GM: It’s a pretty cheap lock, so it only takes you about fifteen seconds to pick it. You hear the satisfying click.
Rachael: Great. I’ll slide my picks back into the hidden lining on my belt before opening the door and slipping through.
GM: You find yourself in the office of Sir Sebastian. An imposing, mahogany desk with a flared plinth dominates the center of the room. Heavy, velvet curtains with gold appliqué seem to swallow the light from the windows. Vivid, arsenic-green wallpaper render kaleidoscopic patterns on the walls. Give me a Search check.
Rachael: 25.
GM: Okay, you find a hidden compartment on the wall, which you open by tracing the patterns in the wallpaper. Inside you find a small, metal ball with black, acid-etched symbols covering its surface. Give me a Spot check.
Rachael: 18.
GM: You notice that there’s a thin seam running around the center of the ball. Give me an Idea roll.
Rachael: 16.
GM: Okay, that just good enough. You realize that the ball can be rotated to form different patterns with the symbols. You experiment for a minute, and find a sequence that causes the ball to pop open. Inside you find Marie Artaud’s ring.
Rachael: Great. I’ll take the ring, close the ball, and get back to the party before I’m missed, making sure to lock the door behind me.
Hopefully the problem here is immediately apparent to you: The GM is cutting off the player’s investigation of the scene by preemptively calling for skill checks. The PC effectively ends up in a kind of “autopilot mode” during which the game ceases to be truly interactive and the player is rendered into a passive audience that can only watch the character’s actions playing out.
It’s rare (although, unfortunately, not unheard of) for this error to be carried out in quite so egregious a fashion, but I’ve found that its less pronounced variants are shockingly common.
THREE-TIERED PERCEPTION
Probably the most common version of this problem that I’ve seen is when the GM preemptively calls for a Search check or similar mechanic. At a minimum, however, a good GM needs to be able to distinguish between three different levels of character perception:
- Automatic Perception
- Spot-type Perception
- Search-type Perception
I’m using skill names from 3rd Edition D&D, but this remains true even in games which don’t mechanically distinguish between these categories. (Pathfinder, for example, is 3rd Edition’s kissing cousin, but lumps both Spot-type and Search-type perception into a single skill.)
If you’re familiar with the Art of Rulings, you may notice how these fall into its three core principles:
- Passive Observation is automatically triggered
- Player Expertise activates Character Expertise
- Player Expertise can trump Character Expertise
Automatic Perception and Spot-type Perception both fall into the category of Passive Observation: Automatic Perception is the stuff that literally anyone standing there will observe. (If you want to think about it in purely mechanical terms, it’s the stuff that requires a DC 0 Spot check to notice.) Spot-type Perception is the stuff that people can notice while just standing there, but might not. (Spot checks are an example of this, but so are Knowledge checks: Anyone can see the large flag hanging on the wall, but only some people will recognize what nation the flag belongs to.)
Search-type Perception falls into the second category, being an example of Player Expertise activating Character Expertise: This is the stuff you can’t see by just standing there. You need to go do something in order to see it / learn it.
Beyond this basic core, there are a few advanced techniques to consider.
Matryoshka Search Technique: This is something I’ve discussed in a dedicated post as a Random GM Tip, but beyond the threshold of the basic Search-type Perception, you can begin to see the game space as nested layers of interaction.
You can actually see this in the example above: Rachael needs to search the room to find the hidden panel. She needs to figure out how to open it. Then she needs to examine the ball inside and figure out how to open that. There’s not one threshold of interactivity; there are many, each nested inside the other.
Superman’s X-Ray Vision: Special abilities (particularly always-on special abilities) can cause some items to swap between the different perception type categories for specific characters. This can result in Rachael and Teresa having different perceptual relationships with a given game space.
I’m Just That Good: What if you’re really, really, really good at spotting stuff? So good, for example, that you might be able to notice the hidden panel in the wall from across the room whereas other characters would need to physically interact with the wall to notice it.
