A fellow who writes a column for my website has been having major difficulties writing one particular column in which he wanted to figure out how to write/run a scenario that (ideally) would not involve any combat whatsoever. He's been having lots of trouble coming up with something more detailed than "See what the players do, and figure out what happens," which I don't consider very good advice for a published campaign module (especially since the scenario so far doesn't contain enough information to figure out what happens). I wound up at one point offering the following advice (below in quotes).

I would love any comments, suggestions and criticisms of the advice, particularly on the following two points:
1) The advice sounds highly Gamist, but all scenarios need some sort of structure and conflict. What kind of changes would it need to support other CA's?
2) Do you think this is good advice for a reasonably novice GM with a D&D background? (I intentionally left out what I consider more advanced techniques - it's not meant to be a complete treatise on scenario design).
3) Anything else?

To preface it with some explanation: The scenario in question will focus on a non-combat situation. The characters will be hired (in some way) by a university to look for a criminal with a reputation for embezzling money from large institutions. The writer has established that the embezzler is a woman who seduces people to help her, is currently acting as a security guard, and has a hacker student at the university helping her. His thought for the scenario start would be to have her give players a tour of the university.

What I recommend at this point is to start thinking about this like a dungeon crawl. In a dungeon crawl, you have a standard format: players enter a new room, kill a monster, and take its stuff. Now, let's decompose this into its basic elements:

PLAYERS encounter a SITUATION in which they must overcome a CHALLENGE, for which they will be REWARDED.

I don't use the caps to yell, just to emphasize the key words.

So, for PLAYERS. Well, we have the group already defined, since this will fit into a campaign. We've got a reason for them to be there. We could still stand to work on a better explanation of why they're there, especially in terms of what they can bring to a university that it doesn't already have.

SITUATION needs to be defined. The scenario itself (catching the embezzler) is like the whole dungeon, so we need small situations that are equivalent to rooms. For example, you start off with a situation in which the embezzler (acting as a security guard) shows the players around the campus. That's a start.

So the situation needs a clear CHALLENGE and a clear REWARD. Why is she showing the players around? What do they stand to gain from it? The more clear you can be, the easier for a GM. For example "She will show the players around. At some point, a student will come up to her, start to say something and then stop himself and leave." The challenge here is mental - players have to recognize the clues and put them together (pretty trivial here except they don't yet know the embezzler is the guard). The reward here is the information that these two are connected.

Now, the difficult part. In a dungeon, everything is connected by doors and halls and no reason or rationale is necessary. (Otherwise, we'd have to explain how a dragon gets in and out through 10x10 corridors and why it lives right next to goblins, who for some reason stay in their own room doing nothing). In this kind of adventure, you have to string things together a little more logicially, but they don't have to connect perfectly.

For example:
PLAYERS: already defined
SITUATION: Fire breaks out in a building
CHALLENGE: Rescue people before they (and you) die from fire
REWARD: As they rescue people, they notice the security guard taking time to disconnect her computer to take it with her.

... which leads to...

SITUATION: Security guard has a computer, and has paid unusual attention to it.
CHALLENGE: Get access to the computer, despite guard's attempts to protect it (by using cryptware on the data and keeping physical possession of it).
REWARD: Bank account numbers with suspicious deposits

... which leads to...

SITUATION: A bank account has suspicious deposits.
CHALLENGE: Overcome bank security to get more information (by computer, by charm or by force)
REWARD: Bank account registered to ...

You see the pattern?

Now, at this point, we've built a railroad. Some players will be happy with that. It's certainly easier to build than something more complex, but I think role-playing can be more. So we could step it up a notch. For example, what if the fire was really a backup plan? Let's insert a situation ahead of that and give it a "PENALTY" which is what happens if the players fail the challenge. Note: in a typical dungeon setting, the penalty condition could be death, but it also might losing your armor to a rust monster, losing hit points to a trap or burning up more spells than you really needed to.

SITUATION: On a guard's tour of the campus, a student reacts strangely toward her/the players.
CHALLENGE: Spot the mistake and find the student
REWARD: Student has information.
PENALTY: No information gained. Nothing happens until a fire breaks out a week later.

Or perhaps an alternate penalty could be this:

PENALTY: As players look for the student, they draw attention. Embezzler tries to have them killed/fired/distracted.

And then...
SITUATION: Embezzler rigs up a drunken party
CHALLENGE: Players must respond to the party
REWARD: Some of the students may provide clues as to how the party started (and who started it and why)
PENALTY: Physical danger or injury.
PENALTY: If players kill or injure students, they may be disciplined by the university.

The second penalty there could lead to
SITUATION: Security guard testifies falsely to university discipline council
CHALLENGE: Prove her wrong.
REWARD: Charges dropped. Plus, she's revealed her hand (at least partly) by opposing the group)
PENALTY: Characters booted from the school. They may suspect the guard, but have lost the university's help.

