People Who Don't Like Dogs In the Vineyard: Anonymous

by Dan H

Dan forgets, for the moment, that this isn't actually an RPG site.
~
Those of you who keep up with trends in the modern Indie RPG industry will probably have heard of a game called Dogs in the Vineyard by a man who goes by the unlikely name of D Vincent Baker. Those of you who don't keep up with trends in the modern Indie RPG industry, well to be honest you probably won't understand very much of this article.

Dogs in the Vineyard has received a large amount of not-wholly-undeserved praise directed primarily at its innovative dice mechanics and its refreshing approach to the way moral issues are confronted in the medium of a roleplaying game.

A lot of people I know really, really like this game. Pretty much everybody on the internet really really likes this game. I've played it, I've read it, I've really tried to get to grips with it, but:

Hello, my name is Dan, and I Don't Like Dogs In The Vineyard.

It's not the setting. I absolutely love the setting (such as it is), the idea of playing wild west Mormon inquisitors is pretty damned cool, and going from town to town laying the righteous smackdown on the prideful, the sinner, and the sorcerer while dressed in a patchwork greatcoat is pretty damned cool.

The things that really bug me are the system, and the philosophy which might or might not underly the system. Perhaps the best way to illustrate the sort of thing I'm talking about would be to provide some salient quotes. Let's take this, from the section of the introduction entitled "What's it like to Play?"
"It may seem odd at first, but the rules are there to support you and make it easy. I can't wait to show you the dice in action! And the payoff is terrific blood, passion, judgement, fire. Real, gripping drama all the way around the table." D. Vincent Baker, Dogs in the Vineyard, p7

Now, those of you that pay attention to the Indie RPG scene are probably aware that "System Does Matter!" is something of a rallying cry for the movement. That being the case, maybe it isn't so surprising that D. Vincent chooses to open his book with a paean to his dice mechanics, but really: "I can't wait to show you the dice in action"? Unless they strip or juggle or something they're just not going to be that impressive. Also: perhaps I'm being a bit too British here, but as a customer, I rather think it's my job to say whether the payoff in this game is terrific or not.

This little bit of gushing sets the tone for much of the rest of the book, Baker keeps lapsing into this fanboyish prose style that makes you feel like you've been cornered at a party by an overenthusiastic geek who won't stop telling you about his homebrew system. And haven't we all been there. (Hell, let's be honest, haven't we all been the overenthusiastic geek).

But to be honest, I could forgive all that, after all there's nothing wrong with a bit of enthusiasm. The problem is that, in my opinion at least, Dogs in the Vineyard seriously fails to deliver on the above promises. Later in the introduction, Baker describes in more detail what the system is supposed to deliver. He tells us that:
"The game's rules' job is to help you, the GM, reveal the pride, sin and corruption in the towns you create ... they work a) by helping you create congregations in turmoil, then b) by seizing conflicts and relentlessly escalating them, then c) by bringing the consequences back home to the players."
D. Vincent Baker, Dogs in the Vineyard, page 9

The problem is that the rules do precisely none of these things. Okay, that's unfair. They do, in fact, help you create congregations in turmoil, but no more than any other game with a moderately detailed setting will. Where the game particularly fails, however, is in "seizing and escalating conflict." This flaw in the game is particularly insidious because "conflict" and "escalation" are real English words with subtle narrative implications, but they are also mechanical concepts in the game, with meanings which are almost but not quite totally unlike the actual narrative concept of an "escalating conflict".

This would not be a problem in and of itself, except that Baker persistently treats game mechanical "Conflict" and "Escalation" as being identical to, and adequate substitutes for, the genuine escalation of a meaningful conflict.

For those who aren't familiar with the way it works, "Conflict" in Dogs in the Vineyard, like in most RPGs which use a conflict resolution system, essentially means "the character is trying to do something which might not work" and "Escalation" is defined on an extremely rigid scale (verbal, physical non-violent, physical violent, gunslinging). This is exacerbated by advice which encourages the GM to push for small stakes in any given "Conflict" making a DitV "Conflict" much more similar to an ordinary RPG "Task" than to anything resembling a meaningful component of a story.

I should probably take a step back here, since I've not really defined my terms hugely well. I haven't actually said what I think "conflict" and "escalation" mean outside the context of the Dogs in the Vineyard core rulebook. Part of the reason for this is that I genuinely don't know. Contrary to what various internet pundits might have you believe, the essential elements of narrative are not easy to define, either in general or in terms of a roleplaying game. If forced to try, I would probably define "conflict" as a tension between two opposed forces or concepts which provides the central drive of a story, and "escalation" as the increasingly extreme acts to which the conflict drives the participants.

The DitV rulebook never directly defines its use of the word "conflict." It seems to use it to mean "situation in which the outcome is uncertain". Examples given for suitable conflicts include "does your brother kill this woman," "do you outshoot the shooting instructor," "do you learn to ride" and "do you get murdered in your bed." They are all, in short, situations in which your character is trying to do something difficult and might fail. This seems, to me, to be functionally identical to every roleplaying game I have ever seen. Actually, there's an awful lot in Dogs which is functionally identical to every roleplaying game I have ever seen. Take, for example, its "relentless escalation" of conflict.

