This writeup will discuss my impressions of the game itself, and my thoughts on the Gumshoe system more generally.
I’ll follow it up with some notes on the adventure I ran, some resources for those who might want them, and also some thoughts on the spin I put on the game concept, since it was a bit unusual.
Since this is my first time running a Gumshoe game, I decided to run a prewritten adventure. It’s actually the first time I can remember ever doing that—I’ve always homebrewed the adventures I’ve run in the past—but I also decided to adapt it a little, because I felt like it would help me get a better sense of how the adventure structure worked.
I was looking for an art-centric spin on Fall of Delta Green (as I’ll discuss next time) and I was surprised to discover that both of the existing adventures for the game are pretty compatible with this art-centric campaign concept. This was especially true of On a Bank, by Moonlight—the Free RPG Day adventure, available from this page—which concerns a weird commune somewhere in New York State obsessed with directed dreaming, cult orgy-rituals, and horrifically-inspired art. I adapted it in a few ways:
- resetting it to a small town in the South of France (but keeping a few American GMCs involved, and retaining some ties to Rhode Island)
- adding some threads that link outward to the rest of a potential campaign—which is to say, links out to a bigger “conspiramid”
- adding some details to amplify the relevance of art to the adventure (even though it’s quite central)
It took me a few sessions to start feeling a little less rusty and also to stop worrying about “messing up” and to start to just simplify some of the rules to suit my group’s interests and preferred style of play. By the end, neither my players nor I were fully sold on the system. Several people (like me) enjoyed how it foregrounded investigation, but found the action side of things somewhat unconvincing.
In any case, I thought I’d talk a bit about the game itself.
I find Gumshoe an unusual system.
It starts with the books: they’re big, chunky, attractive hardbacks, and I think this is (subtly) on purpose: it kind of pushes a button in the brain of a lot of people who started out with trad RPGs back in the 80s and 90s. There’s expansive text in the books, too, with rather involved discussions of rules, organized in a way that reads a lot like trad RPGs.
That said, the system is actually pretty simple underneath all that. Despite all that content, the game is actually quite rules-light: there’s two kinds of actions: Investigative Abilities where you auto-succeed as long as you have the right skill (but can spend points for extras), and General Abilities where you spend points to gain bonuses to rolls with a relatively flat target for success. This ends up feeling very rules-light… and I kind of end up thinking, “Huh, this rulebook could have been a lot shorter.” And it could, in theory, except…
Well, except. The caveat is that there are a lot of finicky little rules, each specific to different Gumshoe systems and tuned to better emulate a given genre. Also, the “action” stuff (meaning anything non-investigative) tends to be finicky, with exceptions, special cases, and the like, depending on which system we’re talking about. This seems especially true for the systems Kenneth Hite has designed to feature hyper-competent PCs, like Nights Black Agents and Fall of Delta Green. I suspect this is on purpose: some of the complaints people made about earlier Gumshoe systems included a perceived lack of crunch in the use of General Abilities, so, like, here’s your crunch. Fall of Delta Green has specialized rules not just Electrocution, but also degrees of electrocution; rules for how PCs shoot their guns dry, shooting a spray of bullets, having weapon jams, and using a range of weapons (from heavy machine guns and grenades to garrottes and ramming one another’s cars off the road). The rules aren’t that heavy, but they sure feel heavier than the simplicity of how Investigative Abilities work.
This can make it challenging for players to adapt to the system: if you’ve got the idea in your head that your character is basically competent at most things, you’ll find the default flat difficulty for most General Actions a little weird to reconcile with this. Of course, you also have points to spend, but that means you have a resource management dynamic to think about, mentally pacing yourself since there are some Abilities that refresh daily, as well as ways to create extra of general purpose points pools through tactical fact-finding benefits refreshes you get by following your Motivations and so on. It’s sometimes hard to know whether to spend boldly, conservatively, or not at all, and at other times it can be hard to narratively explain some of the roll results you get. And, well… some of the “procedural” rules were finicky enough that I ended up handwaving them a bit during play, so that I could keep things moving along instead of having to look them up. This may just be something about crunch: I find myself more attracted to the streamlining that I think characterizes Kevin Kulp’s approach, or maybe the more condition-driven approach of Robin Laws’ newer “Quickshock” version of Gumshoe. (Though I haven’t looked closely at the latter, and Kulp’s Timewatch is, it’s worth noting, one of the thickest Gumshoe rulebooks on my shelf, though that’s in part because it includes a really massive number of alternate campaign frames.)