In some ways, you can actually think of this as a variant of the character possessing a particular special ability (it’s just that their “special ability” in this case is being really, really good at noticing hidden things).
I actually mechanically instantiated this into 3rd Edition: In my house rules, if you beat the Search DC by +20 while making a Spot check, you’ll notice the hidden feature as if you had actively searched for it (either directly, if possible, or through some form of tertiary indication if not; you may note that the latter is effectively introducing a Matryoshka technique). You can do similar stuff with, for example, exceptional successes in Eclipse Phase or point spends in Trail of Cthulhu.
You might be wondering why this is “okay”. Why is this any better than the example of the GM preempting them? Aren’t you still skipping interactive steps?
You are, in fact, still “skipping” steps. But you’re doing so as a reward for character ability. It’s similar to a wizard “skipping” sections of the dungeon by using a passwall spell: Yes, you’re bypassing the “intended” or “natural” path of progress, and there are things you’re losing or missing out on as a result. But you’re gaining a different (and important!) benefit.
That’s why this is an advanced technique: You need to understand the rule in order to know when you can (and should!) break it.
ASSUMING ACTION
The preemptive Search check, however, is just one specific example of the GM making an anticipatory ruling; a ruling in which they assume that the player will make a particular choice and, therefore, skip past the step where the player actually makes that choice.
In this context, you can actually interpret the problem as a scene-framing issue. As described in the Art of Pacing, the GM needs to identify empty time – i.e., time in which the player is neither making interesting choices nor experiencing the consequences of those choices – and frame past that empty time to the next meaningful choice. What’s happening here is that the GM is incorrectly skipping past meaningful choices.
The problems with this are manifold:
- It hurts immersion as the player loses control of their character.
- It prevents the player from actually playing the game as the loss of control results in a loss of interactivity. In this it’s similar to alpha-quarterbacking in co-op board games.
- It prevents the player from making a different and unanticipated choice. The GM is not omniscient, so even when they assume that there’s only one “good” choice to be made, it doesn’t follow that this is the choice which will be made.
- On the other hand, the GM is a little too omniscient. They are biased by their design of the encounter and the wider knowledge of the scenario, which may blind them to the actual thought process the player/character is experiencing.
In this, you can see a pattern of problems similar to run-time choose your own adventure (as seen in GM Don’t List #6).
It is, perhaps, unsurprising that so many errors in GMing technique share common roots. And, conversely, that the solution to those errors are all rooted in a similar ideology.
Once again, I’m reminded of a similar point made by Seth Skorkowsky in one of his videos: https://www.youtube.com/watch?v=pn29aUjo6k0&t=6m42s
Really good thoughts Justin. I am reminded of a game of Edge of the Empire I played once, the characters were outgunned by a faster, sleeker ship, and realised they they weren’t going to win a straight up brawl. As GM I was waiting for them to make some suggestions on what they might do next (into the clouds, fly down into a ravine, land etc). As players they were waiting for me to tell them to roll X skill. It was an interesting moment where we both (players and GM) realised that we were approaching the game style a little differently.
The players had a lot of experience with the exact style of gaming you talk about at the start, and that’s how they were expecting to play. Roll X, roll Y. When put into a scene where they had to make some suggestions about what they might do, they were completely adrift. As GM, I should have provided more of a framework – instead of roll X, Y or Z, I could have offered a range of options etc.
It all worked out in the end, but it was an interesting point about expectations in play – where the GM can make mistakes about assuming things about the players, and the players can be expecting to play in a certain way. I’d like to say our games have improved since that point!
On the subject of Matryoshka techniques, I sometimes find it difficult to indicate to players when there are more layers to what they’re looking at.
This tends not to be a problem with Search checks, because it’s pretty self-evident that you could toss a room. Academic or technical skills (Archaeology, Cryptography, Forensics etc) are more fiddly, due to their abstract nature and the wider gap between Player and Character expertise.
