Deflection vs Failure

[Luke]

Personally, I love failure in rpgs. Failure breeds drama, in my opinion. And it is the act of not failing that makes heroes and tells stories. Success has little to do with history, i've found it's all about who didn't die and who didn't fail.

So one of the game-things I always wanted ask other designer/GMs was what they do with failure in their games. I got a range of opinions from the Forgers: from Vincent's complications, to (I can't remember what Ralph and Jake called it) "altering the intended course of events," to Ron's "don't make a clown out of the secret agent."

The example given for the "altering the course of events" was: An assassin fires and misses with a crucial bow shot on his intended target. What happens? Rather than just saying, "he misses", Jake and Ralph proposed introducing a heretofore unseen element to the story that explains the expert marksman's temporarily faulty aim. The example they gave was: a body guard spots the assassin and raises his crossbow (even though this test had nothing to do with the assassin's ability to move undetected). The assassin is forced to eliminate this threat rather than take out his intended target.

Interesting, but somewhat unsatisfying to me as a GM. It seems to give too much romantic clout to the character when he can never fuck up.

But last night something odd happened in our game. We had a circumstance where some highly competent characters were hobnobbing. A Persuasion test was made and the player failed horribly (Persuasion B7 + 2 FoRKs, she got 1 success). Everyone around the table laughed and jeered (players are so supportive!), but I stopped them and told them what happened: Rather than being laughed out of court and shunned forever, our courtier found herself unable to control the conversation and her "opponent" extricated herself with a meaningful "we'll talk more soon."

This quieted the table down and moved the game along. Though the "failed" player was still a bit incredulous-- she actually asked me, "So she didn't laugh at me because I have boogers hanging out of my nose?" "Nope," I said and told her that the meeting actually went well, she just didn't get a chance to deliver her "message".  The player immediately began to plot of her next move. The failure was water under the bridge.

There actually were a number of instances of this over the course of the night. Competent characters plagued by players' bad rolls. Was it disaster? No, I deflected their intention and explained that they hadn't necessarily failed, but that things had gone differently.  It worked very well. Each instance was accepted without resistance or even funny looks.

"Crucial" failures still loom heavy in my mind-- situations that are do or die-- but setbacks, deflections and inconvenience seem generally better suited to table-based story telling.

hmmm...

-L

[Jake Norwood]

I think the threat of failure is an important issue...in many styles of play it adds to the tension of the roll. The trick, I suppose, would be coming up with an organized way to go back-and-forth from the two kinds of failure.

Jake

[Valamir]

Luke, I think thats a perfect example of what we were talking about.  You have a character who's supposed to be socially smooth and competant in ettiquette who failed.  In the end she still failed in her objective to deliver the message, but she maintained her aura of being socially smooth and competant.  The random roll wasn't allowed to make her look stupid...she just didn't get what she wanted.

I think that's the key to this "deflection" as you call it here.

Consider WHAT the character wants (the goal) and the HOW the character is attempting to obtain the goal as two seperate entities to be manipulated.

In this case failure indicated that the character didn't get the "what"...she failed.  But the failure didn't effect the "how" she still performed competantly.

Its also possible to go the other way and give the character the "what" but have them screw up the "how".  This would be the picking the lock example we talked about where the character needs to get through the door for the story to continue and a series of failed attempts to open the door is considered tedious and boring.  In this case the failed roll may mean the character actually *succeeded* in the "what"...the door opened.  But failed in the "how"....broke the lock pick, set off the alarm, opened the door with a loud creak that someone certainly heard, took so long that guard is just rounding the corner and another roll is required to get through the door before being spotted...what ever.

Elfs makes intentional mechanical use of this second method in the core low cunning rules.  Elfs are constantly screwing up the how, but still manage to achieve the what, using the effect for its comdedic value...a tried and true method used in comedy since the beginning (the fortuitous trip that disarms the bad guys).  But the method doesn't require putting a comedic spin on it.


Failure doesn't HAVE to mean failure.  It can instead mean success with additional complication.  Instead of "No" it can mean "yes, but".

As a skilled GM you have a clear idea of when a straight failure is dramatically interesting and when its just...not.  Instead of fudging the roll, the "yes, but" method can preserve the failure, preserve the statistical sanctity of the dice but instead of screwing things up can be a spring board to further drama.

[Ron Edwards]

Hello,

My thoughts on the matter are presented in the Whiff factor thread. The initial poster made the same error you are making, Luke - perceiving that alternate, non-whiff techniques means characters "never fail."

