Tension, Dice-rolling, and Reliance on Randomness

RPGBloggers seems to be down right now, which is causing me a problem with looking up the source/inspiration for today’s post. I promise I’ll get a link in here as soon as I can find it again…

The simple upshot – the writer was pondering about the issue of assured success in gaming. That is, if player characters are certain they are going to succeed, then what’s the point? Or even more, as the writer put it (and I’m paraphrasing here) – Is it just as special to succeed on that last desperate roll when you needed an 18 (on a d20) as it is when you know you’ll succeed on a 2+?

He further discussed the issue of re-rolls and how a character (or party) can be so good at something that they will never really be challenged. All of these are very important questions in my mind because I want to do something that is probably the worst offender you can imagine in every aspect of these designs – I want to create a diceless game.

In a game that does not use a randomizer, no dice or cards or anything of the sort, it is easy to drop to a binary mechanic of “you succeed” or “you fail.” I remember my dissatisfaction with the Cthulhu Live book when I started reading it and realized that your character was built with static numbers that determined everything. “Oh, you have an EDU of 12? Well, this needs a 15, so you fail.” It was a hard and fast type of binary – either you were good enough, or you weren’t. That didn’t work for me…

But I was thinking about it more after reading this post. Isn’t rolling the dice in danger of the same type of binary? I know the thought is deeper than that, and that the act of rolling of the dice to generate success or failure builds tension (in and of itself) but so what? You still either succeed or you fail. If you need to hit a DC 20 (to use D&D terms) and you only get an 18, you failed the roll. If you need to hit an AC of 23 and you roll a 22 – you miss. If you roll a 23 – you hit. That binary succeed/fail thing is still around and what the dice are adding to the scene is a sort of artificial tension… that is to say, it gives the player a reason to sweat just a little at the same time their character is sweating it out, thinking about what happens if they miss…

And that tension is interesting to me – and the scenario used by the author of that other post gave me a perfect window into the problem. He was talking about a baseball pitcher, trying to strike out a batter and the worry about that potential hit. The tension in that moment though, probably doesn’t come from the pitcher’s skill. Assuming we are talking about professionals here, that pitcher is, presumably, good at what he does. And presumably, as well, the batter is also good at what he does. So the comparative skill of each of them will be called upon for their own actions – but the overall difference in their skill (in gaming terms) probably comes down to, at most, a few points. Ultimately, the tension is built by the value of the pitch/hit in question. If the pitcher is throwing this pitch in the third inning of the third game of the season, it means less than if the pitcher is throwing for the last strike, for the last out, in the last game of a seven-game-long World Series…

And that tension is important. One of the question raised was, “would you even have the players roll for this?” And that’s where I start to get back to my diceless thinking. The majority of what we do in games doesn’t really need a roll. And sometimes, I think we’ve resorted to rolling for things we really don’t need to. I mean, I had this feeling a week or so ago when I was running my Pathfinder game – I really hate calling for perception checks to find stuff. The Troll Chieftain has a key hidden in the skull on his bracelet? That key is the only way to open the treasure chest (without picking the lock – as our party has no rogue)? The adventure says that it’s a DC20 Perception check to find the key. But if my players look at the chest and say, “hey, that key’s here somewhere” and they thoroughly search the freshly-minted-troll-corpse-they’ve-just-created-by-being-awesome, I’m going to let them find that key. After all, why not let them find the key? Once the troll chieftain is dead and the tension is gone, then why fret over it? (Yes, I know, in Pathfinder this is already covered by taking 10/20, but I was making a point which goes beyond just Pathfinder)

Ultimately, as I keep working on my diceless system and thinking back on my diceless play experiences with Amber (and what made those experiences so great) I keep considering what my role as designer is in setting up player expectation. I’m thinking about success and failure and how to achieve more than a simple binary effect and one thing I keep coming back to is the idea that it always involved two things: the first was give and take with the GM – working out that yes, I was probably going to succeed or fail, but I had stages of gaining that understanding (and that seems to have verisimilitude for me) and two – the tension of the scene mattered, but not the tension of, will I roll a 15 when I need a 16? It was the tension of the moment, the tension of making sure that if I – as the GM – was pushing my players, it was because the stakes were high (and if I was a player, were the stakes important enough to build my investment when our GM pushed us).

