get by without any stat blocks at all. DMs who have devoted large amounts of their liesure time to gaming may get better and better at improvisation because they have laid such an extensive groundwork of synapses, but the more you do this in one system the harder it will be to transfer your skills to a different one. Once you understand why it's hard, you can compensate.

 

[EP]

This is why drama class in high school is sooo important for any aspiring GM, especially a class that has improv. Personally, I'm a big fan of improv already, and improv GMing is the next best thing (or the best, depending on how the game goes).

Less preparation also helps keep the game "real," especially if the NPCs don't know the power of your players. That's the main reason why I never plan out anything from the NPCs unless they have been able to study the PCs and gauge an appropriate response to defeat them - like it was said before, the players can smell planning and will always try to thwart it.

 

[DragonLancer]

My improvisational skills are not all they could be. I tend to prepare for about a couple hours each week for our sunday night game, whether reading or making notes, so I would guess that I'm more a "be prepared" style DM over a freeform one.

 

[I'm A Banana]

I pretty much run totally on auto-pilot. I may get an idea for a villain or a setting or a scene that I write down and try to steer towards, and sometimes I'll prep for things that I want to be able to access (ferinstance, I'll prepare certain archetypal NPC types so that I can conjure one up at a moment's notice if need be) if they don't already exist, but everything is meant to facilitate me being able to pull good stuff out of my butt at a moment's notice.

So what makes for a good "improvisational" GM? In my experience I've found that it really depends on...well, experience. If you're really into the settings and rules, and you're really warmed up from regular GM-ing, then it's easier to improvise than if you only run a game every once in a while. I think of it this way--GM-ing is a skill which must be practiced regularly, and it comes more naturally to those who do so. Those who don't--or, in my case, can't--practice regularly benefit from lots of preparation.

There's a few tricks to running games that are heavier in improv that anyone with the inclination can adopt.

Archetypes are your Friend. We've all consumed enough movies and books and TV shows to know how some character types behave. It doesn't take much of a stretch to put that into a D&D game. If the sullen warrior of amazing skill feels insulted, he's taking his magic doodads and going home. If the kindly old priest sees someone in trouble, he'll try to help them, no matter how outmatched he may be. If the dangerous woman notices someone she can manipulate, she'll lure them in with promises of everything and only deliver them death. It pays to work against archetype maybe 1/4 times, to keep 'em guessing, but you can go a long way with these parallels.

Let the Players Drive It. You don't need an answer to a question until it is asked, and once it is asked, you have to make a decision between it being common character knowledge (and then just telling the players the answer that makes sense to you at the time) or not (forcing the characters to research or go ask the right person). At the same time, your characters need to have goals and dreams and hopes that they will pursue.

Anything can spin off into a quest, not everything should. The basic rule is, "If the PC's want something out of the ordinary, they need to work for it." Think of the types of quests in a game like World of Warcraft: you can go slay x number of creatures, y specific creature, collect z number of items, a specific item, etc. All of these basically share the type of "NPC needs something done in dangerous territory." If your players seek something important, tell them they have to go prove themselves, or that the NPC needs special unguents, or that research cannot proceed until some ancient family enemy is slain. In this way, you can "nest" quests, too: To give them the information, the NPC is going to need a ghost haunting his family's old mansion slain. To slay the ghost, the PC's need to research the family history at the library. But they're not citizens, so they can't get in. How do they solve this problem? It's not good to ALWAYS do this (because PC's get frustrated if success is denied for too long), but it lets you pull an adventure out of anything.

Ask questions, only give answers if you have to. When the PC's have goals, they are the ones who need to accomplish them. A PC tells you "I would like to be the greatest warrior in the land!" you ask them "How?" they tell you "I'll slay the current greatest warrior in the land!" From that line, you should have half a million archetypes running through your head already: the arrogant swashbuckler, the unbeatable tank, the mysterious swordsman, etc. Any of these characters could be the current Greatest Warrior. The PC wants to fight them. You need to make that happen: the mysterious swordsman travels alone, so the PC needs to do some reasearch to find him, go into dangerous territory, perhaps find a magical sword that will aid the combat, etc., etc.,...all of this just by asking "How?" Be prepared to prod in one direction or another if they ask you "How do I?" Again, archetypes work as your friend: think of those who have become the greatest warriors before in things that you've read or seen, of how they got there...and think of how to twist them.

Abstract in general, specific in specific. Specificity goes a long way toward creating a believable world, but it's not needed very often. A little goes a long way, and can actually give you further ideas. For instance, you describe a villages houses as "Old. Dilapidated. One has a hole in the roof, another lacks one of the four walls. Rather than being in construction, it looks like they have sat around untended for years." Randomly conjuring that specificity at any random village gives you countless hooks ("why is it in such disrepair? Don't the people care about their buildings? Is something destroying them?"), any of which the characters can explore, if it suits them (or if an NPC needs it). But you don't need to describe the quality of every building, and it's perfectly acceptable to have "Typical Commoner Dwelling" as the norm. If it deviates from that, the PC's may consider it important, but if EVERYTHING is important, then you always mention more than they can explore. Which is good -- it gives them the sense that the world around them isn't ABOUT them. And if the characters never learn what did that, that's fine -- something they don't show interest in is something you don't need to think about, either.

There's more. There could be a whole thread about good improv DMing techniques!

 

[TheCrazyMuffinMan]

I improv nearly every run I DM. I have one fixed setting (Stagnetia), and I improv most of everything else even THEN.

 

[adwyn]

I am poor at organization in the first place so spontaneous is just natural for me. Yet I find some planning does help, so I pre prep encounters with no idea of when and where I'll use them. Likewise I have made up several pre planned locations, each a variation from the run of the mill, that I can plug in as needed.

Thus when the PCs chase some mooks into a building I can grab a preplanned group of thugs and then grab whatever location sheet sounds exciting and appropriate - last week the thugs turned out to be were rats and the location was a halfling industrial bakery with low ceilings, dozens of boiling pots, and many hot ovens. A memorable encounter that just all came together with no real planning.

 

[ivocaliban]

I compare it to cooking. I don't necessarily need a recipe, but I do like to have all my ingredients and supplies in front of me before I begin.

I've spent hours planning out and writing detailed adventures in the past and more often than not they don't happen anything like they're written. Furthermore, some of the best moments in my games have happened when a player did something totally unexpected and I had to improvise anyway. Eventually, it got to the point where they players couldn't tell whether I'd planned an adventure or was playing it by ear.

That said, I still like to have a general outline of what's going to take place. It may be something that takes no more than five or ten minutes to jot down. If there's going to be combat I like to jot down abbreviated stats for common foes and have those close at hand. Once I have those things, I can pretty much handle whatever comes my way.

 

[Graybeard]

I have several NPCs and creatures prepared in advance along with a general outline of the plot as it seems to be going. Other than that, I go with the flow. If the PCs want to go a different way than I had prepared in the plot, then so be it. I will improvise a small side adventure and tie it into the big picture in subtle ways. My group is fine with what I do as long as there is combat for those that want it and a story for the rest of the group.