The Two-Roll Rule: A rule for cases where everyone could make a skill roll

jeffh

Explorer
I noticed the thread "A case where the 'can try everything' dogma could be a problem" elsewhere on this forum, and it occurred to me that a rule I've been using for quite some time may solve the issue Li Shenron brings up there.

I decided to start a new thread so that this topic doesn't get buried in an already-huge thread, and because the original discussion was (nominally) specific to 5th Edition D&D but my suggestion isn't. Indeed, while I've been using this rule in various versions of D&D/Pathfinder for years, I never got around to formally writing it up until fairly recently, when I was doing up the skill system for my own Fantasy Infinity RPG. Fantasy Infinity is, mechanically speaking, nothing like D&D. Both the problem Li brings up and the solution I've been using apply to a wide variety of systems.

(You can learn more about Fantasy Infinity by visiting my site, and actually I'd appreciate any feedback people want to give me on other aspects of the game - but it's probably better if this thread stays focused on solutions to Li's issue.)

Anyhow, here's the relevant bit from my Core Rulebook. As you'll notice, there's very little reference to FI-specific rules, and it could be applied to any version of D&D with a skill system 99% as written. The only system-specific bit I can see, and it's not essential, is the reference to complications, an optional rule that lets the GM create a 13th Age-style "Fail Forward" scenario.

The Two-Roll Rule
Sometimes the whole party must complete a task (swim across an underground river, for example), though only some of them actually have the skills for it. Even a very competent party is unlikely to have everyone succeed at anything non-trivial. This sometimes makes for fun improvisation, but at other times it’s just a pain in the butt.

At other times, it might seem like everyone, independently, could at least attempt a task (like spotting a stealthy enemy). This has the opposite problem; allowing everyone to roll when only one person needs to succeed doesn’t work well, making many tasks far too easy.

In these two types of situations, the GM can have exactly two PCs (out of a group of three or more) make skill rolls. Typically, these should be the two best suited to the task at hand, although if someone has gone out of their way to declare that they’re doing something related to that task, that character should be involved in these rolls, regardless of their relevant skill or lack thereof. (In this case, the GM may decide to cut them a break in the form of bonus dice (see below) if what they announced was a particularly good idea.)

  • If both rolls succeed, the entire party succeeds, with the more skilled characters aiding or covering for the less skilled ones as needed.
  • If neither roll succeeds, no one in the party succeeds.
  • If only one roll succeeds, then that character succeeds but no-one else does. Alternatively, this might be a good time for a complication (see above), giving the character that succeeded some kind of advantage in dealing with whatever the GM comes up with.

When only one character succeeds, that character may be able to leverage her success into something that lets the others succeed too, or at least have an easier time. As a simple example, if only one character succeeds at climbing a cliff face, she can drop a rope behind her to give the others a second, easier shot at making the climb. (This is also a good example of a change in circumstances that makes redoing a failed roll possible.)

Not all situations are handled well by the two-roll rule. The more detailed rules in the specialty descriptions should be used when the details matter, like in combat, where how long it takes a particular character to succeed might matter. When the details don’t matter, use the more abstract rules in this section.

If you discover that it makes a significant difference to the party’s chance of success whether you use the two-roll rule or a more detailed procedure from later in this chapter, use whichever method gives you a better chance of success. (And let us know by responding to the latest blog post at philosoraptorgames.com, because that’s the sort of thing we would want to look at and possibly revise.)
+++++++++++

So, there it is. Does anyone see any issues with it, or have any questions, suggestions, or random derogatory comments about my lineage to contribute?
 
Does anyone see any issues with it, or have any questions, suggestions, or random derogatory comments about my lineage to contribute?
"Issues" in the sense of "not liking the kind of game it is designed for," or in the sense of "unintended consequences even in the kind of game it is designed for"? It's way too gamist for my taste, and the "fail forward/complication" thing especially would be a real turnoff for me, but you must already know about these issues and obviously you're fine with them in your game. I don't see any real unintended consequences--it's just a special case of a group check. For some scenarios, like your rope-climbing example, it actually makes more sense than a standard PHB group check (which is also too gamist for my taste). For other scenarios like stealth it might be slightly worse.

