Thursday 15 November 2012

Cthulubox and the Investigative/Quasi-Railroad Problem

I've been doing some thinking about how you might run an investigative game - Call of Cthulu being the paradigm example - in a sandbox style. The problem with that sort of game is that it is heavily reliant on the players responding to events rather than guiding them by themselves. Something, or somethings, have to happen for them to investigate, and by definition that means that the GM has to spend a lot of setting up quasi-railroads for them to follow.

I say "quasi-railroads" because providing players with something to do ("A young woman mysteriously disappeared last night and her mother asks you to investigate!") is not a railroad per se, but leads down a certain path which has only two outcomes - success or failure in finding out what happened. How you get from A to Z is not set, but you'll probably end up getting there in the end. And by the same token, you can create a huge rumour table all you like - a d100 list of strange occurrences for the PCs to delve into - but at the end of the day, as GM, you have to know where those strange occurrences lead: all you've done is create a list of 100 quasi-railroads, ultimately.

I don't think there is a way around the investigative/quasi-railroad problem - it is probably at its most pernicious if you want to run a police procedural, but it still means that any Call of Cthulu campaign is likewise going to be heavily GM-led, simply by dint of its nature.

One possible way to make a Call of Cthulu game more of a sandbox would be to subtly shift the starting assumption and make sure that the players begin with an understanding that their characters are deliberate seekers of dark powers and magicks who already have some idea that there is a wealth of forbidden knowledge out there - if they could only get their hands on it. Their investigations, in other words, would not be into mysterious events that the GM tosses their way, but an exercise in finding where all of this knowledge lies. They might start off in Providence, Rhode Island in 1922 with a web of contacts, and locations, which they can tap as resources in conducting their own self-centered investigations - my character has heard that it is possible to live forever, and now I'm going to find out how - and this would allow them to drive things along more autonomously.

This would require extensive preparation on the part of the GM pre-campaign - akin to the way I set up urban sandbox games in general (see entries passim here and here), only more so - and also a good, skilled GM who is able to think things up on the fly. As such it may be more trouble than it is worth, though I think it's something I could happily give a try to see what happens.

34 comments:

  1. It seems to me that you're starting with a confused premise here: how does one make investigation into exploration? As you acknowledge in the post, you really can't; the two are different modes. An investigation has a specific goal which players will either achieve or fail to achieve, and that goal is offered by the GM. The players are automatically on something of a railroad (and if they've chosen to play an investigative game, they probably want that). Exploration is more open-ended: the players are discovering what's out there and are free to react to what they discover as they like. I suppose an investigation could become more open-ended if, while pursuing the original goal, the players ran across something they found more interesting and thence redefined their goal, but there are problems with this scenario. Either the players develop an interest in something the GM hadn't anticipated they would, which requires the GM to be very good at on-the-fly gaming, or they get interested in something the GM trails across the path and they're back on the railroad (albeit on a different track).

    As I suggested above, I'm not sure this problem is such a problem. If people choose to play an investigative game, I think they expect a railroad, or at least a road with well-marked signs; they might make a wrong turn at some point and need to backtrack, but at least there is an established road they're meant to follow. That said, I really like your proposal for a sort of "occult search" sandbox; in my experience, it's unusual for CoC players to try being Joseph Curwen instead of Marinus Bicknell Willett.

    ReplyDelete
    Replies
    1. I think your last sentence sort of answers the confused premise problem: "occult search sandbox" is really more about exploration than investigation, right? Or, perhaps, it is the point at which they meet. You are exploring archives, networks of scholars, museums, whatever, rather than geography, but the principle is similar. The task is then, I suppose, to create a sort of hexmap of knowledge rather than physical locations.

      Delete
  2. Actually, the more I think about this, the more it seems like I run standard exploration-style scenarios somewhat like you are describing investigations. I create the basics of a location and some threats and tensions. Sometimes this "creation" is only a few sentences or paragraphs that doesn't get more fleshed out until players show interest in it, sometimes it's a full keyed map. Then, I come up with a few things that are likely to happen if the PCs do not interact with the situation.

    I'm curious how you would categorize this type of play.

    ReplyDelete
    Replies
    1. I suppose you are describing what I was talking about in the latter part of the post, really: it's why the GM needs to be really good at improvising - he just has some vague ideas and notions of NPCs and locations and info, and when the players begin to interact with those things he fleshes them out appropriately.

      Delete
  3. Replies
    1. Yes. I want to run that sort of campaign next, which is why I'm thinking these things over.

      Delete
  4. I had party doing little scenarios and got them exploring HPL county - get a job at Arkham, rent holiday house at Kingsport, go hiking in Dunwich over weekends. Kept going sandbox for months.

    ReplyDelete
    Replies
    1. And was it played completely straight, without any Cthulu mythos stuff in there at all?

      Delete
    2. Well no but if you wander round the HPL county adventures you will find horrible things and get entangled with locals. Keeper doesn't have to lead players.

