As a disclaimer, this post is as much for my own future reference as it is for anyone who might be reading (but don't let that stop you...).
Sean and I talk about this topic quite a bit, particularly since he's still relatively new to DMing. How to prepare for a D&D session is something you learn and refine through practice and experience over years of running campaigns. There's no magic formula, and what works for one person may not work for anybody else. It's like studying for an exam in that everyone's brain processes and retains information differently, so you have to figure out through experimentation what works best for you. That said, while there aren't any empirical rules, I can still share and document my approach.
Two axes: geographical and temporal
As the prospective DM, whether you're kicking off the inaugural session of a new campaign or continuing to run a party with many notches in its belt, the idea of being accountable to an entire fantasy world can feel extremely daunting. Where do you begin? Do you start by fleshing out local governments, agencies, or dungeons that exist across the countryside, or notable NPCs throughout the region? If you're running a homebrew world, the climb can seem particularly steep, but even in an established setting with a wealth of published material like Forgotten Realms, you may feel like the game world holds you to an impossible standard. (In that case, the first thing to realize is that using a published world should never hurt you more than it helps, and if that's happening you may do well to reconsider your choice. Further musings on that here.)
To the question of where to begin your prep work, my unequivocal answer is "here and now." The first thing I need to do when faced with a massive decision tree is prune away as many branches as possible to get to what I care about. No one realistically has time to detail every last person, building, and statistic in a game world, so focus on those that carry the most imminent importance. To wit: I don't need to know what will happen in the party's current location six months from now; I also don't need to know what will happen a hundred miles away tomorrow.
These simple assertions drastically reduce the complexity of the problem. I'm not running a campaign where the long-term outcome is already determined, so events unrelated to the party's current time frame or location are of little concern to my next several hours of DMing. This doesn't mean that I completely eschew what may be happening in other areas, but that I need only have a vague notion of those details, and shouldn't make them the focus of my attention since they're unlikely to impact the party over the short term. Having the PCs overhear in a crowded taproom that trolls have begun encroaching on the north border and nearby villages are petitioning for aid is likely to be enough. It's conceivable that the party could take interest in this hook, so I definitely need to be able to convey anything else the local citizenry might know, in addition to accommodating an initiative by the party to set out for the border. That said, I shouldn't prioritize fleshing out this region unless I fully expect the players to take the lead and arrive at the destination. More likely, I'll end up doing this planning several sessions down the road, if ever.
Rather than dive into rabbit holes that may consume hours of planning that I won't use, I want to concern myself most with happenings around the party in an immediate sense, geographically and temporally. If the session is set to open in a town, castle, or dungeon, I need to be well-versed in the ecology of the site: what individuals or populations reside there, how they interact, what sustains/motivates them, and any pressing urgencies to which they're attending. Ecology feels like a greatly undervalued concept in preparing a game; often, too much focus is given to drawing up specific scenes or encounters envisioned in the mind of the DM. While there are instances where this is useful or appropriate, if I truly understand the ecology of the actors in the game world, I should be reasonably well-equipped to adjudicate any situation that arises during the session, even if I need to ad-lib a few names or stop to roll an NPC's hit points or ability scores. If there are individuals for which I suspect these numbers will be needed, I'll roll them up ahead of time.
The further you look down the party's intended path, the fewer details you need to know. If the PCs are embarking on a journey that leads them through a series of towns and villages, I should know the ecology of each settlement at least at a macro level, but I definitely need to more thoroughly immerse myself in the locations the party will spend time in first. Any substantial event, encounter, or distraction has the possibility, based on decisions by the players and whims of the dice, to sidetrack or significantly delay the party's course. I might put a great deal of work into readying myself to run the first village along the road, but the further out I look, the hazier my view becomes. While I might be able to make a confident assumption that the party will arrive in the first village and take part in its dealings, my confidence wanes considerably with each subsequent hop along the journey. The foggier my take on the party's future, the fewer specifics I prepare. This approach serves me in multiple ways: first, I avoid putting time and effort into preparation that has a higher chance of not being needed, and second, because I haven't invested this effort, I feel less compelled, consciously or otherwise, to force the PCs down a path I prepared, giving them more freedom of choice and control over their destiny. When it comes to actually running the game, I want to be the unbiased referee, not the puppet master pulling the strings. Had I spent hours writing up details on the north border region with the trolls, I'd be more likely to artificially steer to party toward taking the hook, when they may not have any real interest in pursuing.