To go with the original example, it’s not self-evident that Rachael could make an Idea roll to figure out how to work the seam. If she fails, it’s not self-evident that she could go research the item (assuming that’s possible), or take it to the local Thieves Guild.
How do you make these options clear without assuming their intent to use them? Just straight up saying “If you have skill X you could figure it out” solves the problem, but it seems like a soft version of preempting investigation.
Good question.
You’d think the problem of “don’t know to do that” would be one primarily seen in new players, but paradoxically I actually get it the most from experienced players because they’ll have collected whole bundles of preconceived notions of what actions are “allowed” or will be meaningful if they take them.
Some of this is because of a history of being railroaded. (Don’t try to tunnel through the dungeon walls because the GM won’t let you do it, so it’s a waste of time.) Some of it is learning what type of stuff GMs don’t typically prep. (Don’t try to research the reference to Hyperborean violets, because the GM probably didn’t prep anything for that and so you won’t get anything.)
I think this is usually not even a conscious thing: Players have thoroughly internalized the lists of Things We Do and Things We Don’t Do. It’s not that they think about doing X and then consciously reject it because they think the GM will just railroad them; it’s that they’ve learned to just never think about doing X in the first place. That branch of the decision-making tree has been cut off.
In some cases, of course, it just won’t occur to them that a particular clue can be followed up on. If you’re designing with the Three Clue Rule, that’s okay.
But if you’re seeing a systemic failure to take certain types of actions which should be part of play, then the easiest solution is to literally just tell them it’s an option.
You can see this really clearly in situations where the play of a game varies considerably based on whether the players have read the rulebook (and the rulebook’s specific prompts for expected play).
For example, over the past couple years I’ve spent a lot of time running the same Call of Cthulhu and Trail of Cthulhu scenarios for different groups. Groups with players who have read the rulebooks or have a lot of experience with the game will almost automatically go to the library or research the local newspaper morgues for references to whatever they’re investigating. It’s an expected part of play that’s well established in the rulebook. Groups without that experience just… don’t.
(There’s also some divergence between the two games here: It’s laid out as far more “this is the regular procedure” in Call of Cthulhu than it is in Trail of Cthulhu.)
The reasoning here is really easy to understand: If you don’t know that Call of Cthulhu scenarios routinely include prep for relevant newspaper articles, then your thought process (conscious or subconscious) is certainly going to be, “There’s no way that the GM decided to randomly prep some newspaper articles on the off-chance that we’d randomly ask about them, so don’t even bother asking.”
Another example: Technoir is mechanically designed around the default action of “hit up one of your contacts.” If the PCs don’t know what to do next, they should go ask their contact. And the game is designed so that the contact always has a clue or a job or some form of lead that gives them something to pursue. This expected form of play is well described in the book, and the first few times I ran the game, I was running for people who had read the book and just took the default action. The game ran smooth and had a great flow to it.
Then I ran the game for some groups who HADN’T read the rulebook. They had the same contacts on their character sheets, but they just didn’t think to hit them up when they were stuck. The game turned into a grind and it took me a couple sessions to realize that I just needed to literally tell new players: If you’re stuck, do the noir thing and hit up a contact!
Another example: Buying 10-foot poles in old school D&D. If you’ve got a group that knows what they’re used for, then they’ll (a) have them and (b) be using them. If you don’t, then it’s quite likely that the pole just stays on the equipment list.
An interesting phenomenon here, too, is that player knowledge can act like a crystal: Take just one player who knows to look for random newspaper articles or hit up contacts or use a 10-foot pole and the whole group will learn that THIS IS A THING YOU CAN DO. (If you run a large open table, this behavior becomes obvious really quickly.)
Okay, so what can you do:
1. Embrace the “permissive clue-finding” corollary of the Three Clue Rule. If the PCs try to gain information in a way you weren’t expecting, default to the idea that they CAN find something that way. Over time, you’ll train your players not to arbitrarily discard ideas based on second-guessing what’s in your prep notes.
