On Perception and Observation

These should not be skills.

I was a big fan of these types of skills for a long time (roll intelligence + perception, roll your spot, roll your observation check), till I listened to someone else play in a system with these skills.

"Roll your search," *clatter*, "You don't see any traps," "I open the door," "There's a room," "I search it for treasure and traps," "Roll your search," *clatter* "You don't see any treasure or traps," "I search the doors," "Roll your search," *clatter*, "you don't see any traps, the door is locked," "I unlock it," "Roll your open locks," *clatter*.

Is this fun?

I can come up with several objective reasons why it *isn't* fun.
First is the issue of spotlight time. Either one person has the search skill only (or has trapfinding, let's say) and dominates table time and yet all they are doing is rolling their skill over and over (see the second point). Or *everyone* has the perception-type skill and feels forced to train it over and over because of how important it is. In a system like 3.5 it becomes a skill sink on the non-thief characters; in non-level-based skill systems like Hackmaster, they spend all their time training it until it's up to an unreasonable percentage.
Second, the player has no input on the gameplay. The only time their decision making comes into play (their choices, or 'fun') is when they are leveling and they make the choice to increase the skill. After that it's just rolling at the table every time the DM wants to find out if you can see something.

Now, maybe some people enjoy using dice to reach a statistical mean while six people sit around waiting for something to happen. Maybe your game is a complex combat simulation where you go from fight to fight to fight in the course of an evening. I could see it being useful in both these situations.

I'll have three to six fights a night (sometimes as many as ten or twelve). we generally play for about seven hours, and we rarely spend more than two hours in an evening fighting. (And that was for a melee with around 100 participants. We've had 2 other long combats against groups of undead who don't fall down and die and can't be turned). Mostly we spend adventuring, exploring, and role-playing.

Here's the most important point I think. You don't need this skill. Either you tell them the information because you want them to know it, or you wait for them to look for the information. The whole play of the game is wrapped up in the interaction this skill serves to eliminate. "But what if they don't think of where to look?" "How will they know the trap is there?" Let me repeat myself.

The whole play of the game is wrapped up in the interaction this skill serves to eliminate.

It is the players job to notice these things. Take the Find Traps skill - it still exists in my game. We use it very much like a savings throw. The players tell me how they are moving throughout the environment, what precautions they are taking, etc. Then I describe what happens. They often prod ahead of them with a 10' pole, and they move slowly, mapping and examining around for traps. There was a flaming gas vent trap they avoided. . .

"The hallway ahead looks darker."
"Wait, why does it look dark?"
"You don't know."
"I check out the walls and floor, prodding ahead with the pole."
"You notice that the walls and floor appear to have dark stains on them."
"What kind of stains, I rub my finger along them."
"You see that the walls are covered in soot."
"I look closer on the tops and bottoms of the walls ahead."
"In the darkness you see several vents."
"We go the other way."


"Roll your search/observation/perception"
"I got a 17"
"Flaming jets burn you"


"Roll your search/Observation/perception"
"I got a 27"
"You find a trap"

There is a strong movement in modern systems away from the deathtrap. You know why? Because the players don't have any control of the game. In the way I run my game, if a player runs into a death trap it's because they did something stupid. Not because of random chance.


  1. Thanks for spelling this out the way you have. I couldn't agree more!

  2. "Mostly we spend [time] adventuring, exploring, and role-playing."

    That's what I like most about RPGs, too. If you give me enough descriptive information about a box, I can figure out how to search it for secret compartments, regardless of what some % dice say.

    Certainly dice play an important role in RPGs and I think combat is the best example. But I'd rather role-play than roll-play.

    (Has that phrase already been coined? If not, I call dibs.)

  3. In Errant, I actually made these things into saving throws. They are only triggered when something is about to happen, so there is no meta-game information that is given away. The check is called to avoid a negative outcome, not have a positive outcome; hence a save.

  4. Long after the fact, but question:

    I like the idea of certain characters being able to sometimes "spot hidden doors", or otherwise have a better chance of noticing subtle non-threatening things.

    Since hidden doors and their like aren't actively a threat (thus not *requiring* a something like a saving throw), how would you handle this?

    The options that come to my mind are:

    * Just mention the clues about a hidden door to the characters with door-noticing-ability, without requiring a roll.

    * Have folks roll to notice as they normally would, the roll is just triggered by the presence of a skill, not by the player's statement that they are looking. Essentially the roll is a saving throw in the sense that it's triggered by external factors, not a specifically chosen course of action. Maybe they notice the door on a successful roll or maybe you just give them clues on a successful roll.

    * The question is wrong. Secret doors are not fun to search for because it's not fun when you don't find them. Just don't include them. (Note: I can't remember ever having anyone notice or intentionally search for a secret door in my years of 2ed or 3.x, so for all I know maybe the comic drama of Schrodinger's door is not the crazy wild-card of situational fun I'm imagining.)

    What say you?

    1. Thanks for taking the time to comment - I respond to comments no matter what post they are made on.

      Secret doors are an interesting case - here is how I run it.

      Each door has a specific method of opening it (say lowering the visor on a statue)

      The door itself is hidden.

      The mechanisms themselves are always visible objects noted in the encounter.

      Any player my search a 10' x 10' square taking 1 turn to get their chance to locate secret doors - elves get this chance automatically.

      If they succeed, they learn the location of the secret door but not how to open it.

      At this point they may treat it as any other door - either take the opportunity to search out the mechanism for opening the door, or bash it down.

      Each choice must be weighed versus the cost of attracting wandering monsters.

      I will say that my experience when players do interact with the environment and discover the secret door is a sense of real discovery and wonder. It is highly recommended. I have never seen anyone respond to "I check for secret doors, I roll a 17." "You find one." the same way.