The system in DitV is heavily focused on this concept of "escalation". You start off having a conversation, then somebody shoves somebody else, then somebody throws a punch, then somebody pulls a gun. Now to give the game its due, it does produce exactly those sorts of situations: an argument turns into a fistfight which turns into a gunfight. Once again, though, this is exactly the same as every other roleplaying system I have ever encountered. The difference is that in Dogs the escalation is forced by the system, )you have to pull a gun in order to get more dice in your pool), while in a regular RPG the escalation is caused by the sorts of things that actually cause conflicts to escalate in real life (one approach to a situation has failed so we try another, more extreme approach). Because in Dogs shooting somebody is game mechanically identical to talking to them (and indeed, if you started off talking, your interlocutor gets to keep all of his "talking" dice) escalation doesn't really feel like escalation, it feels like a game mechanic. By contrast in a regular RPG pulling a gun actually invokes a whole different set of game mechanics: you literally take the conflict into a different arena.

In other words, I actually think that the conflict-resolution mechanic in DitV provides less drama, less blood, fire, and passion, than a regular RPG task resolution system.

The third claim Baker makes for his mechanics is that they "bring the consequences back home to the players." They do this by the means of "Fallout Dice". This bit is kind of fiddly, so I'll quickly summarise the DitV dice mechanics:

Everybody involved in the "conflict" rolls a number of dice determined by their stats. The initiator of the conflict then "raises" with two of his dice. Everybody affected by that "raise" has to "see" by using dice from their pool the sum of which at least equals the initial "raise." If you use one die to see you "turn the blow" which is good for you, if you use two dice you "Block or Dodge" and if you use three or more dice you "take the blow" which means you take "Fallout". This last concept "Fallout" is what Baker calls "bringing the consequences back home to the players." The problem is that it doesn't.

When your character "takes the blow" you, the player, describe what happens. Does a cutting barb give you a moment's pause? Does a bullet catch you full in the chest? However, you don't actually roll for Fallout until after the conflict is over, so you don't actually know of that bullet is going to kill you, or just graze you a little until well after you describe the hit.

Fallout can do a number of things to your character and you, the player, get to pick exactly what the effects are (it needs to be something appropriate, and it would be churlish to point out that you could deliberately choose ludicrous consequences in order to make the game seem absurd). Fallout can manifest in a number of ways, you can have your stats reduced, you can lose your gear, or you can gain new Traits or new Relationships rated at 1D4 which Baker, incorrectly, believes to be a bad thing. If you take enough Fallout, you might die, although in practice this usually only happens with guns.

Now I can see two major problems with the Fallout system. The first is small, and you can legitimately overcome it by playing the "it's not that sort of game" card. The first problem is that there is absolutely no reason whatsoever not to take the 1D4 trait or relationship as your Fallout. Contrary to what D. Vincent Baker might think, the way the dice mechanics work rolling a D4 while less good than rolling a D6 or a D10 , is better than rolling no dice at all. Since there is no limit on the number of dice you can roll, and no limit on the number of traits you can bring into play in a single raise, and no way to force somebody to roll one trait instead of another, it's always better to have a trait than not to. Furthermore, Traits are always going to be the easiest types of fallout to justify, because a Trait can be anything. If you get shot up real bad, then you can get an "I got shot up real bad" trait. If you get made to feel momentarily guilty by a particularly articulate townsperson you can get a "moment of guilt" trait.

Now if you take too much fallout, things stop being so much fun, your character can get seriously injured or die. Just like wait for it just like in every other roleplaying game you have ever played! The system for PC death in DitV is, contrary to what D. Vincent Baker might think, not much different to the system in any other RPG. It is true that in DitV a character never "takes the blow" unless he chooses to, he can always "Give" on the conflict, but again the same is true in any other RPG, it's the players that choose to start fights, that choose to tick off guys with guns. Furthermore, the GM in Dogs in the Vineyard is absolutely within his rights to introduce a conflict in which the stakes are "does your character get killed?" making the "Fallout is optional" thing rather academic.

Again, this might be best illustrated with an example, from the example "Ambush." What's at stake in this conflict is apparently "do you get murdered in your bed." Here is what D. Vincent Baker has to say:
"My first raise will be to hit you in the head with my axe, I get my Axe dice too! I'm rolling a lot more dice than you, so you'll probably have to Take the Blow. But check it out that means you take Fallout and you get to say how, it doesn't mean you're dead. You aren't dead unless the whole conflict goes my way." D. Vincent Baker, Dogs in the Vineyard, p89

Except that he's wrong. Well, he's not exactly wrong, but he's forgotten two quite important things about the way his dice mechanics work. Firstly, he's forgotten that since the axe-wielding maniac has started with a huge advantage, the conflict actually is fairly likely to go his way, which will kill the character. Heck, if the axe-wielding maniac had rolled really well, the player might have been unable to See the initial Raise, and might have just had his head cut off. Secondly, assuming the guy did, in fact, Take the Blow, he will be taking at least 3D8 in Fallout, probably more, which means he is actually extremely likely to wind up with a potentially fatal injury anyway.