And in this, I guess I shy away from some of the “old school” things. I never liked Wandering Monster encounters, for example. I wanted encounters to mean something when I had them. Better yet, as a player, I often hoped that encounters I found myself in were the result of my own decisions… because it was the tension that made them matter – not the roll of the dice.

Comments? Questions? Thoughts? Anecdotes?

As always, thanks for reading.

Advertisements

15 responses

  1. It may just be changes in what we look from in a game – but the more I know that the results of my actions are up to the GM’s fiat the less agency I feel I have in the game, and the less I care.

    It is why I don’t think I could play Amber again. With no real independent way to affect the course of the game, I’m basically doing collaborative storyteller with a judge – and that requires more trust than I have for finding someone who’s values, preferences, and goals are perfectly in sync with mine – I may be sacrificing the ability to have “great’ to have “good” but it is a price I’m willing to pay.

    1. That’s interesting to me — I certainly understand the feeling of GM fiat that comes from not having dice or rolls or such — but the conclusion I’d sorta reached was that I was stuck with that anyway — whether or not I had dice — as long as I had a GM.

      Meaning, I’m trusting the GM to set the DCs, to design the encounters, to create appropriate challenges, to not abuse their freedom — and the dice (or lack thereof) was a secondary concern in the equation.

      I feel — and I realize that I may only be speaking for myself here — that I have more agency when involved in a game without random elements — because then outcomes are solely based on my RP and my decision-making and my communication with the GM and other players. But I certainly agree that this requires a group effort and a group commitment.

  2. I went through a diceless phase. In the end, most of my players preferred to roll them than not. I believe group dynamics shift constantly. I’ve watched players roll pointless dice even when I wasn’t running the game and the GM be completely annoyed with them. Still, those people wanted randomness to have an input.

    Others want to just play into the story. To make it happen without the chaos of dice. Some individuals can pull that off. Others cannot.

    Blending the two is difficult. I’ve enjoyed both styles.

    1. I guess I’m still in my diceless phase. It’s funny, but I long ago realized that the lack of dice is what made Diplomacy my favorite wargame. I think that a diceless game still requires solid, reliable mechanics — that it’s not just about “playing the story.” That’s what I’m trying to work toward with my current project.

      I wish I knew where the post was (I’m really bad at this) but someone wrote, not so long ago, about how they mixed the two styles — and it was an awesome post… If I can ever dig it up, I’ll link it here.

  3. I agree that a game should have varying levels of success. I use it in my system and it makes for a more rich interaction each time you have to resolve an action.

    The concept could translate to a diceless system but since I’m using a randomizer it’s easier.

    Here’s a random thought for a diceless randomizer. 🙂 Each player (or GM) thinks of a word to describe how their character will succeed at the task. The word has to make sense and describe what is going on or the character automatically fails. Both words are compared. The first four letters of the words are checked to see which come first alphabetically each one that comes first gives a plus to that player. The longer word also gets a point. For example P1 “dashing” and P2 “energetic” P1 gets three points for the D A and the H. P2 gets two points for the second E and for the longer word. P1 wins, possibly having that modified by a skill or attribute.

    It was a spur of the moment thought, maybe it will produce something useful, maybe not.

    1. Your random thought is pretty awesome for an off the cuff idea. I like the basic thing you’re doing there. I guess for my own taste though — I’m trying to avoid making too many “sub-systems” or “mini-games” out of the game. In this sense, I do want play to remain seamless from story — that is, I disagree with the new school sentiment that mechanics should reflect the game and be obvious — one of the things that pulls me toward diceless is that I like my mechanics to become transparent.

      And now I feel like I’m not making much sense… so I’m going to quit there.

      As for the idea of graded success — I suppose for me (again, just my thoughts) I’m not looking so much for a “I succeeded, now by how much” so much as I want more give and take leading up to an eventual “success” of whatever sort the challenge was about…

      And maybe I’m still not making sense?

      Thank you for the insight though. All of you. I appreciate the comments.

  4. Thanks to the Artifact.net (http://www.theartifact.net/) also being inspired to write by the same post I was — I found the original piece that inspired my writing over at Reality Refracted (http://www.realityrefracted.com/2011/07/mitigating-lows-means-mitigating-highs.html) and I appreciate the finding of these — since I apparently couldn’t on my own…

  5. Managing dice rolls and success/failure has been an extremely active topic for a few years now. It’s generated a few really useful maxims:

    Only roll the dice when it matters. If the story will turn out basically the same whichever outcome happens, just run with success for the player characters.