In other words, it seems fine. More power to you.
 

jeffh

Explorer
"Issues" in the sense of "not liking the kind of game it is designed for," or in the sense of "unintended consequences even in the kind of game it is designed for"?
The former is valid too, if nothing else it's often interesting to hear. But the latter is way more helpful for my immediate purposes.
It's way too gamist for my taste, and the "fail forward/complication" thing especially would be a real turnoff for me, but you must already know about these issues and obviously you're fine with them in your game.
I hope these two points aren't meant to be connected (if they are, then something is being miscommunicated here). I don't see fail-forward as gamist at all.

(I should perhaps add that I reject a lot of the assumptions of the Ron Edwards model, though I do find the older rgfa Threefold to be an occasionally-useful framework for thinking about certain topics. There are gamist decisions and maybe even gamist mechanics, but I don't think consciously setting out to design a gamist game will go anywhere useful.)

In any case, in Fantasy Infinity it's optional, and not necessarily the default even in games where it's used.
I don't see any real unintended consequences--it's just a special case of a group check. For some scenarios, like your rope-climbing example, it actually makes more sense than a standard PHB group check (which is also too gamist for my taste). For other scenarios like stealth it might be slightly worse.
Stealth is strictly individual in FI, though I can see that this passage in isolation doesn't make that clear.

I do think this avoids, at least, the problem Morrus pointed out with group checks, that the guy who's really good is often better off without the "help". That may, however, be a gamist consideration of a sort that isn't normally on your radar, if I'm understanding you correctly.

Thanks for the feedback!
 

Morrus

Well, that was fun
Staff member
Total facepalm. Not surprising. Other games are better. I think it needs rewriting.
 
I hope these two points aren't meant to be connected (if they are, then something is being miscommunicated here). I don't see fail-forward as gamist at all.
Not intended to be connected. Fail-forward seems to be slightly more popular among the narrativist-leaning DMs in my experience; I don't see it as gamist.

Either way, this rule isn't designed for simulationists, and that's the camp I'm in. (I'm not married to Ron Edwards' ideas but I happen to find his terminology useful for conveying the gist of my style--don't read too much into it though.)

Peace,
Max
 

jeffh

Explorer
Total facepalm. Not surprising. Other games are better. I think it needs rewriting.
I don't follow. I understand these individual sentences, but have no idea what any of them is referring to or what the reasoning behind them is. Could you be a bit more specific?
 

MarkB

Hero
It seems like such situations could often be handled better by using group checks.

For situations (like spotting an enemy) where the party succeeds so long as at least one person succeeds, use the assisted-check variant - the best qualified person makes the main roll, and others can roll to see if they grant a bonus (or a penalty).

For situations (like sneaking past a guard) where the whole party needs to succeed, use the combined-check variant - each person makes their own check, and if at least half the checks succeed, the task is accomplished.

Either way, everyone contributes and nobody feels left out, but success is not guaranteed.
 

Morrus

Well, that was fun
Staff member
I don't follow. I understand these individual sentences, but have no idea what any of them is referring to or what the reasoning behind them is. Could you be a bit more specific?
Apologies - it was a (very weak) joke meant for somebody else. Completely addressed to the wrong person! Please ignore me! :)
 

Saelorn

Adventurer
So, there it is. Does anyone see any issues with it, or have any questions, suggestions, or random derogatory comments about my lineage to contribute?
One of the problems in letting everyone roll is that it assumes the chances of each character are entirely independent of each other, when the die roll is really meant to encompass a number of variables outside of just your individual skill. The wizard shouldn't get a chance to hunt for food after the ranger fails, because the 2 on the ranger's skill check indicates a scarcity of game (among other things), and the circumstances which cause the ranger to fail would not allow the wizard to succeed. Those circumstances aren't going to change, just because someone else is making the attempt.

This method solves that problem. It recognizes that, if anyone succeeds, it's probably going to be one of the two who are most likely to succeed. If anyone can find enough food to feed the party, then it's probably going to be the ranger (or maybe the rogue/scout), but it probably won't be the wizard.

The downside of this method is that it prevents anyone but the most likely from ever succeeding. With these rules, the wizard will never succeed in finding food after the ranger and the rogue fail. So it's just a matter of whether anyone considers this to be a worse gameplay issue than letting everyone have an attempt, and virtually guaranteeing success on any check that only needs one success (unless the GM artificially inflates the difficulties). That comes down to personal preference, but I see it as a small price to pay.
 

Jhaelen

Visitor
With these rules, the wizard will never succeed in finding food after the ranger and the rogue fail. So it's just a matter of whether anyone considers this to be a worse gameplay issue than letting everyone have an attempt, and virtually guaranteeing success on any check that only needs one success (unless the GM artificially inflates the difficulties).
I think that's fine, especially if the Wizard doesn't have any training in a skill that might help finding food. Actually, I'd expect a wizard (at least in D&D) to know a spell that might help, though.

