DM describes a scene, players say what they want to do and how they want to do it, and DM determines success or failure or if a roll is needed because something interesting can happen because of the roll. As much as I reiterate this flow, I still get experienced players who throw down some dice without being asked and announce "Survival 24 for doing blah blah" or whatever. Did I miss something between AD&D (which I played as a kid) and 5E (my return to D&D two years ago) that made this alright?
The Run a Game Blog has a nice piece on this:
I guess I'm looking for ways that other DMs deal with situations where players roll the dice for skills without being asked to do so. What say you?
OK, this is to answer for "ways that other DMs deal with situations where players roll the dice for skills without being asked to do so."
The short answer is that I tell them not to do it. In my house rules, and at session 0, and reminders if they do it during the game. If they still do it, I ignore it. Or if it's something like "I make a Persuasion check to see if I can bribe the guard" I tell them that "guard looks at the die you rolled on the ground and looks unimpressed." It also might work against them, because if their roll fails, they may have consequences they otherwise wouldn't have had if I had decided that based on their actions they simply succeeded.
But really the answer is why it's not a problem at our table, and it has to do with how I adjudicate skill use in general.
I'll tell you that there's a lot in that blog post and the other posts it links to that I disagree with. First and foremost, the blog's approach is very much focused on the game and fixing the "problem" with guidelines and more rules. Rules about what makes a game "fun" and things like that. His long treatise on the history of traps is heavily predicated on the rule that a thief could only disarm "small traps." While that might very well have been the rule, I don't think that's how people perceived or played the rule.
I'm not concerned about designing encounters that are "fun" or that traps are "boring or stupid unless there's some way to avoid them" and other statements like these. Why? Because those all imply that the fun of the game is engaging the rules, and in particular, rolling dice. That is, setting up circumstances where the players need
to roll their dice, and that the action of rolling dice to determine success/failure is what makes the game "fun." That very premise is exactly what encourages players to want
to self-assign die rolls. (An interesting thing is that the "fun formula" doesn't mention anything about rolling dice...).
Because it's a game, we operate under the assumption sometimes that rules must govern the actions of the game. Over the years many people (including myself) have searched for a better "system" for skill checks. The basic line of thinking is typically something like: Combat is fun and exciting. It's fun and exciting because you roll dice, and that creates suspense. In addition, it's not a "one-and-done" thing, it takes multiple attempts, each with a chance for success or failure, and you have to ultimately succeed in defeating the enemy before they defeat you.
So usually it's an attempt to make skill checks as exciting as combat, and it's almost always focused on making die rolls. Because "automatic success is boring" right?
The second aspect is the way skill checks have changed over the years. When non-weapon proficiencies were first introduced, the assumption was that you would rarely need to make a skill check. You were proficient in that skill, and could therefore just do it. Other characters were not, and therefore could not. For example, swimming or reading. If you knew how to swim, you could. If you didn't know how to read, you couldn't. You didn't have to make a check to see if you could read Moby Dick
vs The Little Engine that Could
. And you didn't need to make a swimming check unless you fell off of a ship in full armor, or in a storm.
When you consider skills from this perspective, they become a story element. Oh, we need to get into this building without detection, who knows how to pick a lock?
As time went on, the rules focused on the times when you'd need to make a skill check. That makes sense, because if a check is needed (there's a chance of failure), you needed to know how to determine success. For completeness, perhaps, we even assigned DCs to very easy, easy, average tests of a skill. The example for average (10) in the d20 SRD is "hear an approaching guard." So without any modifiers we've suddenly defined hearing an approaching guard as something with only a 50% chance of success. Is that a guard in mail with weapons and hard boots walking down a stone-floored castle hall from 20 feet away around a corner? Or a guard in leather on grass from 100 feet away?
It's simple. The perception is that if there's a rule, then we use it, and we use it for every circumstance. Sure, the rule says "if there's a chance for failure" but in 3/3.5e there's almost always a chance for failure unless you take 10 or take 20.
I look back to the 2e approach. If you're trained in something, then you're trained. How trained? Well, the passive score answers that question. If you have a passive score of 15, then anything "medium" or lower is automatic success for you barring mitigating circumstances. I take this a step further, incorporating the old take 20 rules. If there is no consequence for failure, and time is not a factor, then you'll automatically succeed at most tasks that you are capable of performing. That is, if the DC is less than or equal to 20 + your modifier, we know you can
do it. So unless there's something that would prevent you from doing it, you do. As the DM I'll probably let you know it took a few minutes, but you'll still accomplish it without a die roll.
This is further modified by your actions. If there's a door concealed behind a pile of boxes, and you stand 20 feet away and look around the room, then chances are you won't find it no matter what. However, if your Investigation skill or perhaps Perception, is high enough, I might tell you that you notice that the placement of the rooms leaves some space unaccounted for between them. On the other hand, if you go over and just start moving boxes, the door is readily visible at that point, and no check is needed. If, however, the door is not just concealed by the boxes, but cleverly hidden, the act of moving the boxes would prompt me to ask for a Perception and/or Investigation check to see if you actually spot the secret door. If you tell me you're going to feel around the seams of the molding and corners, then I might still require an Investigation check, but with advantage because of the actions you're taking.