      Delete
    3. Isn't that just basically playing it like D&D with different furniture, though? Not that there's anything wrong with that.

      Delete
  5. My refuge has been to create a campaign framework, usually through a larger agency that gathers plot hooks, and then put the players in the role of sifting which leads to follow up. You could do that using Delta Green; our last Cthulhu campaign used a homebrew organization (the NYPD Special Crimes Division) for a 1920's campaign.

    It is an exponential amount of additional work, and you need to factor opportunity costs; what's happening to the other plot hooks when the players choose to focus somewhere else?

    ReplyDelete
    Replies
    1. This is the way I've done it in the past, with varying levels of success.

      Noisms, it may be worth looking at The Armitage Files. It's for Trail of Cthulhu, but it should still work well enough for the original game. I've not read it myself yet -- one member of my group keeps threatening to run it for us -- but I'm told it's an excellent framework and kickstarter for an investigative sandbox campaign.

      Delete
  6. noism wrote,
    "And by the same token, you can create a huge rumour table all you like - a d100 list of strange occurrences for the PCs to delve into - but at the end of the day, as GM, you have to know where those strange occurrences lead: all you've done is create a list of 100 quasi-railroads, ultimately. "
    Could that concept not be extended to say, that when a human being runs a sandbox, they are creating a quasi railroad of all the "lists" the human can store in their head?

    ReplyDelete
    Replies
    1. @Random Wizard

      The list entries are only railroads if you know where they end, not just where they begin.

      Delete
    2. Also, the list itself is not a railroad for several reasons.

      1. It is dynamic (entries can be added and removed).

      2. There is no presumption of when or if PCs will engage any specific situation.

      Delete
    3. What Brendan said. Think of the difference between:

      1. There is a mad wizard living in the local woods.

      and

      1. A woman was murdered last night in the local woods.

      The former is open-ended. The latter has an ending and is hence a quasi-railroad.

      Delete
  7. If you and the players improved up a conspiracy that they were all part of, and the game was them extending and protecting the conspiracy while decent normal folk and the authorities tried to stop them, as if you were the villians in someone elses game, what kind of game would that be?

    ReplyDelete
  8. I think you need a few things for a sandbox mystery:
    1) A big enough conspiracy that clues could be tracked down in many places. In the game I'm running it's "where are the zombies coming from and how can they be stopped?"
    2) A random clue/lead generator. This is important. When the players kill a related enemy you roll up a clue type and decide what it means on the fly
    3) A general idea where some of these clues might lead so that when they decide to go talk to the Professor, or the Occultist Expert etc. he can drop the name of a location or an artifact or something.

    ReplyDelete
    Replies
    1. Billy, can you give an example of your random clue generator at work? It seems like an interesting idea, but I can't imagine how it would work in play.

      Delete
    2. I echo Kelvin's sentiments. I love the idea of a random clue generator, but I find it hard to imagine how it works.

      Delete
    3. As to how it works:

      The table:
      http://billygoes.blogspot.co.il/2012/05/random-clue-generator.html

      I use it:
      1) When rolling up treasure I usually roll on my "something interesting" chart which has a 50% chance of being on my Special Item Table and 50% chance of being a clue
      2) When the party investigates a location in search of a lead

      In case 1) above, it may or may not be relevant to the Zombie mystery, so
      A) it has to do with the main mystery
      B) it's connected to something else, in which case I use a table similar to Noism's Random Mission Generator to determine what the plot it has to do with is

      That way you end up with this sort of noir spy-thriller/detective fiction vibe with lots of different plots going on. The players can pick the clues they follow up-on. Very sandboxy

      Delete
    4. As far as an example:

      http://billygoes.blogspot.co.il/2012/06/polish-resistance-session-6.html
      There were two clues in Session 6. The first was of type B) above(don't remember what the plot was--players didn't follow-up). The second had to do with the zombie plot.

      Actually, for every clue I generate for the zombie plot, there are probably two I make-up on the fly.

      Like in Session 5 when the party had lost the trail of the zombie-making convoy and I rolled-up random zombies, I decided they would be dead soldiers from the zombie-making squad with special insignias on their uniform.

      The point is to make up a lot of leads and see which ones the players follow(most of them they don't, but some of them they run to investigate). I don't have the mystery all figured out ahead of time. You just have to stay one step ahead of the players.

      Like in Session 6 the party said "screw this investigation, we're heading for the Russian occupied territory". All my zombie mystery ideas had been on German side of the border or the no man's land until that point. So for Session 7 I decided that the cultist had been to the Russian side of the border in search of some as of yet undetermined artifact. So when the party told a Resistance member about the cultists, he said "Hey, those guys were here a few months ago!" Well that was it--they dropped everything to seek out information about those guys and chased down more information all-session long.