Predetermined vs. triggered events
In addition to interactions between people and locations, I make sure to plan through any predetermined events that I've "scheduled" to occur irrespective of the party's actions. Ideally, these events have been logically derived from the ecology of the area. For example, if I know that orcs are plotting an assault against a local village, I should determine, before the session begins, when the attack will occur, how many orcs are invading, what the attackers' entry points and strategy will be, and so forth. The date and time can be chosen outright or be assigned a degree of variance (e.g., 1d4 nights after the next full moon, when the orc chieftain gathers his forces on the mountainside). Wherever the PCs happen to be at this time, the raid will be executed, unless they do something in the interim to subvert the orcs. Preparing for events and allowing them to play out regardless of the party's actions helps create realism in the campaign world. Not everything needs to specifically revolve around the characters, even though I might only plan in detail the events in their vicinity.
On the opposite end of the spectrum are "triggered" events, which are common in published modules, especially in manipulative settings like Ravenloft. Triggered events should be used sparingly in any game that doesn't want to predestine the characters' path; examples are akin to "The innkeeper is kidnapped the night after the PCs investigate the cemetery," or "Whenever the PCs walk past the town constable, they overhear him negotiating the assassination of the mayor." These types of events can be exciting, but they also make assumptions about how the characters must act in order to fulfill the DM's narrative. This approach to running a game isn't strictly right or wrong, but comes with strings attached that steer the campaign toward an outcome largely not determined by the players, and this isn't something that's desirable for every group.
Not exactly "points of light"
Finally, trimming decision tree branches and preparing in less detail the further out you go is similar to, but not the same as, the idea of "points of light" (formalized as a setting concept in D&D 4e), which asserts that there exists a collection of known, civilized areas and that everything in between is mostly uncharted or "dark." While the planning approach I've outlined tends to evoke the feel that the party's immediate surroundings are a "point of light" amid the darkness, this is only necessarily true in terms of your own planning as DM. The regions and timelines outside your short-term scope of interest may be extremely well defined in source material or to the inhabitants of the game world. It's only that you, the DM, don't need to concern yourself with details beyond what's relevant to the party given its present course.
I think I'll end here for now, as this post is lengthening and mostly a result of stream of consciousness writing. Happy to read anyone's comments.
Sean and I talk about this topic quite a bit, particularly since he's still relatively new to DMing. How to prepare for a D&D session is something you learn and refine through practice and experience over years of running campaigns. There's no magic formula, and what works for one person may not work for anybody else. It's like studying for an exam in that everyone's brain processes and retains information differently, so you have to figure out through experimentation what works best for you. That said, while there aren't any empirical rules, I can still share and document my approach.
Two axes: geographical and temporal
As the prospective DM, whether you're kicking off the inaugural session of a new campaign or continuing to run a party with many notches in its belt, the idea of being accountable to an entire fantasy world can feel extremely daunting. Where do you begin? Do you start by fleshing out local governments, agencies, or dungeons that exist across the countryside, or notable NPCs throughout the region? If you're running a homebrew world, the climb can seem particularly steep, but even in an established setting with a wealth of published material like Forgotten Realms, you may feel like the game world holds you to an impossible standard. (In that case, the first thing to realize is that using a published world should never hurt you more than it helps, and if that's happening you may do well to reconsider your choice. Further musings on that here.)
To the question of where to begin your prep work, my unequivocal answer is "here and now." The first thing I need to do when faced with a massive decision tree is prune away as many branches as possible to get to what I care about. No one realistically has time to detail every last person, building, and statistic in a game world, so focus on those that carry the most imminent importance. To wit: I don't need to know what will happen in the party's current location six months from now; I also don't need to know what will happen a hundred miles away tomorrow.
These simple assertions drastically reduce the complexity of the problem. I'm not running a campaign where the long-term outcome is already determined, so events unrelated to the party's current time frame or location are of little concern to my next several hours of DMing. This doesn't mean that I completely eschew what may be happening in other areas, but that I need only have a vague notion of those details, and shouldn't make them the focus of my attention since they're unlikely to impact the party over the short term. Having the PCs overhear in a crowded taproom that trolls have begun encroaching on the north border and nearby villages are petitioning for aid is likely to be enough. It's conceivable that the party could take interest in this hook, so I definitely need to be able to convey anything else the local citizenry might know, in addition to accommodating an initiative by the party to set out for the border. That said, I shouldn't prioritize fleshing out this region unless I fully expect the players to take the lead and arrive at the destination. More likely, I'll end up doing this planning several sessions down the road, if ever.
Rather than dive into rabbit holes that may consume hours of planning that I won't use, I want to concern myself most with happenings around the party in an immediate sense, geographically and temporally. If the session is set to open in a town, castle, or dungeon, I need to be well-versed in the ecology of the site: what individuals or populations reside there, how they interact, what sustains/motivates them, and any pressing urgencies to which they're attending. Ecology feels like a greatly undervalued concept in preparing a game; often, too much focus is given to drawing up specific scenes or encounters envisioned in the mind of the DM. While there are instances where this is useful or appropriate, if I truly understand the ecology of the actors in the game world, I should be reasonably well-equipped to adjudicate any situation that arises during the session, even if I need to ad-lib a few names or stop to roll an NPC's hit points or ability scores. If there are individuals for which I suspect these numbers will be needed, I'll roll them up ahead of time.