Best,
Ron

[Paul Czege]

Actually, I think the first discussion on The Forge to cover the topic is this one: http://indie-rpgs.com/viewtopic.php?t=1163">Fortune-in-the-Middle.

Paul

[Gordon C. Landis]

The way I've been trying to think about these situations of late is "different - but not necessarily directly opposed - goals."  That is, the player has one goal in mind for the character's actions in the scene/situation, and . . . another goal (or goals) exist(s).  Sometimes, that goal is just flat-out opposition to the player, and that would usually lead to things like the embarrasing moment the player in Luke's example seemed to be expecting.

But a conflict that justifies activating some sort of resolution mechanic (die roll, whatever) need not only be because of two directly, mutually antagonistic goals - all you need is two (or more) possibilities that interfere with each other to SOME degree.  

The interesting question (to me, from a game design standpoint) is at what point in the IIEE process, and to what degree of clarity, is it best to directly and/or publically identify the various goals.  I take it from Luke's description that he didn't have a particular alternate agenda for the NPC in mind prior to the die roll - which I could see as a GOOD tool for avoiding railroading.  Or maybe it adds to the drama if the players know the "opposition" is up to something other than just opposing their character's actions.  Or . . . well, like I said, it's an interesting question I don't have firm answers for just yet.

Anyway - I was thinking in this area due to a recent playtest experience, and that's what I've thought through so far.

Gordon

[Clinton R. Nixon]

Good characters making screw-ups is, I've found, just about the most demoralizing thing that can happen to a player. Your technique sounds perfect, Luke - embrace it and go.

I suppose I don't have a lot to add besides "hell yeah," as this is one of my pet subjects. I will give an example, though:

In my Sorcerer post-apocalyptic game, one of the PCs was an expert archer. In the midst of our most action-packed fight of the game, the player wanted the character to fire an arrow into a fistfight, helping his friend by wounding the enemy. All the minor riff-raff enemies had been dispatched, and tension escalated around the fight.

Of course, the die roll failed. So, as a GM, what did I do to prevent this from going sour? A new enemy tribesman, one who had been skulking along unseen, leapt up to attack the fist-fighting PC in the back. The archer, poised a bit away, was narrated as quickly shifting his aim and mowing down the mook before he could make his blow. The main enemy wasn't harmed - the die roll had failed - but the character still looked like a bad-ass.

This is pulp storytelling. This is how to avoid deprotagonization in the face of bad die rolls. The key is remembering that looking good and achieving a goal are two separate things, and the die roll controls only the latter.

[Valamir]

Heh...actualy Clinton it was your example from that game that I related to Luke at Gen Con and which he referenced at the top of the thread ;-)

One of the most powerful game moment examples I've ever seen.  I steal it frequently :-)

[Luke]

I think the threat of failure is an important issue...in many styles of play it adds to the tension of the roll. The trick, I suppose, would be coming up with an organized way to go back-and-forth from the two kinds of failure.

Jake

(I hope you don't mind if I discuss this conundrum in relation to my game. It's really the only thing I play....)

Burning Wheel is, at its heart, a simulation. "What would it be like if we were there" it asks often, and then follows quickly with, "What would YOU do?"

Thus far this discussion (starting at GenCon) has made me rethink what BW is all about. Is it really a "storytelling game" like I so desperately claim? No, it's not. It's a vehicle for telling stories whose outcome we do not know.

In a piece of narrative fiction we generally can rest assured, that after the twists and turns of the plot, we are going to be provided with some modicum of resolution. Generally, in protagonist-based fiction, this is resolution is going to be fairly satisfying. The hero triumphs despite sacrifices. Akira Kurosawa's Seven Samurai is a fine example of this.

(Of course, success in that narrative has little benefit for the protagonists, but that's another thread).

The conclusion of the narrative is a convention of the medium of film- and literary-based fiction.

But what happens if we turn this into an rpg?
What happens if the players are now the "samurai"?

Suddenly, that foregone conclusion of protagonist-favored closure has got to be stripped away. Why? Because first and foremost, now we are playing a game. And though there is no winning or losing in rpgs, there is most definitely success and failure.

Secondly, the scent of assured success has to be stripped out in order to create tension among the players. They are asking themselves, "Can we actually hold the bandits off like Kanbei and Kyuzo?" And they are starting to sweat, because there is a chance they might not do it.

Lastly, imminent success (from narrative fiction) must be stripped off, because in the story and roleplay, the characters do not know what will happen. And roleplaying is the other side of our coin.

So, rpgs are different than novels and films. No big revelation there.
Ron brought up the HW wars mechanics favoring the outcome of Aragorn's trip on the stairs at Helm's Deep. Fair enough. And sounds similar to BW's Call-On traits and Saving Grace Artha.