Another example. Picking a lock. You're in a hallway, trying to get into a locked room. There are guards on patrol, so you need to be quiet, kicking it in isn't really an option. You know that there's a guard nearby, and the lock is a DC 12, and you have a +5 to pick it. I know you'll succeed. However, what isn't clear is how long it takes you to do so. So you have a lookout in place to warn you when the guard is coming, and you start to pick the lock (roll a check, and get a modified 8). That's the only roll you're going to make. You don't know the DC, but since you failed by 4, it's going to take 4 rounds to pick. In the meantime, your lookout is keeping you informed of the guard heading this way, and the other characters are whispering at you to hurry up.
One of the mistakes that has been made in the past is the assumption that anything between the "action" is boring. No need to provide full maps to an area, just the places where you'll have a combat, because you'll breeze past the other parts. Each scene is a set-piece. "OK, you've killed 5 orcs, now it's the rogue's turn to shine...you found a complex trap!"
He doesn't like "gotcha" traps. Not just because of the Perception checks that it now requires, but also because they aren't fun. They're just a "hit point tax" (isn't that what most combats are?).
I disagree. First and foremost, traps are placed where they make sense. A creature doesn't place a trap because it's interesting or fun, and despite the "creatures aren't perfect" attempt to reason why every trap must be detectable, it's a cop out. Here's why:
First, in theory, every trap is detectable. But maybe not by you. The 5e Tomb of Horrors sets the DC to notice the pit traps at 12. I think that's ridiculously low. This is, after all, a tomb that has never successfully been plundered in hundreds if not thousands of years. So much so that it's legendary
. You know what legendary means when there's a location that's known to be loaded with monetary and magical treasure right? (Hint - White Plume Mountain
has a turnstile.
The bottom line is, the traps should be deadly, and they should be hard to detect. They aren't a hit point tax, they are a reward for investing your time, training, and resources into learning a particular skill. Just like in so many movies where there's a ridiculously difficult security system to bypass. It's possible, but you need to have the right people, with the right skills.
But the traps also need to make sense. For example, a goblin lair isn't going to have a DC 30 complex trap that will take 3 experts 2 hours to work painstakingly work through. No, they'll have a pit trap with wooden spikes and poison if they can get it. If the trap is a deterrent, it will also be visible. More importantly, if it's protecting their territory against the nearby troglodyte tribe, then it will be designed to thwart them, not a 7th level human rogue that they never considered as a possibility.
So what about the gotcha traps in the hall like the pits in ToH? Well, to start with, your passive Perception may very well be high enough to notice them. Or at least that something isn't right and they'll take a closer look. Moving at a slower speed and taking more time to search will probably give them advantage on their passive Perception, and if they are using tools like a wooden pole to poke and prod ahead, then they'll just outright find it when they push on it with enough force. This isn't boring, it sets the stage. It tells them something about the environment they are in, and that the creature took precautions against intruders. If you fail to discover it, it might be dangerous, perhaps even deadly.
"But that means that players will always be prodding at things and take forever to get down the hall." Well, yes. If they are someplace like ToH, then yes. What's wrong with that? If you're not rolling the dice for every 5 or 10 feet then it just means that you describe the results of their actions a little differently, and it takes them longer to get down the hall. I think we usually underestimate the passage of time in our games anyway.
The other main issue I have with how he recommends you use skill checks is that it overemphasizes the skill check. From what I consider a world-building approach, I like things to be consistent and make sense. This also applies to application of the rules. There's a lot of advice out there like his post that says something to the effect of "don't require a skill check unless there's a potential consequence or it's important to the story." This gets extended further with advice such as to not place a locked door when there isn't something worthwhile inside.
Well, since the players write the story through their character's actions, I can't really tell you what's important or not. If you're exploring a secret laboratory, then different people have different access to different rooms. Most rooms are locked, even if there isn't anything of interest to you. Again, like traps, a locked door should make sense. Not in the context of "is it fun?" but in the context of the setting and scenario. But wait, didn't I say I only have them make a skill check when there's a potential consequence? No, and that's one of the major differences in my opinion. They are always
making skill checks. That is, as a DM I am always measuring the DC against their skill level. The only question is whether the circumstances call for actually rolling a die. That's the judgement call, and can only be made by the DM. We don't call for a balance check ever time somebody walks across the room. But walking across a narrow ledge above a 1,500 foot drop? Sure. If there's a significant chance of failure. It might even be a circumstance where it warrants a skill check or two to use the die rolls to specifically build suspense. Based on their skills you know they'll succeed, so any failure won't result in actually falling to their death, perhaps, but it could have other consequences, or at the very least remind them of their mortality. The point is, even if you're going to utilize a skill check for dramatic purposes, it's a decision the DM makes, not the players.
The bottom line for me is that skill use isn't generally about rolling dice, and I don't think it should be. I think that most of the time, if you're skilled at doing something, then you're just able to do it. It's a tool to help determine success only when success isn't assured, and is needed rarely for somebody who is skilled at something. If you're not proficient, that's a different story. You probably won't succeed without a check, and I'll let you know when you need to make one.