As if his persistent failure to actually understand his own game mechanics wasn't enough, he then rounds the example out with this little gem:
"In most roleplaying games, saying "an enemy sneaks into your room in the middle of the night and whacks you in the head with an axe" is cheating. I've hosed the character and the player with no warning and no way out. Not in Dogs, though: the resolution rules are built to handle it. I don't have to pull my punches! (You've GMed a lot of RPGs before, right? Think about what I just said for a minute. You know how you usually pull your punches?)" D. Vincent Baker, Dogs in the Vineyard, p89

Leaving aside the fact that any sentence which begins "in most roleplaying games" is always going to be a pile of toss, the resolution rules in Dogs are not remotely built to handle people being murdered in their beds, because a massively important aspect of the system is the ability to Give on a conflict, to say "no, this isn't worth taking an axe in the head over." The Ambush doesn't give the player that option.

Also, once again, Dogs in the Vineyard is doing nothing that you can't do in an ordinary RPG. It still boils down to "succeed at this dice roll or series of dice rolls, or you die" and if the GM doesn't "pull his punches" killing the PC with this kind of ambush would be trivially easy: he's got Body 5, Heart 5, the axe is Big and Excellent, he's got the Axe-Wielding Maniac trait at 1D10, the Stealthy trait at 2D8 and so on. The only reason the player character could survive this kind of encounter would be because the GM made it easy for them. You can do exactly the same thing in D&D: the Orc sneaks into your bedchamber, make a DC 15 Listen check to hear him, if you don't hear him he'll need to make a normal to hit roll against an AC of 10, then he'll just do standard damage for his weapon, which any moderately experienced Adventurer will survive easily, then you wake up and can begin to fight normally.

In an ordinary game, having a PC ambushed in their sleep puts them at a large disadvantage. In Dogs in the Vineyard having a PC ambushed in their sleep puts them at a large disadvantage.

And indeed, the closer I've looked at Dogs in the Vineyard, the less and less I can see in it that is actually different to a regular RPG. Perhaps the final nail in the coffin comes from the "Actual Play" example at the back. I won't give the full quote, because it's very long, but what it boils down to is this: there was a problem in a town (the Branch Steward was having an affair with a member of his congregation who was actually in love with somebody else, this was a violation of doctrine that let the demons into the town, hilarity ensued). The players came to town, investigated, and came up with a solution (they packed the Steward off to Bridal Falls City, lied to his congregation about why they were doing it, and set up the girl with the guy she was after). On the way they fought a couple of possessed people. Now admittedly, I'm not reproducing the whole account, and admittedly, I'm being a little flippant in my presentation, but let's just go back to "What The Game Is Like" again: "Blood, passion, judgement, fire. Real, gripping drama all the way around the table" is what we were promised, but the end result seems to look an awful lot like every RPG I have ever been in.

One last quote to seal the deal, from the end of the Actual Play section. This is a quote-within-a-quote, in which D. Vincent Baker describes to us how his players described that session to him after the game:
"Later Tom wrote to me and said "Yeah, your description implied she was only doing it because he was the Steward, and it was cast specifically as 'Brother Malachi is abusing his position`, so it was a pretty natural progression." In fact, in my description, all I'd done is emphasise that he wasn't raping her. Meg and Tom between them had judged Brother Malachi so immediately and so viscerally that his guilt was objective and foregone. And then they lied to the whole congregation to protect a possible future between Avigail and Jonas! Isn't that fascinating? And unbelievably cool? Friend, that's why I play this game." D. Vincent Baker, Dogs in the Vineyard p145

Now there's a whole lot of things here which I find interesting.

Notice that Tom's description of events differs considerably from D. Vincent's. D. Vincent says that Meg and Tom (the players) "Judged Brother Malachi immediately and viscerally." The way Tom describes it, however, it sounds a lot more like he and Meg had just tried to work out what the GM wanted them to do. "Your description implied she was only doing it because he was the Steward" says Tom. Nothing Tom says implies to me that he felt empowered to pass judgement on the NPCs in any way other than one sanctioned by the GM at the start of the game. Nothing implied he was doing anything at all pro-active. The players were presented with a situation by the GM, they investigated the situation the GM presented, and then they took the solution that they felt the GM was leading them towards.

Occasionally, I wonder if my issues with Dogs in the Vineyard are a sign that I'm not capable of dealing with "modern" RPG mechanics, if I can't get over the fact that "I'm a really crappy shot: 1D10" actually makes you better with a gun than "I'm a really good shot: 1D6", or if I chafe at the idea that the consequences of PC actions might be player-controlled and largely positive. In reality, though, it isn't that at all. What bothers me about Dogs in the Vineyard is the very conventionality of it. Sure your character might have traits like "I'm nervous around women" or "I can't shoot straight", but you still achieve your in-character goals (which the game still assumes are identical to your out-of-character goals) by succeeding at in-character actions. You still have to worry about getting killed if somebody fires a gun at you, and judging by the actual play, you still basically run around investigating stuff and trying to work out how to solve arbitrary problems.