Now we've at least got a branching rail road. If they find the kid, they get information which leads them to a different set of situations. The failure doesn't end the scenario (a common problem GMs have with information-based scenarios), it just requires them to get information in a more dangerous manner (either by the fire rescue or by dealing with whatever the embezzler throws at them). And, you can probably see how these situations may interconnect. If she doesn't suspect them now, she may suspect them later. So you can provide the GM with lots of these interconnected situations.

Note: in my published campaign books, you'll this structure at work using different terminology. Each Scene defines a specific situation. Each Act defines a group of closely-related scenes. And each Section groups together the Acts necessary to achieve some larger objective. Combine a series of Sections and you have a full campaign.

For a written scenario, I think this is about as good as you can get. In my own games, I often try to build the "dungeon" such that there are essentially an infinite number of interconnected situations (i.e. I let the players inform me of what they want to do, so they're picking the situation, which is the dungeon equivalent of a room - and chances are good that the players at least partly define the challenge and reward merely by picking the situation). The problem here is that not all GMs know what to do, and you can't exactly write a module that takes into account what the players will think of, though you could write a module that gives lots of options and a detailed setting (thus saving the GM the work of coming up with the "dungeon" and giving him a sense of likely situations and the necessary information to respond to players). For a simple monthly column, I think that's going to be overkill, especially since this scenario is just designed to highlight how a non-combat scenario can work.

I cannot emphasize strongly enough that you buy yourself a copy of "Dogs in the Vineyard" and read its instructions on how to set up a town. What you are asking for is explained in detail, at least as far as designing religion based conflicts in Old West mormonesque townships. But, the basic steps can be applied to a broad range of scenarios.

I can summarize the technique in this fashion:

1) Figure out who is doing something naughty and why are they doing it.

2) Figure out how those actions are affecting bystanders. The nastier the activities, the more of an injustice they will be to others.

3) Write up the NPC's in detail, both the villains and the bystanders, paying particular attention to their motives and relationships to one another. Stats and skills are optional, motives and relationships are not. You probably want to also focus on the particulars of how the bad guys are going about their bad-guy activities.

4) Figure out a way to hook the characters into the story. (i.e. 'A man staggers out of the shadows with a panicked look on his face. He cries weakly, "Keep this away from Belazar" as he shoves something into your hand before collapsing on the ground dead. You see a knife in his back...').

5) Stop. Just stop. Don't tell the GM where the story is going to go, because nobody knows where a story is going to go until it has gone there. In other words, "See what the players do, and figure out what happens" isn't such bad advice, provided the GM has suffient information to "figure out what happens" in response to the players' actions.

Keep in mind, a plot is made up of the villains and heroes taking rational actions based on the unfolding situation. You cannot know what actions the villains are going to take after the first scene because you cannot predict the actions that the heroes are going to take. Set up a scenario in sufficient detail so that the GM knows the kinds of things the villains can do along with what they are willing to do. Set it in motion. And, just leave it at that.

Don't assume that a novice GM with a D&D background is incapable of handling the role. Sure, they will be a little raw at first, but they'll catch on quickly enough. In fact, a novice may be able to handle such a setup more easily than an experienced Game Master, because he hasn't been brainwashed to think that a module must necessarily be set up as a dungeon crawl.

John Kirk wrote: 4) Figure out a way to hook the characters into the story. (i.e. 'A man staggers out of the shadows with a panicked look on his face. He cries weakly, "Keep this away from Belazar" as he shoves something into your hand before collapsing on the ground dead. You see a knife in his back...').

I thought the advice was great. It really helped explaining it kind of as a flow chart of events.

I think you would have to make it clear that the flow chart doesn't cover all possible events. Just like in a dungeon crawl the party might decide to bypass parts of the dungeon, flee from some challenges, or abandon it altogether. To give the DM the tools to deal with these situations in a primarily non combat game he has to know the details of the characters. This would be analogous to knowing the AC or HP of monsters.

You also have to make sure to tell the DM to throw in a red herring once in a while. :)

Just to add something to what Kirk is saying, is that if you are building the mystery above do not make the clue the reward in every scene.

The clue should be the sole goal of the scene, in this way you won't end up with an adventure runing to a dead end. Give the main clues out, then have the reward be exstra clues. Check out this thread for more on this.

John Kirk wrote: Don't assume that a novice GM with a D&D background is incapable of handling the role. Sure, they will be a little raw at first, but they'll catch on quickly enough. In fact, a novice may be able to handle such a setup more easily than an experienced Game Master, because he hasn't been brainwashed to think that a module must necessarily be set up as a dungeon crawl.

jdagna wrote: One of the key problems he had was a story in which nothing happened until finally it ended. (Basically, his three ideas for leads either revealed the whole plot in one roll or were missed entirely).

xect wrote:

jdagna wrote: One of the key problems he had was a story in which nothing happened until finally it ended. (Basically, his three ideas for leads either revealed the whole plot in one roll or were missed entirely).

