Roll dice or say "yes"

[Kristian]

There's a piece of text from the rules of the game Dogs in the Vineyard (a really good game, by the way) which is also quoted in the rules book of Burning Wheel. I'd be interested in hearing what you guys think about this. It goes like this:

Every moment of play, roll dice or say "yes."

If nothing is at stake, say "yes" [to the players' request], whatever they're doing. Just go along with them. If they ask for information, give it to them. If they have their characters go somewhere, they're there. If they want it, it's theirs.

Sooner or later - sooner, because [your game's] pregnant with crisis - they'll have their characters do something that someone else won't like. Bang! Something's at stake. Start the conflict and roll the dice.

Roll dice or say "yes."

This is a very hard rule to play by if you're used to traditional, DnD-style gaming. Or at least it has been for me and my group. The principle isn't hard to grasp. But when you use this concept for the first time it makes a radical change in your play style. The (by now) classic example is the locked door:

The characters come to a locked door. They have to get it open to continue, so the thief starts picking the lock. Normally, most GMs (including myself) would go: "Roll a lockpicking check. The difficulty is X" or whatever. The thief rolls and either he gets it or he doesn't. If he gets it, cool, the group continues down the hallway on the other side. If he doesn't, fuck, they're going to have to try and bash it in or find another way.

Here's the thing, though: Everyone at the table wants the characters to get the door open. Having them find another way is just a waste of time. So why are we rolling dice? Years ago, I might have just fudged that roll or told them they'd gotten it open even though the thief didn't roll high enough. Today, that seems kind of silly, but it wasn't until I read that little snippet of advice that I realized there was another way of doing it.

Using the "roll dice or say yes" approach to the same situation, you can either go "Right, don't roll. You unlock it" or you can go "Okay, here's how it is: The difficulty is X. If you succeed, the door opens. If you fail, the door still opens, but you set off an alarm further down the hallway. You'll be through the door, but you'll have a serious fight on your hands."

I realize this isn't the best example, but I think it works. I don't use this method all the time, but in key moments this is great for building up tension. The main thing, for me, is getting rid of all the meaningless rolls that fly across the table during a gaming session. A character is out gathering information about the orc king they're going to visit tomorrow, and fails his check. So he's spent a whole day doing nothing, and wasted game time. What if, instead of "Either you learn something or you don't" it's "You learn soething, but if you roll poorly, the orc king gets word that you've been asking about him." Suddenly the player has to decide whther it's worth it or not.

This goes double for a game like Burning Wheel where rolling skill checks is how you increase your scores, so in that game, every roll has to count.

I'm not saying, always give the players what they're asking for and just stick a hazard on it. And a lot of the time I still use the traditional method. "Oh, you wanna haggle? Roll your bargain skill vs. his willpower. If you fail you don't get it half price." Coming up with the right stakes for "roll dice or say yes" on the fly isn't always easy either. But when you do, it makes for some really exciting moments.

[Delbareth]

I must admit that I'm a little bit disapointed!
I'm always very cautious when I see that everybody forget that the RPG must not only be funny one afternoon but lots of years. What happen to the game pleasure when you know that your GM will let you open any door you want (as far as he want it also), or will let you steal anything, etc...

Ok, I'm not standard neither since I'm a at the very opposite of that type of game. I really think that GM and players build the story. The player aren't there just to execute the will of the GM, and the GM has not an absolute control of the story. I want them to think about the best action without knowing anything of what will do the GM, i.e that I will help them "just because the story is much more cool that way". I want them to succeed because they deserve it and not because it's more fun to succeed than to fail.  And I want them to know that, and they have to do real choices!  As a GM I also do a lot of test of chance (seen by the players) in order to choose between possibilities. It add a lot of suspense for players but also for me. And they know that I'm not the absolute master of their destiny. Several very interesting adventures have begun by something which was not planed (a failure on something the GM wanted to succeed).

I think it's a way to enjoy the game years after years without lassitude.
But it's only my opinion... 

[Kristian]

Hey, guys. Sorry it's taken me a while to respond.