And that, Friend, is why I don't play this game.
~

bookmark this with - facebook - delicious - digg - stumbleupon - reddit

~
Comments (go to latest)
Jen Spencer at 15:48 on 2007-03-29
How did you feel about Care Bears in the Vineyard?
Wardog at 15:51 on 2007-03-29
Care Bears in the Vineyard sounded WELL COOL - I think the problem is the game is inherently pretty exciting, it's just the system doesn't support it. I have to play Dogs at some point. At the moment I'm just voicing randomness.
Dan H at 11:01 on 2007-04-02
Care Bares in the Vineyard was cool, but it highlighted a lot of the issues with the system. Instead of thinking to myself "What would my character do in this situation" I was thinking "how can I use my 'Obscurantism' trait". Similarly, I didn't escalate conflicts because I cared about them in character, I escalated conflicts as an alternative to taking Fallout. It was all good fun, but it wasn't "Blood, fire, judgement: real gripping drama all the way around the table."
I stumbled across this article looking for somthing else,but it makes me wonder if you ever revisited this game?

From my perspective, as someone that still loves this game and ranks it as one of the best, you appear to be viewing the game text through a distorted lens. Where you see a game that doesn't actually depart from standard games and does things strangely, I see a game that rarely if ever overlaps with my more traditional gaming background and does things elegantly.

I agree the text of the game is rubbish. The style is just outright annoying and it actually serves to obscure some of the game's basic procedures. For example the escalation rules you describe are not correct but ninety percent of readers of the text make the same mistake. They assume escalation to be only one direction from talking to gunfire when you can escalate from any arena to any other. This is a problem with the text, it needs a total rewrite in my opinion.

Anyway, my comments are not aimed at converting you, more wondering if you even looked at the game again?
Orion at 14:22 on 2015-03-16
This game does have one or two nice features that conventional/old school RPGs don't. The big one, for me, is the ability to play out one "conflict" to settle something that another game would model as a huge number of discrete actions or not model at all. In Dungeons & Dragons, sneaking through an enemy camp requires some arbitrary, unknowable, and likely very large number of hide and move silently checks. Fights with multiple participants grind through hideous numbers of die rolls. Whether a character learns a new skill is either deterministic or by fiat. It's genuinely awesome to have a game where "can I infiltrate the enemy camp?", "can I beat up the gang of thugs?", and "can I learn to ride?" are able playable "conflicts."

Unfortunately, Dogs isn't the only game that does this; lots and lots of indie games do, including better games from the same designer. Dogs isn't your only option for that feature, and it's not even a good option. Whatever gains in speed of play you might hope to gain from reducing "fight the gang of thugs" to one conflict, you lose when you realize that resolving a single conflict may require an absurd number of operations, and that the game actually encourages players to drag out every damn scene by rewarding the mechanically for filibustering. If you just keep changing the topic frequently enough to touch on everything in your backstory, you will eventually roll all the dice on your sheet and win. Make sure to carry a bag full of weapon, swing each one once, then drop it and grab the next one.
Arthur B at 15:51 on 2015-03-16
This game does have one or two nice features that conventional/old school RPGs don't. The big one, for me, is the ability to play out one "conflict" to settle something that another game would model as a huge number of discrete actions or not model at all. In Dungeons & Dragons, sneaking through an enemy camp requires some arbitrary, unknowable, and likely very large number of hide and move silently checks. Fights with multiple participants grind through hideous numbers of die rolls. Whether a character learns a new skill is either deterministic or by fiat. It's genuinely awesome to have a game where "can I infiltrate the enemy camp?", "can I beat up the gang of thugs?", and "can I learn to ride?" are able playable "conflicts."

To be fair, I think a certain amount of the above either comes to received wisdom in terms of refereeing and/or designing adventures for D&D rather than the actual written rules, a certain amount comes down to the axiomatic assumptions of the rules, and a certain amount is orthogonal to D&D's concerns. To break down each of your examples:

- I am not aware of any edition of D&D which specifies hard and fast rules requiring heaps of rolls to infiltrate the enemy camp. In fact, I'd say that in most recent editions I'd say that good DMing practice would involve just setting a single difficulty level for that and having the character make the appropriate roll, unless you wanted to do something like have an onion layer-type setup, where the deeper the PC penetrates the more beneficial stuff they get out of the infiltration but the greater the danger they are in if they're caught. The latter case is the sort of thing which the extremely binary setup of Dogs is ill-suited for, whereas a game like D&D can handle that more granular situation by adjusting the target number or applying penalties to die rolls accordingly. (Of course, in an investigation-and-espionage themed game, resolving the infiltration of an enemy base in a single die roll would miss the entire point of play, much as rolling "do you defeat the dungeon?" as a single conflict would ruin old-school dungeon crawling.)

- Fights typically requiring an extended series of die rolls to resolve is a specific game design decision. In general, the more rolls you make to resolve any particular fight in a game, the more likely it is that stronger enemies will beat weaker enemies and superior tactics will overcome inferior tactics, because whilst a single roll here or there may be a fluke over time a series of rolls will tend towards the average. Conversely, if a conflict is resolved in a single (or very limited number) of die rolls, suddenly blind luck becomes a much greater factor and flukey results are more common.