But in that moment of running up those stairs, the player who controls that character MUST feel like they are walking the knife edge -- that their "life" is in danger. It is both necessary and appropriate to the moment being played out at the table.

So this brings me to "Well, what happened at the table?"

Gordon mentioned it: Where in the IIEE are we?

I think it is a games' use of the IIEE that really determines what failure can mean in the course of roleplay.

During the overwhelming melee, did Aragorn's player say, "I retreat?" Had he been repeatedly warned by the GM that his situation was worsening? Thus a desperate Speed test is made to reach the door before he is overcome? (In BW it would be Speed vs Speed). He fails to retreat. What happens? What needs to happen so the laws of the GM's universe are not broken?

Or was it one sweeping "attack" that Aragorn failed, as Ron suggested? Aragorn's player defends the stairs with an attack action and fails the roll. Thus he successfully defends the stairs, but is nearly overcome while falling back.

Where does the contract of engaging in deadly melee/play get acknowledged? Aragorn quite knowingly was risking his life for what he did. When is it time to pay that price?

In a simulation game, we are playing out blow-by-blow, and, i think, we acknowledge that no matter how cool you THINK you are, there is always a chance you fall flat. Which is a pretty fair simulation of life. As far as I can tell from all my research, life and history is far more replete with failure than success. Failure that often forces people to choose and other path or failure that just says, "you're wrong." Thus, I feel it is important not to over-romanticize a player's role in a story. There is always a chance they'll fail.

But wait, being "wrong"... wouldn't that be determined in the Intent phase? A wrongheaded action is undertaken, can it be successful at all? ::shakes head::

I digress.

But i am also aware that failure in life often brings knowledge, too. Or success in another, unexpected, way. Hence, "Eureka."

In this vein, BW rewards failure. 99% of abilities can be advanced on failure alone, but you've got to try.

But on the conflict resolution results side there seems to be some unsatisfactory options: dead-end failure (the door is still locked and there is no other route), crucial failure (death), and narrated "yes, but".

Agreed, dead-end failures are the worse form of railroading. If you are going to give the players only one single option, then you might as well just make them 100% successful.

Narrated, "Yes, but" doesn't sit well with my simulationist self. Sometimes we just lose the race, or fuck up our project. And we usually do this when we think we are at our best.

And crucial failure seems to be the only extant tangible effect in most rpgs, especially simulationist. Without some supporting superstructure that plays with a character's success, goals and personality, death is the only real threat an rpg character lives under. Everything else is just gravy. That's not cool, in my opinion. Bankers should be allowed crucial failure. They should be allowed to fuck up so badly they bankrupt themselves. Just like a swordsman risks his life in every melee...

As it stands now, I actually tell my players if a test is do or die. They dutifully and gleefully pony up artha and traits to ensure victory. But, at the same time, I NEVER only have one option for the players/characters. At the very least, they can always go back. More often than not, there is a another way around.

And for other, non-crucial failures, I usually have the test just waste time, energy and money. Players and characters are free to try again.

Yet here I am, rambling away, still vaguely dissatisfied with the whole lot of it.
-L

[Mike Holmes]

First, Luke, first, you missed Ron's point above. Nobody is saying that one should use such techniques all the time. Failure, out and out, messing up, didn't get it done, failure, should occur some of the time. You have to understand that first. Using these sorts of techniques doesn't work if you use them every time.

This solves the "tension" issue. Jake proposes a randomizing method. OK, roll a d6, even, you just failed, odd, use another of these techniques. Problem solved. The player still has to worry that he'll mess up. But it'll only happen half as often, and the rest of the failures will make up for that, by actually being "protagonizing" moments.

But you don't even have to do that. It can just be GM fiat. In telling a good story, just use the methods when they make sense to you to use them, and hit the players with failure when it makes more sense to do that. You'll know when, you're already talking about it in the posts above.

Point the second, you're mixing up Sim and Gamism, and a bunch of other things. I won't get into that too deeply, but the most important thing is that you've identified a need to stay out of the metagame in terms of describing events. That is, you don't seem to see the character abilities as player empowering devices, but as actual ratings of a construct in-game. And that's fine. It does mean that some of the sorts of techniques are unavailable. That said...

Point C, there are ways to do this sort of stuff that don't require going to the metagame. I keep explaining this, and people somehow forget that the idea exists. Bascally, the fact that a character has to fail doesn't mean that they have to look bad doing so. In fact, I think that descriptions of these sorts are an artifact of a certan style of play where the GM is out to hose the players. It's easiest to describe with an example (I wish that I could find the damn thread where I give the examples):

Bad Way GM: You failed your persuade? Boogers!