    If you’re going to roll the dice, be prepared with both a success and a failure. Too many GMs have gotten trapped by calling for a perception-style check to notice a vital clue, then having to backpedal when no one in the party succeeds in finding the plot. If you can’t afford to have them fail, you can’t afford to have them roll.

    Failing the check doesn’t necessarily mean failing. To use your example of the Troll King and the key, don’t have a failed perception check mean that the characters simply can’t find the key. Instead, maybe it means that the Troll King fell with that arm underneath him. Getting to the treasure now involves moving the large, smelly corpse and cleaning the gore off the bracelet. Have failed checks introduce complications, not blocks.

    Similarly, you can use the dice to inform the quality of success, rather than simply the presence of success. If it is worthwhile to assume that the character succeeds (e.g., leaping a chasm where failure means death), the check is really used to determine how well or poorly the attempt goes (e.g., a “failure” might result in hitting the edge of the chasm and needing a lift up, while a “critical success” might imply that the character nimbly and lightly crossed the void).

    1. Lugh, thanks for the comment. I don’t think I’ve seen you around here before — so, welcome.

      I agree that dice rolling is only really necessary when success actually matters, and I’ve been in enough games where rolling replaced playing — but that’s not really what’s driving me…

      My problem with examples using the “failed checks introduce complications not blocks” idea is precisely that — players should/could be narrating their search of the body (to use the troll again) rather than rolling first — and to use your example (which I don’t mean to pick on — it’s just a really good example and stays with the troll) there’s really no way to use this without it being clumsy… either the DM has to effectively tell players “well, you don’t find anything but…” and convince them to flip over the body OR many players will assume that if they say “I’m searching the body” and that they roll and the DM says, “well, you don’t find anything” that they will indeed have already done things like flip over the body… A DM who micromanaged my character’s actions to the extent this example would require would upset me as a player, not encourage me. And I can only speak to my own experiences — but I’ve experienced this type of play with many groups.

      To take it further — if a DM were to say something like — “Well, you don’t find anything, but you noticed he was wearing a huge bracelet during combat and that’s currently trapped under the body because of how he fell…” then effectively, the DM just gave a failed roll a “do-over” and guided the party overtly at the same time. If you are going to do that — you might as well just give them the key to begin with, right?

      I dunno — I really don’t mean to pick on this one example — I just find the current thinking about rolling more discouraging than encouraging.

  6. Well, to elaborate on the example, before the player even rolls the GM has decided that the character will find the key. That is a given. The point of the roll is to determine the, for lack of a better term, cost of finding the key.

    If the roll is a critical success, the troll happened to break the bracelet in the fall, and the key fell out on the floor in plain sight. (Negative cost, as the search check actually takes less time than it normally would.)

    If the roll is a regular success, the troll fell with his arm either on top of the body or conveniently outstretched, so that access to the bracelet is easy. (Normal cost.)

    If the roll is a failure, the character figures out that the key is in the bracelet, but getting to the bracelet is going to require some effort, and is going to be gross. (In addition to the time spent searching, the player also has a RP opportunity to be too disgusted or too weak to take advantage of the success and needs to ask a party member for help. If you want to stick with mechanics, you might require a second strength or willpower check to dig out the arm, but I wouldn’t do that.)

    If the roll is a critical failure, the character actually finds a different key tucked in the troll’s belt, and wastes ten minutes trying to figure out why it doesn’t fit the lock. After that ten minutes, the character finds the real key. (Additional cost of extra time spent on the check.)

    The point is to stop looking at the roll as giving a simple yes/no answer, and instead giving an idea of how complicated/costly/awkward an earned success would be. Really high rolls mean that the character not only succeeded, but looked good doing it. Really low rolls mean that the character had to spend extra time, or got their clothes dirty, or maybe even pulled a muscle. But, the basic task still got done.

  7. I think — for my own play sensibilities — that I have a problem starting right from the first assumption… that the GM has already decided the PCs get the key.

    Now, that’s fine, if that’s what you want, but if you are already planning on them getting the key (which in this case I was) then why not just give it to them? There doesn’t need to be a “cost” associated with it if the assumption is just to give it to them anyway. And the scenarios you outline for the degrees of success/failure presume on the part of the PCs. Which I prefer to avoid.