The former is better for a game which emphasises clever tactics on the part of the players and rewards efforts made to stack the odds on their part before combat is joined (and, indeed, to encourage players to back down from or avoid combats where the odds are clearly against then). The latter is better for a game where a sudden reversal of fortune could happen to anyone, there's only so much massaging of the odds you can do, and entering combat is always highly risky. It isn't really a flaw of D&D that it chooses to be the former sort of game rather than the latter and then makes appropriate design choices to implement that.

- Learning new skills is the sort of thing which the assumed timescale of a D&D adventure doesn't really allow for and tends to be kept for downtime. "Do I already know how to ride?" as a related question can come up and is, as you say, either deterministic or based off fiat, though at the same time pretty much every RPG session you'll encounter includes a bunch of details which are at least deterministic or based off fiat; for instance, the design of a Dogs adventure is largely deterministic on the part of the referee, and so far as I can recall you can't use a conflict in Dogs to actively contradict a fact about the gameworld the GM has already established.

To a large extent, though, pretty much any conflict in any RPG can be reduced to a single roll: simply get the group's agreement that nobody particularly cares to spend a lot of time on this matter and have someone make an appropriate roll using the game's typical action resolution system and an appropriate difficulty level. Boom, done.
Orion at 17:50 on 2015-03-16
I've played and run D&D 3.5 extensively. I'm coming to that conversation as an outsider or as a detractor; I would say that D&D 3.5 is one of the best RPGs ever made. One of the assets of third edition D&D was that it had actually quite detailed rules for a number of things, which made it substantially less subject to fiat ruling than almost any other system. Unfortunately, some of those rules simply don't work, but are still explicit enough to seem binding. Stealth and perception are probably the most egregious examples. If you want to creep past a room with 5 guards in it, then you'll be making a hide and move silently check, and the DM will be rolling 5 spot and 5 listen checks. As you go through the complex, it's not really clear whether and when you need to roll more hide checks, but it is clear that the DM is rolling dice for every observer. That's both extremely time-consuming and extremely deterministic. I've both played in and run games where we played it this way, but it resulted in everyone choosing to not use those skills very often. You certainly could choose to bypass those rules entirely and let someone make a single roll (hide? move silently? one of each?) to learn some arbitrary information, in the sense that you could do anything your group agrees will improve their experience but that's really a level of divergence from the explicit rules that I think it passes beyond "adjudication" to become "changing the rules."

There is absolutely nothing wrong with a game that provides for complex tactics and in which resolving combat is intended to be time-consuming because it's intended to be enjoyable. I like that kind of game, so I play a lot of D&D. There also ought to be games where fights can be resolved relatively quickly in order to get back to the action. That's an obvious idea that few games have really delivered on, so I think it's worth taking note of when someone does pull it off. Most games that are supposed to be "less combat oriented" end up like Vampire, where fighting is about as complicated as in D&D but much stupider.

Although I want to reduce reliance on fiat, it's not because I have any great distrust of GMs. You're quite right to note that the conflict rules in Dogs actually do very little to tie the GM's hands, and players still can't easily swim "against the stream" of the GM's intentions. I like rules that cut down on fiat because reliance on fiat actually imposes a cost on the GM as well as on the players. It's better to outsource as much decision-mkaing to the dice as possible, so that the GM can spend more of their attention and creativity on the bits that are most important to them. Rolling dice, even for something basic like "is my character a good rider," may well generate a more interesting answer than anything the player OR the GM would have come up with on their own.

Again, I wouldn't recommend actually playing Dogs. You really don't have to look far to find a better game; even Vincent's following project, Apocalypse World, has almost all the upsides and only some of the downsides; I just do think the folks who found it new and exciting were onto something.
Arthur B at 18:29 on 2015-03-16
I've played and run D&D 3.5 extensively. I'm coming to that conversation as an outsider or as a detractor; I would say that D&D 3.5 is one of the best RPGs ever made.

Sure, but equally it seems odd to compare D&D to a game which isn't trying to do what D&D is trying to do when there's ample other games that came before Dogs which tried to do what it tries to do to varying levels of effectiveness.

Unfortunately, some of those rules simply don't work, but are still explicit enough to seem binding. Stealth and perception are probably the most egregious examples. If you want to creep past a room with 5 guards in it, then you'll be making a hide and move silently check, and the DM will be rolling 5 spot and 5 listen checks. As you go through the complex, it's not really clear whether and when you need to roll more hide checks, but it is clear that the DM is rolling dice for every observer.

FWIW, 5E has made this very elegant - the skill list has been greatly rationalised so you just have a flat "Stealth" skill, and observers have a "passive perception" score which is effectively the DC required to sneak past them, so to sneak past 5 guards you just roll Stealth and see if you can beat the best passive perception score amongst them.

