As arbiter of the game world, the referee fields a lot of questions. Whether explicit or implied, these questions are usually of the form “Can I?” Such as “Can I climb the wall?” or “Can I sneak past the guard?” Even questions which don’t seem to follow this form often do. For example, if a player asks “What is the statue made of,” what they’re actually asking is “Can I tell what the statue is made of?” And while the full answer to these questions will be more complex, it will always boil down to either Yes, No, or Maybe.
There’s a well known dictate of improv comedy: “Always say yes.” As a fundamental rule, it’s useful if you need to create a coherent narrative for an audience. Unfortunately, some misguided folk have spread around the idea that the rule works just as well for D&D. It does not, for the simple reason that D&D is not a performance. The game is not meant to move smoothly through the familiar narrative notes of exposition, action, climax, and resolution.
In D&D, it’s important that it be possible to fail. Not just once, at a dramatically important moment, but over and over again until the failure becomes boring and you have to choose between continuing to bang your head against the wall, or going off to do something else. (The beauty of D&D, of course, is that you can go do something else in the game). If all the referee is needed for is to say “yes,” then they should just be another player at the table. The group can all participate in improvised fantasy theater for the amusement of themselves, or an audience. Given the proliferation of minor celebrities streaming their games on twitch, I suspect we’re going to see a lot of that.
That being said, I don’t want to make it sound as though saying “yes” is bad. If anything, I think the above misconception is so particularly dangerous because it’s so close to the truth. If you take the “always say yes” motto and apply it to D&D, you will have a good time. Then, if you’re observant, you’ll start to notice how all these dice and rules and systems are just getting in the way. You might reasonably think they’re getting in the way because they’re bad, when in fact they’re getting in the way because they’re meant to support a D&D game. But if you “always say yes,” you’re not playing D&D; you’re playing an improv game, with unnecessary baggage.
A better dictum might be “Try to say yes.” Think about the situation the players are in. Is the thing the they asked to do something they reasonably could do? If so, then say “yes!” Don’t muddy the game with unnecessary barriers, but do bear in mind what barriers exist, and enforce them. Never say “no” just because saying “yes” would trivialize a challenge. Taking clever action to trivialize a challenge is half the fun of good D&D.
This isn’t all that difficult to do. As referee, you have all the details of the game world in your head. Even the ones you haven’t bothered to come up with yet are in your head. And, as an adult human, you’ve spent a not-insignificant portion of your life observing other humans. You’ve got a good idea of what they’re capable of.
So when your players are in a room with a statue, and they ask “Can I push the statue over?” just look at the image in your own head. What kind of statue were you picturing? Do you think an athletic person could push it over? If so, say “yes”. If not, say “no,” and explain why. Is the statue too heavy, is it bolted to the floor, or is it just magically un-push-overable? If you’re not sure whether a person could push it over or not, say “maybe.” I’ll talk more about maybe a little further down.
First, while we’re still on the subject of “yes,” I want to talk about qualifiers. Qualified yeses give the players complications to overcome, and are almost always more interesting than a simple “yes” or “no.” Which isn’t to say you should invent complications that don’t exist, but you should take a moment to think about the specifics of your player’s proposal. What problems might they encounter?
A chainlink fence is a good example. If your players want to climb over a chainlink fence, you can’t really say “no” to that. Climbing over a chainlink fence is easy. You yourself have probably done it many times. But, it’s also noisy.
Instead of just saying “you make it over the fence,” you can say “Yes, you can climb the fence, but someone may see or hear you.” Razorwire is also a common feature of chainlink fences, so you might say “yes, you can climb the fence, but you’ll take damage from the wire, and there’s a chance you’ll become tangled.” The more you try to spot these hiccups in your players actions, the more your players will think about their actions. Your game challenges them, and they’ll be more engaged with it as a result.
Which brings us to “Maybe.” Maybe is easy: if you don’t know whether you should say “yes” or “no,” roll a die.
In a lot of cases, the die you should roll is spelled out by the rules. “Can I stab the goblin?” roll an attack. “Can I find food in the wilderness?” roll a Bushcraft skill check. These pre-established cases are easy to resolve, but just because the rolls are established in the game’s rules, doesn’t mean the referee shouldn’t consider whether “yes” wouldn’t be a more appropriate answer. “Can I stab the sleeping goblin?” Yes! Anyone who makes you roll for that is an asshole.
It should be noted that the inverse is not generally true. If the rules have established a roll that determines the success or failure of a specific type of action, it’s almost never appropriate to say “no.” Better to simply penalize the roll. After all, skilled foragers may still be able to find food in a barren landscape, it just probably won’t taste super good.
Then there is the other kind of maybe. The ones without any pre-established resolution mechanic. You still need to roll dice, but which ones?
Some folks use roll-under ability score checks. They figure out which of the scores best represents the kind of effort needed to successfully accomplish what the player wants to do, and have the player roll a d20. If the roll is equal to or under their ability score, the check is a success.
Roll under checks are an elegant solution. As far as I’m concerned, they’re just about the only good justification for having ability scores at all. But, since I think the ability scores are kinda sloppy, and want to move away from using them, I avoid this method. Instead, I just pick a chance-in-six that seems appropriate for whatever the player is attempting. I default to a 50/50 chance (1-3 success, 5-6 fail), and modify up or down based on circumstance, and any clever planning the characters put into their attempt.
That is all I have to say. This post is done now.