    Look at it this way — Why not simply put the key on a chain around the chief’s neck and then have a PC see it during combat? Then when the slimy, bloody, burned, acid-scarred body falls over, and the PCs are left to get the key — then they actually have to make the decisions and have the discussions you bring up. If they’ve already made the search checks though — then I, as a player, would be annoyed when the DM started layering stuff in… and making assumptions about the way we searched.

    I would submit, alternatively, that the point is not to use the roll unless absolutely necessary and instead to teach players to play actively by offering them good story reasons to do so. Heck, if you want to really use the skills — then have the PCs make Perception checks right after combat (or during, if you prefer) and when the fight is over you gently remind them about the key they saw if necessary.

    Many systems don’t have more than a binary yes/no built into the skill system anyway and those that do often have an artificial set of designated levels — I’d prefer, speaking for myself — to build that into the PCs decision-making and describing what they do proactively instead of reactively as post-skill roll solution does.

  8. OK, let’s break down that assumption that they get the key first. On the one hand, you can simply decide as GM that they get the key, and skip the roll altogether. It is important for whatever reason that the group gets the key, so they do. IME, this tends to strongly devalue perception (and, in other situations, knowledge) checks. There is never any reason to put points into perception, because you will always succeed or fail as the plot demands. This is doubly true if you turn a search check into an exercise of the players describing where they search, and giving them success if they choose to search in the right place. Or, alternatively, if you simply assume that the PCs are taking appropriate search actions, which you’ve already said annoys you.

    On the other, you can keep the success/fail mechanic. If the player rolls well enough (possibly modified by declared actions), then the group gets the treasure. Otherwise, they don’t (or, at least, have to think outside the box to get into the box). Now you are removing some of the power of the GM to dictate the story, but you are putting that power into the dice, not into the players. In some situations, that’s great. In others, as previously noted, it can really screw up a storyline.

    My example allows the GM to dictate the events necessary to move the story along, but then allows the dice to flavor those events.

    As to presuming on the part of the PCs, I think that you can’t get around that. Asking them to detail their search procedures is generally going to just be tedious, unless they are dealing with working around a trap. You can run with certain assumptions based on the fact that a) they are generally intelligent people, and b) are probably professionals. That’s what the skill ranks mean. As long as your flavor isn’t harming the characters unduly, isn’t radically against their established concept (e.g., a historically paranoid thief deciding to just yank open a door), and moves the action along in an interesting way, I don’t see the problem.

    Also, I don’t think that my example really made much in the way of presumptions. The PCs stated that they were searching the body, or maybe even specifically searching the body for the key. That statement requires certain actions. The die result tells me that their actions, along with their previous experience, reveals certain information. In the case of either a critical success or a success, acting on that information (taking the key) has no ill effect, and going through the motions of checking the bracelet for traps has no benefit for timing or atmosphere, so I assume that they take the key. In the case of a failure, acting on the information *could* have an ill effect, so I tell the player what was discovered and allow him to decide. In the case of a critical failure, acting on the misinformation only has the ill effect of lost time, which is the intended cost of the poor roll. Again, assuming that you as GM know that time is not of the essence, there is little benefit from negotiating each individual step of the action.

    As a note, it doesn’t have to be the GM who decides what the side effects of the low die roll are. If you have players you trust, and who are comfortable taking narrative control, you can certainly let them tell you what failure means in this case. You say, “You know the key is hidden in the troll’s bracelet, now tell me what happened in finding it that caused you problems.” Maybe the player surprises you with something like, “Well, in the course of searching the body, my rogue had actually slipped that bracelet under her shirt to keep for herself. It’s only after ten minutes of fruitless searching that she thinks to pull it out and check it. Explaining to the rest of the party how the bracelet ended up inside her shirt takes another ten minutes and some quick thinking!”

  9. Lugh :

    OK, let’s break down that assumption that they get the key first. On the one hand, you can simply decide as GM that they get the key, and skip the roll altogether. It is important for whatever reason that the group gets the key, so they do. IME, this tends to strongly devalue perception (and, in other situations, knowledge) checks. There is never any reason to put points into perception, because you will always succeed or fail as the plot demands. This is doubly true if you turn a search check into an exercise of the players describing where they search, and giving them success if they choose to search in the right place. Or, alternatively, if you simply assume that the PCs are taking appropriate search actions, which you’ve already said annoys you.