It can actually be helpful in some cases to literally say, “I didn’t prep anything for that. Give me a minute to figure it out.” And then give them something. That REALLY teaches your players not to self-censor or second-guess on the basis of whether or not they think prep was likely to have happened.
(There are cases where saying you didn’t prep something is problematic or breaks immersion or gives away information about what is and isn’t important. But there are just as many cases where there’s no downside to it and, IME, players love it.)
2. As with my Technoir example, preemptively establish certain norms of play for the current game. This is one of the reasons why, when I’m running OD&D for new players, I will literally read page 8-9 of The Underworld & Wilderness Adventures book, which more or less lays out the major conventions of the OD&D dungeon crawl.
3. Prompt them during play. I prefer to be fairly cautious with this approach (and let the players figure out what they want to do for themselves), but you can:
(a) Be aware of moments when the players are literally asking for help. “Do I recognize the symbols?” “You don’t off-hand, but you’re pretty sure Drosnan’s Runic Concordance might be able to point you in the right direction.”
(b) Preemptively hint at a Matryoshka path that can be followed. “The staff is covered in strange runes. They look strangely familiar, but it would probably take some research to figure it out.”
(In the latter case, you might be thinking they’ll go to a library, but they could easily surprise you in terms of how they choose to research it. They might look for experts in the field or hit up a personal contact, for example. Remember to roll with it.)
Thanks for the detailed response! This is all very helpful.
As to your suggestions:
1. Good point. Permissive clue finding is something I employ, but I’ve deliberately tried to obscure whether a clue is improvised or prepped. I’ve done this for immersion’s sake, but I think you’re right that this kind of metadata can be beneficial.
2. “When in doubt, do the genre thing” is a great idea. I have referenced film and TV examples when starting a new setting, but only to help convey the tone. I probably need to spell out exactly what elements of those settings will be an expected and useful part of gameplay, to make sure we’re on the same page.
I recently ran a heist scenario that went pretty haywire, partly because I didn’t clarify that infiltrating and casing the target would be not just useful but ESSENTIAL. The players got hung up trying to plan with only what they found through off-site research, not realizing it wasn’t enough.
That was despite them knowing the tropes of heist movies. More clarity from me at the outset would have helped, but when things hit an impasse I should have intervened and explained why.
3. Talking to my players, they prefer me telling them what skills would be applicable to a situation. They think that asking would create more inefficient back-and-forth than it’s worth, and it’s hard to argue with that.
A compromise could be to treat this as a kind of passive awareness, triggered by active examination. The archaeologist handling the artifact thing would probably recognize if it’s significant without having to ask. The cryptographer would spot the telltale signs of steganography etc.
(This does mean that in practice a character without the skill would still get “If you had skill X, you could tell more”, but if the alternative is that everybody cycles through looking at All The Things to trigger all possible passive awareness then it’s just easier to skip that and chalk it up to “Hey college boy! Get over here and check this weird thing out!” Now that I say that, I may ask my players to roleplay that 🙂
That said, I’m still hesitant because I also REALLY prefer not to prompt, so I’ll see how far I can go with suggestive description.
I also generally feel better filtering any hints through character abilities. Beyond the initial description suggesting the first examination, one idea I’ve had is to link indications of the next layer to the Margin of Success.
If the PCs find a bronze medallion with a Lemurian sigil on it, a Partial Failure results in “You recall a description of something like this in the journal of the Westminster Archaeological Society, but you’d have to dig to find the specifics.”
This might require a research roll or it might just take additional time. The idea is that Partial Failures give them enough information to attempt to recover that layer of the Matryoshka, at a cost. Only a full Failure leaves them with no leads, which seems appropriate.
Partial Success would answer that layer without a lead to the next one, but full Success would grant an answer and a lead to follow up on.
That is essentially the logic behind Search checks. You can’t examine something you don’t find. Success or failure at the first layer gates access to the next one, unless player expertise trumps it.