You certainly could choose to bypass those rules entirely and let someone make a single roll (hide? move silently? one of each?) to learn some arbitrary information, in the sense that you could do anything your group agrees will improve their experience but that's really a level of divergence from the explicit rules that I think it passes beyond "adjudication" to become "changing the rules."

This may be where our differing experience of D&D comes into play; I've tended to avoid 3.5 precisely because of the preponderance of explicit rules that previous editions didn't need and subsequent editions have quite effectively streamlined (though 4E and 5E make very different decisions about just what to streamline).

It's better to outsource as much decision-mkaing to the dice as possible, so that the GM can spend more of their attention and creativity on the bits that are most important to them.

What if decision-making (as a whole or on specific subjects) is the part that is actually important to the GM?
Orion at 20:12 on 2015-03-16
Stealth is always a tricky case to end up debating, because it's an extremely hard design problem that almost no designer has ever gotten right. It's especially difficult to discuss because it's a classic case where players (here including game masters) tend to skim the relevant sections and assume it said something that made sense to them, rather than engaging with what's there. For example, this:

observers have a "passive perception" score which is effectively the DC required to sneak past them, so to sneak past 5 guards you just roll Stealth and see if you can beat the best passive perception score amongst them.

Is just wrong. The sidebar on page 60 says that you can use Stealth to hide from anyone whose passive perception you can beat, but it also says twice that you will be revealed if you make noise. There's no rule given for for being quiet. Because the language is focused on the sneaking character ("if you make noise, you give away your position") rather than on the perceiving character ("you may be detected if an nearby creature can hear you"), I think the most defensible reading would be to default back to the general-case for skill use, where the DM assigns a DC based on the perceived difficulty of the task. So, if the orc guards have 12 perception, you may only need to roll a 12 on stealth to hide from them, but actually getting past them might require a DC 20 (hard) "creep through piles of dead leaves" check. It's not clear whether this would be the same die roll or a second roll, although let's be charitable and assume that they want you to do it the reasonable way (with one roll).

It seems so obvious that there should be a "beat the guards' passive perception to sneak past them" rule somewhere that it's easy to assume it must be there, but it's just not. The plain text of page 60 tells you that stealth is the skill you would use "slink past guards," but not what you would need to roll. Page 69 says that if you beat the enemy's passive perception, you can surprise them in combat, but it doesn't say that you could avoid the combat instead. There actually is no rule empowering a PC to do any kind of scouting, burglary, or infiltration.
Arthur B at 20:57 on 2015-03-16
I think you are either cherry-picking a lot from the sidebar on page 60 or using an out-of-date version of the basic rules PDF - on the current version, at least, the full text paints a very different picture.

On making noise, it specifically clarifies making noise as doing something like knocking over a vase or shouting a warning; it also says you remain hidden as long as you are quiet, which is not the same thing as silent. In addition, contested rolls are only called for for any creature that "actively searches for signs of your presence" - there's a whole passive perception section which specifically states that passive perception is the DC to avoid notice by creatures who are not actively looking for you.

It's also manifestly untrue that there's no rule empowering a PC to do any sneaky stuff - on page 60 of the basic rules it specifically says "Make a Dexterity (Stealth) check when you attempt to conceal yourself from enemies, slink past guards, slip away without being noticed, or sneak up on someone without being seen or heard." "Slip away without being noticed" would allow avoidance of combat; "slink past guards" covers exactly the sort of situation we have been talking about.
Orion at 22:49 on 2015-03-16
I'm using the basic rules pdf, v 0.2, which I've just downloaded from the Wizards website today.

It's also manifestly untrue that there's no rule empowering a PC to do any sneaky stuff - on page 60 of the basic rules it specifically says "Make a Dexterity (Stealth) check when you attempt to conceal yourself from enemies, slink past guards, slip away without being noticed, or sneak up on someone without being seen or heard."

If all a game does is tell me what stat or skill would be relevant, I don't actually count that as a rule. I look at it this way: in a "no rules" freeform session, each PC can do whatever the DM feels it's reasonable for that PC to do. Assigning numbers to the PC's skills doesn't fix that by itself. If the DM is picking arbitrary DCs between 5 and 30, PCs can still only do whatever the Dm feels like it's reasonable for them to do. The main reasons we add rules are to give the players some ability to predict what their PCs are capable of, and something to appeal to if their ideas about the direction a scene should go are out of sync with the DM.

I understand that you feel 5E's stealth rules are more robust than that, and I'm interested in that discussion, but I think this is a sufficiently important point of method that it's worth discussing on its own.
Arthur B at 13:54 on 2015-03-17
I understand that you feel 5E's stealth rules are more robust than that, and I'm interested in that discussion, but I think this is a sufficiently important point of method that it's worth discussing on its own.

I think it is worth breaking this apart from the stealth discussion, not least because we've established that stealth in 5E doesn't use arbitrary DCs - the DC for avoiding the notice of someone who isn't actively looking for you is their passive perception, and someone who is actively searching gets to roll a contested Stealth-vs-Perception test with you.

The way I see it, any tabletop RPG boils down to an extended conversation between the referee and the players; this being the case I don't see "the PCs can only do what the DM feels it's reasonable for them to do" to be a problem, because the process of negotiating with the referee and convincing them to set a favourable DC or provide you with suitable bonuses to your roll and so on is a legitimate part of that conversation.

If anything, I would say that this capacity to negotiate with the referee is a necessary fix for the inflexibility of the rules, rather than inflexible rules being a defence against a hostile referee. If the outcomes you are getting in a game flies in the face of your understanding of your character's capabilities, that usually comes down to one of two problem:

- As regularly happens with White Wolf games, the fluff about what you're supposed to be able to accomplish with a particular skill level doesn't really map to your chances of success according to the rules as written.
- The referee themselves has a decidedly different picture of your character's capabilities than you do.

Detailed rules to fall back on can solve the second problem if the system actually supports your vision for the character, but they can't solve the first problem, whereas loose rules combined with a referee open to negotiation can solve both problems.

Moreover, DCs fixed by the rules don't really solve the problem of referees using unfavourable odds to shut down player actions they aren't interested in, because the referee is still free to throw whatever opposition they like at the PCs, however unfair. (See, for instance, the axe-wielding night visitor Dan talks about in this article.) In a loose system a mean referee can just set the DC to hit a goblin at 30; in a tight system a mean referee just substitutes a more impervious monster in place of the goblin.

There is no rule-based solution to playing with someone that isn't fun to play with. You just chalk it up to experience and stop playing with them (or if they're a lousy referee but fun as a player, stop playing games where they are refereeing).
Orion at 22:04 on 2015-03-17
Let's set aside the hostile-referee case, because it's not that interesting. I do think that rule-driven DCs help there too (they make it more obvious that you're being set up to fail, and they give you the chance to succeed by doing something the MC didn't expect and thus didn't foreclose), but a bad ref still makes for a bad game.

So, let's assume a referee of good faith whose sensibilities are similar to that of the players. In fact, let's go to the opposite extreme -- suppose that every negotiation ends in complete agreement and that no one is dissatisfied with any of the rulings made. I want to argue that even in this case, mechanics that require negotiation imposes substantial costs. When you depend on the referee's adjudication, you don't know what you can do until you check with your ref. To do that, you need to get your ref's attention and they need to spend at least a moment of time considering it. If a combat scene or action scene is happening, that means you may not be able to decide your move in advance, because you may need to wait for your turn to come around so you can ask what DC your MC would set for some jump, or swim, or climb you may be considering. Outside of combat, when your group is making larger-scale plans, rule-generated DCs are even more beneficial for two reasons. First, as in combat, it just saves time if you know what you can do without having to ask. Second, more importantly, it saves you from screwing yourself over on what I like to call "implied tasks" or "secondary tasks." Basically, a primary task is something your character freely chooses to do. You can always ask the referee immediately before choosing it what your chance of success is. An implied task is a check you end up being forced to make because of an earlier choice. Implied tasks are dangerous because we rarely think to ask about them, and rely more on unstated assumptions.
Swimming is a great example of an implied task. If there's a rapid river below my character, and I'm thinking of trying to cross the river on a narrow beam, I'm probably going to ask how difficult a balance roll it would be. I'm less likely to remember to ask how difficult the swim roll would be if I fell in. In that situation, while my focus is elsewhere, I'm likely to just assume that my character can or can't handle the rapids and that my referee probably feels the same way. If I assume "can" and the ref doesn't feel the same way, I could get myself into trouble. On the other hand, if the DC for an armored human to stay afloat in rough water is something I can just look up, I would have already check whether my character can survive a dunk in the pool, and if my character cannot survive taking a dip, I would know that going in an be sure not to put myself in that situation.

Basically, I want to say that the downside of negotiating with your referee isn't the risk that you might not get your way, it's the time spend doing the negotiation.
Arthur B at 11:52 on 2015-03-18
I want to argue that even in this case, mechanics that require negotiation imposes substantial costs. When you depend on the referee's adjudication, you don't know what you can do until you check with your ref.

When you depend on static DCs set by the rules, you don't know what they are until you check with the ref. Sure, you might know that goblins have AC (whatever) as standard, but what if the goblins' wizard ally has enchanted their armour without you knowing it? In practice you always need to check what the current DC for a task is with the referee anyway because you don't have the same access to information the referee has except through discussion with the referee.

To do that, you need to get your ref's attention and they need to spend at least a moment of time considering it.

To do *anything* in an RPG you need to get the ref's attention and they need to consider it. What tabletop RPGs have you played where PCs can go off and do a whole bunch of shit by themselves without notifying the GM?

If a combat scene or action scene is happening, that means you may not be able to decide your move in advance, because you may need to wait for your turn to come around so you can ask what DC your MC would set for some jump, or swim, or climb you may be considering.

And again, because you don't know whether your jump would be impeded by an invisible wall, or your climb would be complicated by a grease spell on the cliffs, or whatever, you still need to check what the DC is even in games where those are set by the rulebook.

First, as in combat, it just saves time if you know what you can do without having to ask.

Except because you never know what hidden factors might be at work, you still need to ask in systems with rules-generated DCs anyway.