Delbareth, I think maybe I've said this the wrong way, and I know I didn't come up with the best examples above. I'll try and explain it a bit differently, 'cause I think I agree with you on most points. As I wrote, I'm not saying just give the players anything they want. That would get old pretty fast, as there wouldn't be any adversity at all. What I'm saying is, make every roll count and get rid of boring dice rolls, like...

Like... if a player says his character wants to climb up a tree to look around. I would never ask him to roll a climb check. I would just say "alright, you go up the tree," 'cause the alternative would just be boring (i.e. he can't climb up the tree). But if the same player says he's climbing up the wall of the enemy baron's castle to get in behind the guards at the gate, or climbing down a steep cliff to save one of his friends that's hanging on by his fingertips – now you got something worth rolling for! Something's at stake. If he fails, it's got a real consequence.

Also, I think it's worth remembering that the GM is usually the one who creates whatever situation the characters find themselves in at any given point. I know that the players' choices have taken them to that point, but it's the GM who describes the obstacles and the environment (at least in most games), so if a situation turns up that requires the players to succeed at a certain check, then that's just poor GM'ing (at least it seems that way in all the examples I can come up with in my head – I might be wrong, though) and shouldn't have happened.

Does that make sense? What I mean is that, for example, when you as a GM do tests of chance, like you wrote, you're still the one who decides what the possible outcomes are, right? So, while it's still a test of chance, you decide what those chances are. So, whatever the dice decide it's still you, the GM, that's decided what happened. And hopefully whatever happens is cool, 'cause you chose cool stuff as possible outcomes.

This is the same thing I was trying to say in my first post – Namely, whenever you roll dice, every possible outcome should be cool, whether the characters "succeed" or "fail". Otherwise, why are we even rolling? If success = cool story, and failure = endless rerolls, repetitions and boredom, why are we rolling? On the other hand, if success = cool story, and failure = cool story, we've got every reason to roll. Of course it should be better for the characters to succeed, but for the players and the story, failure should be just as good.

I realize I'm not reinventing anything here, just stating something I think needs repeating now and again.

[Delbareth]

Ok Kristian
With these example I'm closer to your point of view. It is obvious that all dice roll should not be done (to walk between puddle, to speak with somebody we disagree with...) Of course! But what I wanted to say is :
1- some rolls should appear to be completely uninterresting at the first sight, but a surprising failure could lead to something interesting
2- if you suppress ALL rolls that are not crucial, you will have some undesired aftermaths (see the other post)

Concerning "test of chance", you say that as a GM I am the one who decide. Yes but... when the result is seen by the player (he throws the dices) and if it's 98 over 100. He know that something terrible should happen. If he's at see (my last session), it can be a sea dragon or some thing very dangerous. But if it's 62, of course he don't know what is the limit in my head... if it's only slightly above (a small problem) or largely above (a big problem), or even under the limit...
Such "test of chance" are important for me. I don't want he think that some monster will attack just because he's at sea. A lot of sea travels can happen without problem and it can be one of these. To keep a certain interest in the game, I think that PC are more lucky (to find treasure) and more unlucky (to have problems) that the standard person, but I don't like when they are powerful magnet for monsters.

In fact, this opinion is mainly due to my experience as a player. I hate to see that the GM will do his maximum to bring me to the death door, but after that a "miracle" occurs which help me to kill the evil guy! I hate to know that whatever clever is my plan, it will always fail just enough to bring me big trouble. I hate to realize that if I act stupidly, the result is the same. I hate to know by advance what will do the GM just because he always react with stereotypes (for examples, to be at sea = to be attacked by pirates).
That's why "test of chance" enable to add something that neither the player nor the GM can predict...

[Drul Morbok]

The post is old, but maybe someone is still interested:
I think the rule is good, but a bit incomplete. It should, in a very precise form, be something like (using some D&dD3 terms):

- If the player is undisturbed, in no special hurry, and can generally perform as he like, assume he rolled an average result. If that would be enough, say "yes". That might apply to the somehow talented climber climbing the most suitable tree around, who canalways search for optimal grip.