  5. What you're saying here applies just as much to any other mechanic.

    Either you let them hit the bad guy because you want them to, or you wait for them to say they're hitting the bad guy in the right way and then they do.

    Either you let them jump over the crevasse because you want them to, or you wait for them to lay a board down and walk across it.

    If you're making them roll Perception to notice that there's a fire in the fireplace that they're standing next to, you're doing it wrong. If you're making them roll Search to discover that there's ermine robes in the chest when they've just told you that they're opening the chest and looking inside, you're doing it wrong.

    Skill checks are useful for determining outcome when the outcome is indeterminate. The GM could, of course, just arbitrarily make such decisions. But, personally, I find it useful to be able to consult the mechanics as an oracle.

    For example: Do the PCs notice the orcs sneaking up on them before the orcs try to turn them into pin-cushions?

    Similarly, if the PCs specifically look under the bed they'll see the chest there. If they specifically look inside the chest, they'll see the ermine robes. If they specifically check the pockets of the ermine robes, they'll find the key to the secret vault.

    But what if they just say, "Okay. I want to poke around inside the chest. Anything interesting in there?" Is that good enough to find the key? Is it good enough to find the secret compartment on the bottom of chest? What if the secret compartment can only be opened from the outside of the chest? Is the fact that they specifically said "inside the chest" mean they can't find it? Or is the fact that they're clearly interested in the chest good enough cause to notice that the bottom of the chest is a couple inches higher on the inside than the outside?

    As I said, the GM could certainly just make arbitrary decisions. But, personally, I find it useful to occasionally say something like, "Okay, there's a chance they'll notice that. Let's find out together."

    For more of my thoughts on this, check out The Art of Rulings.

    1. First, I would like to say that I am a huge fan of your blog - it is one of the main influences that has encouraged me to write my own blog, and I have a tremendous amount of respect for you and your thoughts.

      I firmly believe in the DM as impartial adjudicator. This means that he has no wants or desires to how the game resolves.

      Your first example is a case of surprise which has been part of D&D since the very beginning. It is a valued and interesting roll because the possibilities for success or failure are significant among other reasons. Although this rule also can be trumped with player skill

      I answer your second question with yes. Let me elaborate.

      "Is that good enough to find the key?"
      "Is it good enough to find the secret compartment on the bottom of chest?"
      The response is, "The bottom of the chest is covered in a rich red cloth and does not appear to be as deep as the chest is tall."
      "What if the secret compartment can only be opened from the outside of the chest?"
      "It appears to be solid on the inside. If there is a secret compartment, you don't see anyway to open it from the inside. Would you like to hack through the wood or check on the outside of the chest?"
      "Is the fact that they specifically said "inside the chest" mean they can't find it?"
      No. I talk at length here about why pixel bitching, which is what you're describing has nothing to do with skill light or skill heavy games.


    2. "Or is the fact that they're clearly interested in the chest good enough cause to notice that the bottom of the chest is a couple inches higher on the inside than the outside?"

      Yes. The chest is an excellent example. Most of the time they just take the goods in the chest. Any additional questioning will be instantly rewarded with the solution.

      But this isn't just about chests. Encounters are constructed. There is always a clue to anything of interest. There are also red herrings. My example is the classic one from the DMG, where Gygax says, on Page 99:

      "First, the others checking the containers find that they held nothing but water, or are totally empty, and that the wood is rotten to boot. You see a few white, eyeless fish and various stone formations in a
      pool of water about 4' to 6' deep and about 10' long. That's all. Do you wish to leave the place now?"

      "Wait! If those fish are iust blind cave types, ignore them, but what about the stone formations? Are any of them notable? If SO, I think we should check them out."

      "Okoy. The fish are fish, but there is one group of minerals in the deepest part of the pool which appears to resemble a skeleton, but it simply - "

      I believe that as you do different player capacity is more important and more engaging then character capacity. Other than that, as you can see here I agree with many of your points about why and when I should roll.

      I do not think, after player expertise has been activated, that character expertise should then trump that - which is what happens if the character fails the roll.

      As to the pixel bitching regarding traps. All traps are obvious ("There is a dead body in front of the chest"), or they automatically get told "Do you wish to search for traps" on doors and chests. They may do this and use player skill to determine if there is a trap, but if they find the trap the roll can't make them fail.

      I primarily run 1st edition which has surprise rolls and Find/Remove traps skill - which are the only valid uses I consider perception to have. The term perception is too broad for what the useful functions of the skill are.

      In other words - this article excepting certain stylistic differences is a direct outgrowth of my reading of your articles and those on ars ludi regarding game theory.

      Thanks for taking the time to comment.

    3. Also, I'm pretty sure this is a conversation we had 11 months ago in the comments of the article you linked to.


  6. There’s a big difference between an attack roll and a jump check.

    An attack roll is part of a system that involves (generally) multiple die rolls and resource management. Not to mention tactical decisions.

    (And by “tactics” I mean things like maneuvering to emphasize your side’s strengths, the enemy’s weaknesses, and leverage terrain. Not choosing which special move/feat/power to use. Though, when you’re using those things, that’s yet another reason why an attack roll doesn’t stand alone.)

    A jump check (generally) has none of that.

    Pulling the attack roll mechanic out of the combat system and using it to resolve a jump (generally) makes the game a lot less fun for me.

    I can agree that there are gray areas between a sure failure and a sure success that are appropriate for a die roll. I can’t agree that skills and skill checks—as they exist in any of the games I’ve played—are the way to handle them.


Related Posts Plugin for WordPress, Blogger...