Swimming is a great example of an implied task. If there's a rapid river below my character, and I'm thinking of trying to cross the river on a narrow beam, I'm probably going to ask how difficult a balance roll it would be. I'm less likely to remember to ask how difficult the swim roll would be if I fell in. In that situation, while my focus is elsewhere, I'm likely to just assume that my character can or can't handle the rapids and that my referee probably feels the same way. If I assume "can" and the ref doesn't feel the same way, I could get myself into trouble. On the other hand, if the DC for an armored human to stay afloat in rough water is something I can just look up, I would have already check whether my character can survive a dunk in the pool, and if my character cannot survive taking a dip, I would know that going in an be sure not to put myself in that situation.

This seems to be a huge stretch. Why, exactly, is it more likely you will forget to ask "Just how swimmable just this river look anyway?" than it is for you to look up the "swimming in armour" rules in your rulebook?

For that matter, are you constantly looking shit up in the rulebook whenever you are called on to make a decision, including the DCs for failure conditions like falling off the beam into the river? I point-blank refuse to believe that this is faster than just checking with the GM.

Plus, yet again, you're assuming that you have a perfect knowledge of the situation which by definition the players in the typical GM-plus-players tabletop RPG format don't have. How do you know that the beam isn't polished with nigh-invisible grease? How do you know the river isn't an illusion?
Orion at 15:28 on 2015-03-18
For that matter, are you constantly looking shit up in the rulebook whenever you are called on to make a decision, including the DCs for failure conditions like falling off the beam into the river? I point-blank refuse to believe that this is faster than just checking with the GM.

It took me 2 minutes and 40 seconds to look up the following: Swimming in rough water is DC 15, and with the -7 armor check penalty for masterwork full plate, I would need a +12 swim bonus to automatically succeed by "taking ten." I'f I'm being shot at, I can't "take ten," but I won't actually sink unless I fail by 5, so if I have a +18 swim bonus I'm in no danger. Climbing a tree is DC 15, while climbing a generic dungeon wall is DC 25. Walking a 2-inch beam is DC 20, unless it's coated in invisible grease, in which case it's DC 25.

Suppose that the group wants to take out a fortified camp of enemy goblins. While the referee is describing the front walls, the gatehouse, and the visible defenders, it occurs to me that it might be a good idea to ignore the front gate and make a sneak attack by crossing the river and scaling the back walls. I can check the books to see whether that's something I would be able to do while another player is asking for more information on the the front of the fort, or negotiating with an NPC about acquiring equipment. If I decide it's a good idea, I'll still have to clear it with the referee, but if it looks impossible, I don't waste their time. If it does look doable, I can open talks with the referee like this: "Hey, the players handbook suggests that swimming rough water is DC 15, and climbing a dungeon wall is usually DC 20. Does "rough water" sound like a fair description of the conditions? Does my character have reason to believe the walls might be more difficult than that?"

How do you know that the beam isn't polished with nigh-invisible grease? How do you know the river isn't an illusion? I would say that this is a good time to remember that There is no rule-based solution to playing with someone that isn't fun to play with. You just chalk it up to experience and stop playing with them
Arthur B at 18:34 on 2015-03-18
It took me 2 minutes and 40 seconds to look up the following: Swimming in rough water is DC 15, and with the -7 armor check penalty for masterwork full plate, I would need a +12 swim bonus to automatically succeed by "taking ten." I'f I'm being shot at, I can't "take ten," but I won't actually sink unless I fail by 5, so if I have a +18 swim bonus I'm in no danger. Climbing a tree is DC 15, while climbing a generic dungeon wall is DC 25. Walking a 2-inch beam is DC 20, unless it's coated in invisible grease, in which case it's DC 25.

"What's the difficulty on swimming that river?"

"In your full plate? 22, and you won't be able to take ten because the goblins are firing at you".

This is an exchange which shouldn't take more than 10 seconds, and takes about as much time as saying the following:

"Based off the book, it looks like the DC for swimming that river should be 22 - I'm assuming I can't take 10 because of the goblins, right?"

"Right, and that DC sounds correct to me".

Don't get me wrong, in some types of game extensively nailed-down DCs can be an advantage, but I really don't buy that efficiency is one of those advantages. There's just as much scope to get into boringly drawn-out argument over hard and fast rules as there is over matters of GM fiat.

How do you know that the beam isn't polished with nigh-invisible grease? How do you know the river isn't an illusion? I would say that this is a good time to remember that There is no rule-based solution to playing with someone that isn't fun to play with. You just chalk it up to experience and stop playing with them

OK, but setting my arbitrary examples aside you'd agree that there are numerous reasons why a hidden variable might be at work in a game situation which you are not aware of but the GM is which doesn't necessarily equate to the GM being an arse, right? And even if it doesn't happen very much, the fact that you personally never know whether it's the case means that even in a fixed-DC system, you always have to check your DC figuring with the referee anyway - at which point you're spending just as much time talking to the referee as you would in a fiat-DC system.
In order to post comments, you need to log in to Ferretbrain or authenticate with OpenID. Don't have an account? See the About Us page for more details.

Show / Hide Comments -- More in March 2007