The further you look down the party's intended path, the fewer details you need to know. If the PCs are embarking on a journey that leads them through a series of towns and villages, I should know the ecology of each settlement at least at a macro level, but I definitely need to more thoroughly immerse myself in the locations the party will spend time in first. Any substantial event, encounter, or distraction has the possibility, based on decisions by the players and whims of the dice, to sidetrack or significantly delay the party's course. I might put a great deal of work into readying myself to run the first village along the road, but the further out I look, the hazier my view becomes. While I might be able to make a confident assumption that the party will arrive in the first village and take part in its dealings, my confidence wanes considerably with each subsequent hop along the journey. The foggier my take on the party's future, the fewer specifics I prepare. This approach serves me in multiple ways: first, I avoid putting time and effort into preparation that has a higher chance of not being needed, and second, because I haven't invested this effort, I feel less compelled, consciously or otherwise, to force the PCs down a path I prepared, giving them more freedom of choice and control over their destiny. When it comes to actually running the game, I want to be the unbiased referee, not the puppet master pulling the strings. Had I spent hours writing up details on the north border region with the trolls, I'd be more likely to artificially steer to party toward taking the hook, when they may not have any real interest in pursuing.
Predetermined vs. triggered events
In addition to interactions between people and locations, I make sure to plan through any predetermined events that I've "scheduled" to occur irrespective of the party's actions. Ideally, these events have been logically derived from the ecology of the area. For example, if I know that orcs are plotting an assault against a local village, I should determine, before the session begins, when the attack will occur, how many orcs are invading, what the attackers' entry points and strategy will be, and so forth. The date and time can be chosen outright or be assigned a degree of variance (e.g., 1d4 nights after the next full moon, when the orc chieftain gathers his forces on the mountainside). Wherever the PCs happen to be at this time, the raid will be executed, unless they do something in the interim to subvert the orcs. Preparing for events and allowing them to play out regardless of the party's actions helps create realism in the campaign world. Not everything needs to specifically revolve around the characters, even though I might only plan in detail the events in their vicinity.
On the opposite end of the spectrum are "triggered" events, which are common in published modules, especially in manipulative settings like Ravenloft. Triggered events should be used sparingly in any game that doesn't want to predestine the characters' path; examples are akin to "The innkeeper is kidnapped the night after the PCs investigate the cemetery," or "Whenever the PCs walk past the town constable, they overhear him negotiating the assassination of the mayor." These types of events can be exciting, but they also make assumptions about how the characters must act in order to fulfill the DM's narrative. This approach to running a game isn't strictly right or wrong, but comes with strings attached that steer the campaign toward an outcome largely not determined by the players, and this isn't something that's desirable for every group.
Not exactly "points of light"
Finally, trimming decision tree branches and preparing in less detail the further out you go is similar to, but not the same as, the idea of "points of light" (formalized as a setting concept in D&D 4e), which asserts that there exists a collection of known, civilized areas and that everything in between is mostly uncharted or "dark." While the planning approach I've outlined tends to evoke the feel that the party's immediate surroundings are a "point of light" amid the darkness, this is only necessarily true in terms of your own planning as DM. The regions and timelines outside your short-term scope of interest may be extremely well defined in source material or to the inhabitants of the game world. It's only that you, the DM, don't need to concern yourself with details beyond what's relevant to the party given its present course.
I think I'll end here for now, as this post is lengthening and mostly a result of stream of consciousness writing. Happy to read anyone's comments.
This is a collection of good advice, and I take a similar approach in a lot of ways.
ReplyDeleteYour comments regarding ecology are spot on. Too often the details are glossed over to get to the "good stuff" (which too often means combat), but there is a wealth of roleplaying opportunity in event the smallest details. Something as simple as a claw mark on a cavern wall, even if it was just meant to be dressing, can completely turn the course of a campaign with the right players, when presented the right way. Matt does a good job at not dispelling cosmetic dressing for what it is--if we want to spend a half hour mulling over some detail he knows to be minor but we find some potential significance in, he lets our imaginations run.
At the same time, there's a point at which there's too much focus on that kind of thing...that's a difficult balance to find as a DM, and comes with experience, I think.
For me, one thing I do is roleplay out potential conversations with notable NPCs in advance of the session. This helps me assign the characters a voice--not always a literal voice, but style of speech. Are they cagey when asked direct questions, gruff, seductive...I'll respond to expected questions in first person as the NPC, and start to form pathways so that the conversations play out more naturally in game.
Great stuff, though.
Thanks, good advice on your part as well. In truth, I feel like this post only scratches the surface of a very deep topic. I might try to write more on this, someday, but for now I wanted to put down thoughts that were ruminating while getting ready for this week's game.
Delete