      Or in session 4. I still hadn't determined if the zombies were made my mad scientists or cultist. Then the party questioned a witness so I had to make up my mind and I went with cultists.

      So my point is:
      a) tons of clues/leads with a generator to help--not just related to the main mystery
      b) let the mystery evolve on it's own--you only need very general ideas in advance

      Delete
  9. I think there was a series of posts on the Alexandrian about this issue:

    http://thealexandrian.net/wordpress/15151/roleplaying-games/game-structures-part-5-mysteries
    http://thealexandrian.net/wordpress/1118/roleplaying-games/three-clue-rule

    ReplyDelete
  10. I hate to sound like a broken record but when it comes to investigative games I highly recommend anyone and everyone interested in them to check out GUMSHOE and InSpectres (Trail of Cthulhu mentioned by Kelvin Green is a GUMSHOE game and InSpectres has and it's the net).

    GUMSHOE by Robin Laws (Robin Laws man!)
    http://en.wikipedia.org/wiki/GUMSHOE_System

    Trail of Cthulhu
    http://www.pelgranepress.com/site/?page_id=242

    UnSpeakable - Cthulhu for InSpectres free pdf
    http://www.memento-mori.com/inspectres/unspeakable.pdf

    ReplyDelete
  11. Long ago I ran a fairly sandboxy campaign of COC... based in San Francisco during the psychedelic 60s.
    We went with the assumption the PCs were 'looking for trouble'... they weren't naive innocents. My original idea was based a bit on HPL's 'The Hound' that featured grave robbers who owned their own copy of The Necronomicon. Our group wasn't nearly as nasty as those fellows... but they were a bit more craven than most COC groups. They weren't out to save anyone, though they often did... they were experimenters and searchers and thrill-seekers.
    There were a number of other local groups and organizations I set up as competition, strange occurrences, conspiracies. Nothing seriously world-threatening for the most part. It was very street-level and most of the monsters were human (Charles Manson put in an appearance).
    In a way, I suppose it played a bit like 'Vampire'... different groups secretly jockeying for power/knowledge... while all sorts of horrid things are lurking, waiting.

    As for 'Trail Of Cthulhu'... the system annoys me (a LOT), but the books make nice supplements for COC.

    ReplyDelete
  12. Like Barking Alien, I hate to sound like a broken record (as I'm always touting UA), but the opening chapters of Unknown Armies describe exactly what you note here — there's a wealth of occult knowledge and power out there, if you can find it. The authors seem to encourage characters to look for it rather than GMs to lead them to it.

    ReplyDelete
  13. I think the problem is that so many investigative plots follow a "ball of string" model. The clues and events form a chain from hook to solution. A Jigsaw puzzle is probably a better metaphor for a sandbox style investigation. Finding all the pieces is only part of the solution. Figuring out how they fit together and trying various solutions will provide a lot of the action of the campaign.

    Especially since this will be taking place in a "live" environment. Rather than a point-&-click computer adventure where things wait until you solve the puzzle that advances the plot, incomplete and imperfect solutions will also impact the campaign.

    ReplyDelete
  14. I think you've fairly summed up why so many of the Call of Cthulhu campaigns I'm aware of (those related to me by others, and those I've played in and tried to run) have spectacularly failed (the most recent a near-TPK that resulted in half the players quitting): there are two different modes of play involved here, investigation and exploration, and to do Cthulhu "properly" both the players and keeper have to be in investigation mode. Which is not to say you can't dismantle Call of Cthulhu, lift bits out, and add them to your own exploration game...

    ReplyDelete
  15. Here's a post Zak did about how you can frame an investigative adventure in a more free sandboxy style.

    http://dndwithpornstars.blogspot.com/2012/02/hunterhunted.html

    ReplyDelete
  16. I think at this stage *you people* would be better off not using jaded cliches like 'railroading' and 'sandbox'. As in any form of discussion If you can't say what you mean without using cliches then I suggest you don't know what you are talking about.

    Now, the difference between investigation and exploration is merely that the former more accurately describes the style of gaming where a DM is more explicit in describing events occurring independently of the players.

    >> They might start off in Providence, Rhode Island in 1922 with a web of contacts, and locations, which they can tap as resources in **conducting their own self-centered investigations** - my character has heard that it is possible to live forever, and now I'm going to find out how - and this would allow them to drive things along more autonomously.

    **conducting their own self-centered investigations**

    That my friends is called a **campaign** to be distinguished from **one-off adventures** which are much more in danger of reducing player autonomy because the players DON"T KNOW ANYTHING ABOUT THE DM"S WORLD.

    I am struck over and over by how subtle and fugacious these concepts are to the majority of you who can't sustain a campaign of your own design.

    ReplyDelete
    Replies
    1. I can't for the life of me figure out who this comment is directed towards, to be honest.

      Delete
    2. Himself at a guess.

      Delete
    3. Me neither, but at least I learned a new word.

      Delete