Ah, the curse of investigation. It seems to me to be next to impossible to predict how long it'll take the characters to figure it out.

jdagna wrote: I would love any comments, suggestions and criticisms of the advice, particularly on the following two points:
1) The advice sounds highly Gamist, but all scenarios need some sort of structure and conflict. What kind of changes would it need to support other CA's?
2) Do you think this is good advice for a reasonably novice GM with a D&D background? (I intentionally left out what I consider more advanced techniques - it's not meant to be a complete treatise on scenario design).
3) Anything else?

I'm not sure what kind of critique you're looking for, but here goes:

1) it's probably better to have at least two ways to transition from scene to scene. In this example, if the the PC's don't think much of the security guard and don't "click" on her, the scenario is stuck: I can't see any other planned-out link to the next scene. That's inherently bad design IMO because it practically guarantees that the scenario plans will become useless, unless your players are extremely well-trained.

2) each link in the chain should be interesting in some way -- the fire is dramatic, but there's not much exciting about stealing someone's laptop or looking for embezzlement. Many player groups are going to be tapping their toes impatiently, wishing they could find the next clue in line so that something more interesting happens. If the plot has a boring-but-necessary part, I find it good practice to add an interesting subtext, a colorful NPC, a cool setting, or something similar to the scene.

3) I think the structure of Rewards and Penalties is likely to piss off and de-protagonize the players, unless carefully managed -- some of those penalties are really rough!

Robin Laws' book on GM'ing talks about this scenario design technique in a bit of detail, so it might be good to look at that to avoid repetition.

Thanks for the feedback guys!

It looks like the consensus is that my traditional scenario design advice from the core book is generally better than this approach, since most of you guys are recommending that approach instead. I've always leaned toward more of the background-based plots that let players define more of what they want to do. It certainly works better to handle the curveballs players inevitably throw you.

I think I'll just chalk this up as advice that clicked with that one particular person but stick with my usual plan for more general purposes.

jdagna wrote: It looks like the consensus is that my traditional scenario design advice from the core book is generally better than this approach, since most of you guys are recommending that approach instead. I've always leaned toward more of the background-based plots that let players define more of what they want to do. It certainly works better to handle the curveballs players inevitably throw you.

I've done a lot of scenario-writing in my time, although it's not my most cherished occupation. I'll try to share some thoughts on writing scenarios where combat is not essential.

1. Think elements, not drama.

2. Think interaction, not conflict.

3. Try to go with the flow, but don't let the flow overcome you.

1. ELEMENTS
The setting should be prepared with a lot of elements; characters, NPCs, locations, occations, tensions, popular issues, and a few preplanned scenes.

"Characters": the ones played by the players. These should have some goal (preferrably mutual), something to strive towards...
Example of "occations": a funeral, a ball, any kind of seremony or festivity.
Example of "tensions": racial, gender, generations, families, any kind of laden relationship between groups.
"Popular issues: a single sentence on each issue, made as a list for the players to use in in-game conversations. No hidden agenda, just the ordinary gossip about wheather, celebrities, recent and well known happenings, old grievances, etc.

2. INTERACTION
The method should focus on interaction with the players by using the elements; meeting the initiatives of the players, feeding their perception of the setting, creating an illusion of "independent" happenings in the setting, weaving the elements with inner logic and drive. It should also try to stimulate interaction between players.

Example of meeting initiatives: player say "I investigate". GM say "You find scary stuff".
Example of feeding: player say "I investigate". GM say "The formal licence to investigate any kind of crime is given by National Security".
Example of creating illusion: player say "I investigate". GM say "Another man seem to be asking the same questions..." or "You find some material indicating a tax fraud, but thats not what you're after..."
Example of inner logic and drive: GM say "The scary stuff is not the content of the documents, but the fact that they all have govermental letterheads". GM say "Two guys from National Security has asked about you". GM say "This private eye has worked this case for years, and he is sceptical to your involvement in it".

3. FLOW
Try to be in flow with the happenings of the game. Stay ahead, but don't jump to conclusions. Let the players concentrate on their own actions, not your storytelling. Try to explore the setting together with the players, as to make the game a "joint venture". Do not fear for your lack of imagination. If you need a break; take a break. If you need impulses; ask the players for proposals. Try to be a tool for the game (not an almighty God). Get in flow with the game!

OK, this is more than instructions about how to write a scenario without combat. The problem is; combat is commonly used as the great finale in the drama, and as such it is often given defining power over the dramatic structure, defining the ideal structure like this:
- Start
- Build up (with some twists)
- Finale
- End

You may use this without combat too, but you are better off initially if you place your faith with the theme and not with a specific dramatic structure, at least if you want to give some general advice on doing scenarios without combat. There is no ideal structure. You need to pick and choose your structure, and the way you apply it, based on theme, conflict, participants, genre, etc.

And you need to ask what game you are giving advice about. If the game in question has half its rules occupied with guns and how to use them, your are probably striving in vain.