Good Way GM: You failed your persuade roll? You discuss the issue bringing to bear your impressive skills of persuasion, but your target turns out to be on the ball today, and says that they want to think it over.

I mean, where in the rules does it say that a bad roll has to equal something insulting to the character? Not to say that you can't use that occasionally in the right circumstances. But how is it the standard? Let's look at the result of a big failure (lots of success margin by the opponent, or a rolled "Crit Fail" or whathaveyou).

Good Way GM: You failed your persuade roll? You discuss the issue bringing to bear your impressive skills of persuasion, but the character notes that your signet ring gives you away as a member of the opposing guild. He rises up suddenly and calls for the guard.

The dice represent the luck involved in the situation. Part of the target's "defense" in this case is their ability to percieve things correctly. So it makes sense that the die roll in this case means not that the character did a bad job, but that the opponent did an exceptionally good job. Or fate intervened in some typical way to mess things up. The dice are randomizers that represent the minutae that we don't want to make up in play, nor do we pay attention to in RL, that are the actual determinants of what happens in a "random" situation. Use that.

In the classic "dead-end pick the lock" situation, you have several options. Basically, you have to look at the attempt in slightly larger than "task" terms. It's a conflict. In this case a door stands between the character and what he needs. Failure doesn't mean that the player didn't get through the door, necessarliy, and shouldn't if the game will halt were that to be the case. Instead, it means that something didn't go right with the attempt. What are your options?

A) The metagame idea would be to say that somebody comes along. This would be pure luck, however, and does not pertain to the skill at hand at all. So we want to avoid that.

B) You can say that the character failed to pick the lock, but that there was a window open nearby. No good. Again, somewhat metagamey, and worse, it makes the character's efforts pointless. Why even bother rolling if you know that the GM will just provide a way out of trouble in the end?

C) You can say that the character made a lot of noise, and this means that a squad of ninja are on the other side of the door when it opens. Sure this goes with the "new conflict" rule, which is cool, and it relates to the skill used. But the player will feel the metagame there again (unless you're playing hardcore illusionist). So this is not a great option either.

D) You can say that the character made a lot of noise, and this means that the guard at the last post comes looking for you. Now this is a good solution. It means that the character's skill is directly involved, and the negative result is something that makes sense within the established facts of the world. You set up some tension when they slipped past the guard in the first place, now you are escallating that to discovery.


Think of it this way. When a player says that they're trying to go through the door, they're making a lot of implicit statements. Select the one that makes sense of the situation, and have that be the subject of the failure.

1) I want to get through the door undetected.
2) I want to get through the door unharmed by the trap I detected in the lock.
3) I want to get through the door quickly because time is of the essence.
4) I want to get through the door without leaving evidence of tampering.
5) I want to get through the door without the fragile lock breaking, making it impossible to lock behind us.

All of these allow the Character to actually get through the door (thus portraying their ability), and yet to fail in a way that means further conflits. I'm sure more will come to you on the spot.

And if you want to let them fail, there's always Illusionist techniques (or No Myth play) that can be used to fix the situation as well. There are no "dead-ends" in Illusionism.

Mike

[Matt Wilson]

Hey Luke, you'll also see something like this in the copy of PTA I sent you, on page 31.

A lack of advantages should never make a character look inept, unless it has been established that the character is in fact inept at the specific task being attempted.
EXAMPLE: John is playing Gustav, who is trying to intimidate a bartender, and John rolls no advantages. Matt, playing the bartender, rolls and gets an advantage. There needs to be a good reason why a hardened mercenary would have trouble scaring the pants off a bartender. Maybe Gustav didn't pay his bar tab last time around, or the bartender knows something about him. Or maybe it turns out that the bartender is a retired mercenary.

[Gordon C. Landis]

Well, just one little twist I want to add here, in reponse to:

As it stands now, I actually tell my players if a test is do or die. They dutifully and gleefully pony up artha and traits to ensure victory.

Sure, that can work just fine.  But there's an alternative approach that (in my experience) doesn't lose anything in terms of the "I'm at risk here" feel, but that works better for some play styles.  I think it was 7th Sea that taught me this, or maybe just discussions on the Forge . . . Anyways, let's flip those two sentences I quoted around. How we know when a test is do or die is determined by when the player ponies up a ton of artha and traits (or whatever).  They indicate to the GM and the other players when they consider the in-game situation appropriate for having the continued existence of their characters on the line.

Not better, just different - and it doesn't work if you seriously want a "every combat really can be deadly" feel.

Gordon