All that said, in terms of crunch, I also think fluency with the system was part of the issue for me. I’m at a point where learning new, complex rules just doesn’t appeal to me so much, especially when a more simple, streamlined, and consistent resolution method could work just as well. Heck, before running Fall of Delta Green, I spent a few months working on a spy/thriller game that uses Forged in the Dark resolution rules for action sequences, but hacks in Gumshoe for investigation. (The system uses tags for Investigative specializations instead of outright Investigative skills, and as with Quickshock, dispenses completely with Investigative point spends for extra information.) That said, my experiences as a player (with Fall of Delta Green and Trail of Cthulhu alike) have been smooth and positive, so I feel like perhaps if I were more comfortable with the rules, players might have felt a little more comfortable with this stuff, like I did when I was playing with GMs more experienced with the system.
There were other ways that the game system didn’t really gel with the group. It’s not that they hated it, they just felt generally unconvinced by it, not really sure that the added work involved in getting comfortable with the system was worth it in terms of how it pays off in play experience. They also felt a little uncertain about how to get a handle on the resource management side of using General Abilities.
(To be honest, I experience this kind of discomfort in any game that has an element of resource management: it always takes a while to get used to the uncertainty and pacing of resource spendsm whether it’s spellcasting in an OSR game, or uses of an ability in 5E, or Luck points in a story game, or whatever. We probably notice it more with Gumshoe because the resource management mechanic is so foregrounded and crucial to how action sequences resolve, since we’re used to relying on luck instead of having control of the narrative pacing of our character’s successful actions.)
All that said, for me the biggest challenge so far seems to be scenario design. I certainly don’t feel confident enough in my abilities to improvise mysteries of the sort that are played out with Gumshoe, and constructing an adventure was a daunting prospect when I sat down to start taking notes and tried to develop a mission. For that reason, I decided to run the Free RPG Day adventure On a Bank, by Moonlight. However, I also decided to adapt and expand it slightly, re-setting it in a small village in rural France. I’ll talk more about that in a follow-up post. but I will say prep seems like a somewhat intensive undertaking. (Even just adapting the adventure to France, weaving in a few more narrative threads, and expanding some material for action that went beyond what was written in the adventure took me about five to seven hours.)
I’ll talk more about how the game went next time, but for now, I want to focus on what I learned from adapting and running the adventure. On one level, my sense is that Gumshoe adventures are actually pretty simple: they’re a kind of loose web-of-clues pointcrawl type thing. But on another level, it’s a tricky balancing act: you want to introduce clues at a rate that makes sense, as well as introducing opportunities for action that alternate with the clues somewhat. Note, I said opportunities for action. Players sometimes decline the opportunity, and to a point that’s okay, if it makes sense for their characters… though eventually, the action should find them.
Anyway, here are some of the things I figured out while adapting the adventure, and running it:
- Hyperlinks. Seriously, for me these are mandatory. Right now, I’m new to the system, but I cannot imagine running Gumshoe scenarios from print books. I mean, I imagine you could, and I am happy to have the books in order to read and digest them, but for actually running adventures, I am blown away by how much it helps to have hyperlinks. Part of this is because a good Gumshoe adventure is a web of clues and leads, a bit like one of those crazy string boards you see in investigative TV dramas. That means players can go from point A to point R, then to point C, and then congregate hours later at point P. If you’re flipping back and forth in your book trying to keep things straight, you know what you end up with? A+R+C+P = CRAP, but misspelled. But with hyperlinks (and searchable text), it’s more of a walk in the “PARC,” to use the French spelling of the word. (Ha, see what I did there? Terrible, I know.) Seriously, hyperlinks make it way, way more manageable. I was a bit shocked to see that the PDF version of On a Bank, by Moonlight didn’t have hyperlinks built in.
- Clues need to suggest… but not say too much. This is actually somewhat tricky: you want every clue to be a hint that something is off, something is up, something is happening or not happening, and that leads toward the Horrible Truth of the investigation; but it should never rule in or rule out everything else, not on its own. As Gumshoe authors have repeatedly noted, you probably don’t need to purposefully throw in red herrings: players will create those on their own by misinterpreting things that aren’t Core Clues, or even by temporarily making incorrect assumptions about Core Clues. This is more about shattering the knowledge of what’s going on to the point where they have to reconstruct it from shards that have scattered due to the impact of whatever got them called in to investigate in the first place.
- Not every clue should be a lead, but most scenes should include at least one and often more than one potential lead. Characters need “places to go.” These aren’t always literally places, of course: sometimes they’re people who need to be tracked down or talked to, or items that need to be found. But the point is that mysteries are about following leads to retrieve clues. If you only get clues in a scene, you can backtrack, of course, but lots of clues function as leads: they don’t just tell you something about the Horrible Truth, they also tell you about somewhere to go (or something to look at, or someone to talk to) in order to leverage more information about the Horrible Truth. So, logically speaking, most nodes in your web of clues should offer at least one clue that functions as a potential lead. Once leads start to converge a certain amount, your group will start to veer toward that point of covergence, so making sure there’s a good distribution of other things to look into, and a certain amount of blindness about the Horrible Truth, is a good idea.
- GMCs (that is, NPCs) should each have a perspective and their own problems and concerns—like in any RPG—but, and this is the special thing in an Investigative scenario, they should be partially blinded by them. It’s worth it to give every one of them a story, an angle they’re working, and a blind spot. Make them unreliable but also crucial. The Chief of Police in On a Bank, By Moonlight is a great example: the man’s life is on a downward spiral, and he has a few primary emotions you can return to when you’re asked things he doesn’t know (like anger at the commune, frustration at being left out of the loop and not taken seriously, and worry about his daughter). He also has his own agenda, and if you keep him focused on that, the players will have to work a little to get what they want out of him… but they can also do it without him realizing what they’re up to, too, if that’s what they want.
- Floating Clues are important for spotlighting and pacing. Floating clues are clues that aren’t tied to a specific scene or locale. It’s a good idea to have a few of those on hand, in case players go off the beaten path in ways you don’t expect. This is especially true if the PCs split up to do different things: if a PC spends the evening doing something like forging documents or meeting a contact, a Floating Clue coming up can be pretty handy.
- A Clues/Leads checklist. I didn’t make one of these for the adventure, but I think I will try it out for the next one I run, because I feel like it would be useful, especially for improvising Floating clues, keeping track of how much information characters have. The trick here would be to make a list with checkboxes that contains enough information to make the clue obvious at a glance, but also fits onto a single page, with highlights for Floating Clues that could be improvised into any scene.
- Floating Action Scenes are also important for the same thing, but for a different reason. That reason is simple: action scenes are a simple way to add spice to a mission, to shake things up a bit and cost the PCs resources, to kick up the pace, raise the stakes, and so on. This is the RPG extension of the Raymond Chandler’s policy, “When in doubt have a man come through a door with a gun in his hand”: to do that, you need to have some handy guys to send in—their stats, their relevant point pools, or at least a note on appearance or a possible clue they might have on them. It doesn’t take much prep—probably just some stats and a possible clue on an index card—but prep is helpful for having things to throw in on short notice.