    — I was reacting to what you said here — your point was an assumption (from your previous post) that the PCs will/should get the key. And I’m okay with devaluing skill rolls — in the interest of improving player interactions with the game. Narrative work by a player should supersede dice-rolling when possible, with dice-rolling as a mechanic to resolve the action in it’s finality — not the other way around (IMO).

    On the other, you can keep the success/fail mechanic. If the player rolls well enough (possibly modified by declared actions), then the group gets the treasure. Otherwise, they don’t (or, at least, have to think outside the box to get into the box). Now you are removing some of the power of the GM to dictate the story, but you are putting that power into the dice, not into the players. In some situations, that’s great. In others, as previously noted, it can really screw up a storyline.

    –My contention would be not to create the situation in the first place by not doing something that creates such a situation. And my goal is to get the power out of the hands of the dice but I’m not really sure how my suggestion takes away any ability from the GM or the players?

    My example allows the GM to dictate the events necessary to move the story along, but then allows the dice to flavor those events.

    –I trend the exact opposite direction. The PCs should act, then the dice resolve their mechanical actions — but if not necessary — don’t bother.

    As to presuming on the part of the PCs, I think that you can’t get around that. Asking them to detail their search procedures is generally going to just be tedious, unless they are dealing with working around a trap. You can run with certain assumptions based on the fact that a) they are generally intelligent people, and b) are probably professionals. That’s what the skill ranks mean. As long as your flavor isn’t harming the characters unduly, isn’t radically against their established concept (e.g., a historically paranoid thief deciding to just yank open a door), and moves the action along in an interesting way, I don’t see the problem.

    –Put it down to experience then. I don’t presume on the part of my players… ever. Even the most innocuous things have led to arguments in my many years on both sides of the screen, some I’ve been a part of and some I’ve just watched in awe and despair…

    Also, I don’t think that my example really made much in the way of presumptions. The PCs stated that they were searching the body, or maybe even specifically searching the body for the key. That statement requires certain actions. The die result tells me that their actions, along with their previous experience, reveals certain information. In the case of either a critical success or a success, acting on that information (taking the key) has no ill effect, and going through the motions of checking the bracelet for traps has no benefit for timing or atmosphere, so I assume that they take the key. In the case of a failure, acting on the information *could* have an ill effect, so I tell the player what was discovered and allow him to decide. In the case of a critical failure, acting on the misinformation only has the ill effect of lost time, which is the intended cost of the poor roll. Again, assuming that you as GM know that time is not of the essence, there is little benefit from negotiating each individual step of the action.

    –That was my point: That negotiating each little step of the process, even if it’s after the skill roll as you suggest, is a waste of time. Set the situation up to allow the PCs to make their choices prior to rolling (if rolling is even needed) and avoid the issue altogether.

    As a note, it doesn’t have to be the GM who decides what the side effects of the low die roll are. If you have players you trust, and who are comfortable taking narrative control, you can certainly let them tell you what failure means in this case. You say, “You know the key is hidden in the troll’s bracelet, now tell me what happened in finding it that caused you problems.” Maybe the player surprises you with something like, “Well, in the course of searching the body, my rogue had actually slipped that bracelet under her shirt to keep for herself. It’s only after ten minutes of fruitless searching that she thinks to pull it out and check it. Explaining to the rest of the party how the bracelet ended up inside her shirt takes another ten minutes and some quick thinking!”

    –And that’s just it, they don’t really need my permission to have narrative control. I don’t restrict that in my games much at all. I’m saying that by encouraging more narrative control you could effectively change the skill dynamic to not require as much rolling, and that rolling — when you do it — should not be narrated after the fact, but before.

  10. arg. So my response is messy. I hope you can follow it and I’ll know better than to do that again…

    Thanks for the conversation.

  11. So, basically you’re just saying that that option doesn’t do much for you, because of your particular style. Right? In that case, no problem. You’ve clearly gone much more for the first maxim: Only roll the dice when it matters.

    Personally, I can’t get into the “roll first, then describe the actions to fit the roll” mentality. It just doesn’t work for me. I much prefer to describe my attempt, roll the dice, and let the dice tell me if I succeeded or failed. That’s just my style. As such, I get a lot more mileage out of “a failed roll is not necessarily a failed action.”

    YMMV, of course.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: