What The Rulebook Didn’t Include: Gumshoe (Part 2 of 2)

Note from the editor: This is part 2 of guest author Andrew Brehaut’s deep dive into making Gumshoe work. Part 1 is available here.

Interpersonal Interactions

Gumshoe places all interpersonal abilities under the investigative ability umbrella, therefore they are never rolled. This results in three common cases:

  • The player is looking for information (aka a clue): They approach the NPC in a way that corresponds to an interpersonal ability they have, and they get the clue.
  • The player is looking for something besides information. This is probably the realm of a special benefit: They approach the NPC in a way that corresponds to an interpersonal ability they have, they make a spend (or push), and they get the resulting benefit.
  • The player approaches an NPC in a way that does not correspond to an interpersonal ability they have: they are rebuffed.

The particular ability used colors and positions the narrative. Using intimidation may complicate matters later on, for example, where flattery might not. If such an approach occurs but the NPC still has future (core) clues, you need to walk the line between letting them off without consequences and blocking access to the clue. Often a good solution here is to require a spend of a different interpersonal ability (reassurance or credit rating for instance).

I’ve found that one of the ramifications of the way interpersonal abilities work, as well as the procedural nature of the genre, is that character interaction with NPCs tends to be more abbreviated than in a more dramatic character focused game of the sort readers of this blog tend to enjoy.

Last pointer: Sometimes an NPC will provide multiple clues, and the players will need leverage to get the latter clues. This should be provided to the players as a clue itself: If the player tries to intimidate the accountant to get access to an important ledger but he doesn’t budge, then tell the player that accountant seems more scared of someone else. That’s the clue.


Gumshoe games’ combat systems tend to either be simple (in games like Trail of Cthulhu), or require players to learn special rules (in games like Night’s Black Agents). 

Simple combat requires the GM do a lot of lifting with fictional positioning and description to make things exciting. A fight is not something to just drop into a scenario that feels like it’s slumping to add a burst of excitement, and the participants in combat have few mechanical levers to operate tactically. Worse, the result of a fight going badly can seriously stall momentum: not only do the players not get what they want, they will likely want to hunker down to recover general ability points. I prefer to use a Contest instead, for example a chase. The results of Contest failures tend to introduce complications instead: The chaser is led into an ambush or is separated from the rest of the player characters for instance.

The latter case you can rely on combat for excitement, assuming the players have bought in and learned those special rules.

Antagonist Spend Strategies

In all non-QuickShock Gumshoe games, you need to come up with a strategy for your NPCs to spend their points. Here are my heuristics:

  • For any character you do not expect to survive the fight, spend the entire pool of points for an NPC.
  • Base an NPC’s spends on their disposition and awareness of the PCs.
  • Have the character reevaluate their opponents and adjust your spend strategy at some point in the fight.

My basic spend strategies are as follows:

  • High constant spending: 3 points per attack (4 if the opponents have good cover). This will basically guarantee a hit every attack against even skilled humans. An alert, dangerous threat.
  • Low constant spending: 0 or 1 point per attack. Lacks awareness, or doesn’t acknowledge the PCs as a risk.
  • Ramping up spends: Start with 1 point per attack, and increase it to the maximal as the fight progresses. Panicking or reassessing the threat.
  • Tapering off spends: Start with the Maximal spend, and then decrease to 1 point per attack as the fight progresses. Fatigued or cautious enemies, reassessing the threat.
  • Alternating high and low spends: Default to 0 or 1 point spends, and use 3 point spends intermittently to mix up the rhythm of the fight. This enemy needs some time to recharge, or perhaps responds to a surge of adrenaline when hit.

Some heuristics for modulating those spends:

  • Have tactically aware NPCs adjust their spend for cover or threat level (they are aware of the difference between trained and untrained combatants and will take the different hit thresholds, or potential damage dealing into account).
  • Monsters and inhuman enemies who can’t distinguish between a professor and a soldier should spend the same amount per target regardless of what you the GM know about hit thresholds.
  • High hit threshold or heavily armored enemies have the luxury of frugal spending.
  • Human enemies who have not been in a proper fight should spend less by a point (at least). They aren’t familiar with high hit-threshold targets.

Shaping a Gumshoe Mystery

One quirk of genre that you will need to master is the independence of antagonist and protagonist for a large section of the episode. The protagonists chase the shadow of an antagonist until they have enough clues to piece together an identity. Similarly the antagonist may not be aware of the investigators on their trail and thus won’t react until the protagonists start closing.

I talked last week about how the scenario is structured around an A- and B-plot, with the mystery forming the A-plot, and the protagonist’s story forming the B-plot. This is true of episodic play, and for a one-shot, you’ll typically just want the two to be much more closely entwined.

My basic formula for creating the A-plot is to start with the B-plot: Pick some aspect of the character I want to focus on this episode, and figure out how to reflect that as an A-plot. Look at facets of the character like drive, pillars of sanity, sources of stability etc for inspiration. For example, Antiquarianism, but too much: The antagonist’s obsession with the past is harming people in the present. Or duty might be reflected in a character’s sense of duty being misplaced.

Early in a scenario I like to have a 2:1 ratio of “investigative” and “general” scenes. As the scenario progresses, this shifts toward almost exclusively general scenes. In the latter half of the scenario when investigative scenes have mostly dropped away, introducing more antagonist reaction scenes is helpful; these become a source of action oriented sources of information for the players.

For a mystery game like Trail of Cthulhu I like to start with an investigative scene, while for a thriller game like Nights Black Agents I prefer a general scene.

Think of the ‘spine’ structure suggested by the official GMing advice as providing two things: a safety check that you have considered at least one path of clues from A→Z, and a clock (events applying ever-growing urgency) helping you pace the mystery. What it should not be is the exact map of how the mystery is to be solved.

Finally, I’ve found that giving the players some way to visualize the mystery helps a lot: I have a pinboard I place on the table and the players write the clues down on small tabs of paper and pin it to the board. I’ll have prepared a large period map in the center of the board to give things a sense of geography. String is optional but encouraged.

The Recap

  • Gumshoe is all about spending limited resources, and luck is a very small part of the system.
  • Keep clues small and frequent.
  • Players should be spending their investigative points for special benefits.
  • Neither combat, nor nuanced interpersonal scenes, form the backbone of your games: rely on the mystery for that.
  • Build your mystery’s A–plot from characters, even though it won’t be directly about them most of the time.

Another note from the editor: Apparently it’s always all capped, as well: GUMSHOE. That’s on me, I changed it, Mr. Brehaut is blameless.

Leave a Reply