- Not all “Action” needs to be “Action,” it just needs not to be investigative action. Characters discovering that they’re being followed by an unmarked car could be an action scene. Discovering a bug in their hotel room could be action. A character having a surreal experience in passing that doesn’t seem obviously tied to the investigation (but might be tied to the campaign) could be action. A random encounter with someone unrelated to the investigation—a mugger, a nosy journalist, a local scumbag—could be action. The whole point here is that stuff happens that triggers the possible use of General Skills, to break up the pattern of characters receiving information based on their Investigative Skills
- GM collaboration may need to be more of a thing because of PCs specialized knowledge and skills. In D&D, part of the fun is characters getting up to ill-advised hijinks: if a player wants to make their character do something that probably isn’t a good idea, you damned well let them, make them roll, and then have fun with the jubilation (if they succeed) or the consequences (if they don’t). It’s kind of win-win. Though it sometimes happens that a player will ask you about something your character would know how to do, but which you don’t—it happened today in a 5E game, in fact—it’s not that common: the skills system, and the kinds of challenges presented in a D&D game, tend to be familiar and easy to handwave with standard fantasy-world assumptions. You can do this is a Gumshoe game, too, but in the games where PCs are by definition hyper-competent, if the player is steering their character toward a rookie mistake, you should be ready to step in and help them out a little. If a player’s sweating small stuff that their character wouldn’t, or seems to have forgotten something their character wouldn’t, it’s good to gently remind them of it—indirectly if you can, but directly if that doesn’t work. “From your training, you can see he’s pretty drunk and won’t notice the missing document: you don’t think you need to risk putting it back, if you can just leave it elsewhere in his house,” or “The one thing that gives you pause is that van. If someone’s watching—and you’ve said you think that someone is—then they probably have night-vision of some kind. Its a starry night, so there’s enough light. They’d probably see you pretty easily… if someone is watching from the van. You’re not sure.”
- The game tends to be played in the theater of the mind, so you will want to work your mental scene-setting muscles. More than once, players found themselves having to ask me to describe a room in more detail. This didn’t used to be an issue for me, but I think all the battlemaps in online games in recent years have spoiled me a bit… or maybe I was just feeling overwhelmed by all the clues I had to juggle. I’m not sure.
- You may need to remind players about what they’re automatically succeeding at. When asking players to roleplay an exchange where they’re using an Investigative skill (or an Investigative spend) that is guaranteed to get them whatever relevant information is available in the scene, it’s worth reassuring them—at least the first few times—that the roleplaying isn’t going to somehow result in failure, and that you’re just looking for details that come out during the exchange, or giving them a chance to present their character’s personality, style, and M.O. People used to traditional games may well assume that the roleplaying—or even the detail you ask for when you say, “Sure, how are you going ask him that?” might modify their chances of success when in fact you’re looking for a little scene color, or for future repercussions of the exchange.
Overall, my group’s response to Gumshoe wasn’t necessarily negative, but it wasn’t extremely enthusiastic. I don’t blame them, and I’m not sure how I feel about the system myself. From past experiences as a player, I love how the system foregrounds investigation and makes it feasible to run a game where investigation is the focal activity. It definitely has its strengths.
Still, I can’t blame those who find the resolution system somewhat unsatisfying—even if I feel tempted to think that getting more comfortable with it would fix that. We’d opened up the can of worms of a much bigger mystery by the end of my version of On a Bank, by Moonlight, and we’d like to revisit the world, we may not do it with the Gumshoe ruleset. Even if I don’t complete my own FitD/Gumshoe hack, there’s always External Containment Bureau (a paranormal/investigative game that uses FitD rules), to which it would be pretty easy to add Gumshoe-like Investigative abilities if we wanted.
All that said, I’d like to revisit Gumshoe, if only because I have so many systems on hand, and I’m curious whether a different version of the system, especially a simpler one, might not appeal a bit more to my group, and be easier for me to run, too. Probably the main reason I feel that way is that when I’ve had a chance to participate in games of Trail of Cthulhu and Fall of Delta Green games as a player, I didn’t really have any issues at ll. It was all pretty smooth for me, except a little uncertainty about spending General skill points at first.
But anyway, for now, it’s someone else’s turn to run a game for a while!
Anyway, I’ll follow up next time with a discussion of the adventure I ran. Coming soon…