The Alexandrian

Don’t prep plots, prep situations.

That’s a maxim I first started preaching on the Alexandrian back in 2009. And one of the key things I talked about in Don’t Prep Plots is that you want to focus your prep on developing toolboxes instead of contingencies. Prepping contingencies catches you in the Choose Your Own Adventure trap, where you waste a lot of time trying to second-guess your players and developing mutually contradictory material for every possible choice they might make.

I’ve seen a lot of GMs, both before and after I wrote Don’t Prep Plots, discover the virtues of this lesson. And what frequently happens is that they begin applying the lesson at the macro-level of their scenario design, but continue making the same old mistakes at the micro-level. This is ironic, because it’s actually the micro-level stuff that is frequently the biggest and most useless time sink.

I’ve been spending a lot of time with Eternal Lies lately, so I’m going to use it as an example of what I’m talking about.

SPOILERS FOR ETERNAL LIES

St. Luke's Hospital - Malta

In Malta, the PCs find a hospital where two boys are being held by cultists and fed a “treatment” regime which is actually making them sicker. The published adventure anticipates that the PCs might try to rescue the boys and this happens:

Leaving the Hospital Superbissima is not especially difficult if the PCs are just visitors or patients. An Investigator may use Medicine or Disguise (Difficulty 5) to pose as another Investigator’s doctor and authorize the transport of the character to another hospital. (This probably requires the authorizing Investigator to sign a few forms.) A bit of Reassurance convinces hospital clerks that a patient feels fine and is ready to depart. A Reassurance spend convinces a clerk to the point that he or she doesn’t even write a suspicious note in anyone’s file or form much of a memory of the character — good for anonymity, if they need it.

The real challenge to escaping from the hospital appears under two other circumstances: when Donovan’s guards are there, watching who comes and goes, and when the Investigators have patients with them that are under Dr. Solazzio’s “special care.”

(Note that Investigators in the hospital as patients can get dosed with Nectar or put under Dr. Solazzio’s care for two broad reasons: either the cult recognizes them as enemies and decides to make use of the PCs’ vulnerability, perhaps to draw out their comrades, or you as Keeper decide that Dr. Solazzio’s attentions fall on them for purely dramatic reasons such as for pacing or exposition.)

If the Investigators attempt to slip out in pursuit of Donovan or while his guards are in the hospital, see the scene Malta 4,”Pursuing Donovan,” p. 232, for stats on those guards. They take immediate note of strange behavior in their vicinity.

If the Investigators try to get Alexi or Monte (or both) out of the hospital, they have to get around some nurses and orderlies and go through the security staff. (They may even have to get through Donovan’s own guards, if they time their escape poorly.) This is not all that hard to do, really; it’s just hard to do anonymously. Waving guns around or making a total of a 2-point Intimidation spend is brutish enough to clear a gap through the hospital staff. Getting Alexi and/or Monte out of the hospital likely leads to a citywide manhunt for the boys and their “kidnappers,” with both crooked and legit police on the search for the Investigators.

Thus the Investigators may want to wear surgical masks or Disguises, since even a simple disguise (Difficulty 4 or 5) at least renders a character unidentifiable to witnesses who might be called upon, in this era without security cameras, to assert that “yes, that’s the one who carried that boy out of the hospital.” A surgical mask gives any character a dedicated pool point of Disguise within the hospital. A doctor’s white coast grants an additional pool point of Disguise. Either or both gives you a good reason for some NPCs to make lazy assumptions (“Just another surgeon, I guess”) or obstructive assumptions (“Excuse me, doctor, can you help me?”) about the character, depending on the success of the attempt. NPCs might challenge a disguise if the Investigators draw attention to themselves or interact with an NPC. Not all interactions call for the Difficulty 7 test befitting proper impersonation. In this context, it’s easier (Difficulty 5 or 6) to impersonate “a doctor” than it is to impersonate a particular Doctor (Difficulty 7). A Medicine spend might count as a point toward a Disguise roll if, for example, an Investigator wants to portray a visiting specialist (“Didn’t you get my telegram?”).

All disguises are temporary affairs, anyway, buying just enough time to take action, stymie the opposition, or delay consequences for recklessness or failure.

Once the Investigators get out onto the street, they must have somewhere to take the rescued boys (though see “The Knight”). Thus the situation gets more complicated. You may want to call for a Stealth or Shadowing test (Difficulty 5 close to the hospital, Difficulty 4 after that) to describe the Investigators’ attempted escape from the scene of their rescue. Alternately, they may try simply Fleeing the scene until they can make a single Stealth test to hide.

If the Investigators have devised a whole scheme for rescuing the boys and getting them free of the cult (perhaps involving fake papers and a ship out of Malta), let them explore it. If it proves to be too much of a distraction from the job at hand, gloss over details or assume that the characters succeed rather than testing for every damn task. Securing papers for the boys might just require a Law or Streetwise spend among the right contacts, for example, and getting the boys to safety might simply involve flying them out in the Winston-Rogers plane to Sicily and then sorting out the rest of it between Locales. If the logistics of a complicated rescue seem to be spoiling the players’ good feeling for doing the right thing, make things easier on the players (even if things stay complicated for the characters between scenes).

You can immediately see that there is a ton of verbiage being dedicated to specific plans that the PCs may or may not actually come up with. What I’m suggesting is that the prep for this should look something more like this:

ESCAPING THE HOSPITAL

There are two exits from the Hospital Superbissima: The front door and a side entrance used by employees. There are tall windows in the patient wards, but many of these are not designed to be opened.

NURSES: Collectively, the nurses of the Hospital Superbissima benefit from Awareness +2 to notice investigators snooping around areas they don’t have permission to be in or any other suspicious activity happening in the hospital.

HOSPITAL SECURITY: Security around the hospital is surprisingly tight. In addition to two guards in the lobby, a guard is also found at the nurse’s desk on each floor of the building.

[Insert a stat block for the hospital security guards here.]

DONOVAN’S GUARDS: If Donovan is on the premises, his guards are stationed at the entrance to the third-floor Intensive Care Ward and will respond immediately to anything they view as strange or suspicious.

[Repeat the stat block for Donovan’s guards here for easy reference.]

GETTING THE BOYS OUT: Getting Alexi and/or Monte out of the hospital, this likely leads to a citywide manhunt for the boys and their “kidnappers,” with both the crooked and legit police on the search for the investigators. Securing a place of refuge where people can’t spot the boys and report them may be difficult. Getting them out of the country may require securing (or falsifying) legal papers.

SIDEBAR: Remember that the Knight may already be watching the investigators at this point. If they get into trouble in the hospital, he’s likely to step in and help them out. Or at least offer them a place of refuge.

ON THE USE OF TOOLS

And that’s pretty much it. What’s the distinction here, beyond a greatly reduced word count (and, thus, work load)?

As the GM, note that if your players were to propose any of the escape plans proposed in the original text, you should be able to look at the tools provided in the second description and figure out what the result or response will be. More importantly, if the players propose some completely different plan (calling in a bomb threat, rappelling through the windows at night, taking hostages, casting a spell to escape with the kids to another dimension, etc.) you should also be able to pick up the tools (the layout of the hospital, the nurses, the security guards) and figure out what will happen.

While prepping the adventure, you don’t need to think to yourself, “What will happen if my players decide to disguise themselves as doctors? Well, I guess they’d make a Disguise check. I better write that down!” Not only because it’s self-evident, but because there are at least a half dozen other possibilities for what they might do. Whatever work you’re putting into trying to figure out the myriad tactics the PCs might employ in a particular tactical situation, you’d be far better off making your toolbox larger and more interesting.

PREPPING YOUR TOOLBOX

But what constitutes the difference between a tool and a contingency? For example, isn’t prepping this information about the hospital dependent on the PCs going to hospital? Doesn’t that make the whole thing a contingency? And even if we lay that concern aside, how do we actually identify what the useful tools are? I mean, the bad guy might have a collection of early Picasso paintings. How do I know whether to prep the Picasso collection or the alarm system on his windows?

I find this generally boils down to two questions.

First: What will the PCs be interested in? Not what they’re going to do, but what they’ll be interested in. Where their focus will be. (In a mystery scenario this is relatively easy to predict because it usually equates to wherever the clues are pointing them.) It’s possible to get pedantic and argue that “being interested” constitutes an action, but I think the distinction being sought here is generally pretty clear.

Second: What are the NPCs’ plans? These can be specific to the events of the scenario (“they want to blow up Woodheim”) and possibly even specific to the PCs (but not specific to the actions of the PCs) if their current plan is aimed at the PCs (like Lex Luthor obtaining kryptonite to deal with Superman). But most of the plans you’ll be looking at will actually be a lot more general and long-standing than that.

The hospital scenario is an example of this: The cult doesn’t want the boys to escape. What precautions are they taking to prevent that from happening?

Similarly, if you were designing a mansion for a mob boss you’d ask questions like: What type of security system does he have? What does he enjoy doing at home? What does his daily schedule look like?

If the mob boss is currently planning to assassinate the head of the local Triads, then you’d start asking yourself questions like: Who does he hire to do that? If things go bad, what resources does he have to protect himself?

All of these questions will guide you towards creating either the long-standing status quo or the current trajectory of action that the PCs are going to be thrusting themselves into. And what you want to focus on is that situation which exists without the PCs and let your players worry about the thrust.

Which, I suppose, ultimately brings us full circle:

Don’t prep plots, prep situations.