- Otherwise let' not forget that you can reroll many skill checks. If a rogue fails at picking a lock, that doesn't mean he'll never be able to open that special lock. It just means he didn't succeed in that round. Unless he fails critically, he may try again next round. So if your player would succeed by rolling often enough, let him succeed without rolling in the time it take to roll often enough. The standard in D&D is 20 rounds, but I'd be flexible enough to reduce the time somehow if actually only a 11 or 12 was needed - should it ever matter.
That might apply to a rogue picking a complex lock that's within his gernal abilities.

Whenever the player is distracted by enemies, in a hurry, threatened by a failure, or in a situation where "there is just this try", role the dice (or let the player roll).

[Kristian]

Hey Drul. I'm still interested. Still struggling with doing this correctly, even after four years. 

As for your elaboration on the rule, sure. I totally agree. I think there was even a rule in DnD 3rd that was close to what you're describing, right? The take 20 rule, or take 10, or something. Basically, we're all agreeing that boring stuff shouldn't happen, and that you should just hand it to the player if it's just a question of time and effort, and nothing else is at stake. Nobody wants to sit there and go "You failed? Okay, roll again, please." twenty-seven times.

The specific quote in the original post is from the game Dogs in the Vineyard, where you only have conflict resolution mechanics. Meaning, there are no rules for task resolution. If you're trying to hit a can fifty meters away with your poistol, the game doesn't care. If you're competing to see whose best at hitting the can, or if hitting the can is somehow important to your proving yourself to someone else, then the game cares, and the mechanics come into play.

But this is a game that's very focused in its scope. The game is about conflict and not much else. When applying the principle to games like DnD, for example, where doors have hit points, and you have details for falling down X number of meters, I use it to cut down on the fluff. In the same way I try to eliminate endless shopping and haggling scenes, by just going "What do you buy? Right. Note it on the sheet and let's go," I try to get to the meat of the story, drama and tension by making sure that, when the dice come out, it's because something cool or important is going to happen.

But the root of this is actually one step earlier, I think. In the locked door example above, the door shouldn't even be there in the first case, if nobody cares about it. If I, as a GM, already know that the thief isn't going to have to roll to pick the lock, why are we even talking about the locked door at all? What I mean is, my example above actually assumes that you're trying to fix a broken situation, rather than illustrating what to do to avoid that situation. The rules for picking a lock shouldn't even figure into my considerations unless I know this lock will be important, just as I shouldn't focus narration on a door that's just going to be open in a minute anyway.

[Laurent MEKKA]

I think it's in 13th age (a take on D20 rules inspired by D&D and by indie games with more modern mechanics) that present the concept of "fail forward".

You roll the dice, if you win, you win, as ever.
If you fail, you manage the action but there's a complication.
Example : You try to open the locked door, you fail the roll, you still open it but you made too much noise and guards are coming. Or you broke the lock and cannot lock the door behind you.
If you were going stealth mode, and the guards are coming, you have to find a way to dispatch them in silence, or to hide, in order not to alert everyone.
It's more interesting than juste keeping the door close and looking for another way that doesn't exist or to retry the roll again and again.

I use this "fail forward" thing as much as possible and it led to more interesting situations.

[David Roomes]

I like that "fail forward" approach. I'll have to remember that. And yes, it's much better to have success with but a complication rather than outright failure. Anything that slows down the game is bad. Even if its realistic. Better to keep things moving forward and keep things fun. Besides, as you point out, complications lead to more fun.

[Drul Morbok]

I also like this approach, since i find it rather...honest:
In a game world, obstacles tend to be there for some story purpose, and the player proup won't accept failure anyway - and more often than not, neither will the DM

I mean...the game is about challenging players, not about wiping out those who fail. Nothing is gained by ending a campaign with all characters' death only because of failed rolls (if bad decisions/roleplaying lead to dice rolls resulting in such an ending, that's an entirely different matter).
And if you place an item inside a locked box, this will rarely result in the group moving on without the item because the group comes to the conclusion that this lock would either be impossible or not worth to open after a failure.