FURTHER READING
The Railroading Manifesto
Node-Based Scenario Design
Gamemastery 101

Share on TumblrTweet about this on TwitterShare on StumbleUponShare on FacebookShare on RedditShare on Google+Digg this

9 Responses to “Don’t Prep Plots – Tools, Not Contingencies”

  1. Jan says:

    It’s so hard reading the original example and the “tooled“ version. It reminds me of all the time when I spent HOURS reading long and unstructured texts and leafed through pages of awfully distributed information. It could have been so easy to make it “tooled“ and saving me SO MUCH time. Also I found it very strange that authors are making assumptions when don’t know a slightest bit about the players, characters not the gm. It’s hard enough to do this with your regulars, but add an authors of a piece to be sold its insane… I sincerely hope that adventures will be written in a new way soon.

  2. kelvingreen says:

    This is the big problem with Eternal Lies and what made me give up on it and move on to other things; whenever possible it will use 400 words when 40 would do.

  3. Justin Alexander says:

    Re-reading this little article now that it’s gone live leaves me a little unsatisfied. The problem is that, for the sake of conciseness, I’m only showing one slice of the hospital’s description (the precautions taken to prevent the boys from escaping), and this ends up carrying with it the connotation that I’m prepping a toolbox that can only be used for that one thing. Hopefully the nature of that narrow slice is clear enough that it’s not needlessly confusing the tool/contingency distinction.

    There should also be more emphasis on the fact that the difference in length is not the key virtue. You could write up notes that were just as long and contained just as many details as the original version as long as all of those details were tools and not contingencies. (And the result would be a much richer scenario with a lot more usable depth.)

    Also tempted to go back and add in a section showing what these notes would look like in my actual prep notes (instead of as polished paragraphs). For example, I’d write things like:

    NURSES: +2 Awareness collectively

    GUARDS:
    – 2 in lobby
    – 1 on each floor @ nurse’s desk

    Which is more than sufficient for recording my intentions when the only person I’m communicating with is myself.

    @kelvingreen: It’s why I’ve found myself less and less willing to use published material over the past few years. I just literally cannot run material effectively out of most scenarios published in the last 10-15 years: The useful information is just buried inside massive amounts of text that’s been narratively-organized instead of utility-organized. It takes a very high quality of material for me to want to put in the effort to put together game-ready prep notes.

  4. thekelvingreen says:

    Yes, I was hoping that books like Vornheim would bring about a revolution in adventure design, where utility would be the foremost concern, but it seems that no one got the hint.

  5. Justin Alexander says:

    The problem is that there’s a reputedly sizable portion of the audience who are buying adventure products in order to read them as pulp fiction.

    And while i can understand the appeal of that (there’s a free-form and imaginative version of Choose Your Own Adventure that happens in the interstice of closure during the reading of a really great adventure scenario), I wish that the scenario writers weren’t beginning to cater to that audience to the exclusion of people who actually want to use their products at the gaming table.

  6. Brotherwilli says:

    Have you had the opportunity to review Monte Cook’s new format used for some Numenera adventures? I haven’t had a chance to run any yet, but on review they seem to genuinely be polished adventure notes.

  7. Gamosopher says:

    It’s probably not a coincidence, but that reminds me a lot of open world game preparation. What you describe is basically how Stars Without Numbers advise GMs to prep their sandbox. It also reminds me of Jason Lutes’ Servants of the Cinder Queen adventure (and the advices he gives in his just-kickstarted The Perilous Wilds).

    Everyone agrees that a pre-written adventure is there to reduce the workload of the GM, but maybe there is a confusion about how a written adventure is supposed to help : it’s there to reduce the time the GM spends prepping its game before play, not to reduce the work the GM does to make interesting stuff happens during play. Prepping tools reduce the prep by giving the GM what he needs to make up stuff on the spot during the game, while prepping contingencies gives ready-made interesting stuff that will happen in play (well, the illusion of it.)

  8. Justin Alexander says:

    “Not to reduce the work the GM does to make interesting stuff happen during play.” That’s just spot-on, Gamosopher.

    @Brotherwilli: I’m waiting for my copy of that Numenera adventure collection to arrive, actually. I looked at the free preview a couple months ago, and what I saw was very promising. And the effort to make published adventure notes so accessible that you can run them the same way you can run a board game is, obviously, something I support 150%.

  9. d47 says:

    Justin wrote: “The problem is that there’s a reputedly sizable portion of the audience who are buying adventure products in order to read them as pulp fiction.”

    I definitely fall into the category of people who read adventures more than use them, but I still prefer your tool format to the overly prescriptive example. I suppose that kind of detail could be useful for beginner GMs, but it is boring reading.

Leave a Reply

Archives

Twitter

Recent Posts


Recent Comments