The question to me is:
How many checks are there that only need one success? Finding food seems to be an activity that allows for multiple successes to lead to better results, i.e. find _more_ food.

The thread that inspired this discussion was mainly about Knowledge skills. Knowledge skills, imho, also work best, if multiple successes can be used to improve the initial result, i.e. hand out information in small increments. This also simulates that if one pc remembers something and tells the others, it may remind them of something they initially didn't think of on their own.
 

Saelorn

Adventurer
The thread that inspired this discussion was mainly about Knowledge skills. Knowledge skills, imho, also work best, if multiple successes can be used to improve the initial result, i.e. hand out information in small increments. This also simulates that if one pc remembers something and tells the others, it may remind them of something they initially didn't think of on their own.
I guess it depends on which you prefer - that the ranger is likely to remember some detail about magic that the wizard has forgotten, or that the ranger should almost never know something about magic that the wizard doesn't know.

For my money, the second generalization seems way more reasonable. In my mind, the only time the ranger would ever know more about magic is if they grew up in a region with very specific magical traditions with which the wizard wasn't familiar (e.g. the ranger grew up in a land full of undead, but the wizard had only read about them in books). This is doubly important in a system utilizing Bounded Accuracy, which greatly increases the chance of any chump randomly making a difficult knowledge check after the expert fails (i.e. the die vastly outweighs the modifier, so the ranger can roll a 15 with no bonus and easily beat the wizard's score of 5 + significant bonuses).
 

DMMike

Game Masticator
The Two-Roll Rule
Sometimes the whole party must complete a task (swim across an underground river, for example), though only some of them actually have the skills for it. Even a very competent party is unlikely to have everyone succeed at anything non-trivial. This sometimes makes for fun improvisation, but at other times it’s just a pain in the butt.

At other times, it might seem like everyone, independently, could at least attempt a task (like spotting a stealthy enemy). This has the opposite problem; allowing everyone to roll when only one person needs to succeed doesn’t work well, making many tasks far too easy.
I think these premises need consideration.

When must the whole party complete a task? If only half of them can cross an underground river, well, either the successful ones lower the drawbridge from the other side, or the non-swimmers find another route. If this is just a way to address splitting the party, you will hopefully agree that splitting the party occasionally is not a bad thing.

When it comes to the second case, I typically just ask for one roll from the best candidate, or ask everyone to roll, and whoever rolls highest gets the worm. For what it's worth.

The wizard shouldn't get a chance to hunt for food after the ranger fails, because the 2 on the ranger's skill check indicates a scarcity of game (among other things), and the circumstances which cause the ranger to fail would not allow the wizard to succeed. Those circumstances aren't going to change, just because someone else is making the attempt.
+1. If your specialist rolls low, it's not because he's rolling low. It's because the wind is wrong, or the gods have cursed the party, or...
 

jeffh

Explorer
One of the problems in letting everyone roll is that it assumes the chances of each character are entirely independent of each other, when the die roll is really meant to encompass a number of variables outside of just your individual skill. The wizard shouldn't get a chance to hunt for food after the ranger fails, because the 2 on the ranger's skill check indicates a scarcity of game (among other things), and the circumstances which cause the ranger to fail would not allow the wizard to succeed. Those circumstances aren't going to change, just because someone else is making the attempt.
Thanks for spelling this out. At the time this post appeared I was thinking about a "by the way, you can justify a lot of this in a simulationist way" further follow-up to Hemlock, but then you went ahead and wrote it more articulately than I was likely to.
The downside of this method is that it prevents anyone but the most likely from ever succeeding. With these rules, the wizard will never succeed in finding food after the ranger and the rogue fail. So it's just a matter of whether anyone considers this to be a worse gameplay issue than letting everyone have an attempt, and virtually guaranteeing success on any check that only needs one success (unless the GM artificially inflates the difficulties). That comes down to personal preference, but I see it as a small price to pay.
It's also sometimes possible for the weaker (at the skill in question, not necessarily physically) characters to aid the stronger ones, as Mark also points out. He labels it as an alternative to my rule but there's no reason you couldn't use both together. (The current draft of my rules doesn't explicitly acknowledge this possibility, but contains nothing that would invalidate it.) In this way you can at least sometimes have everyone participate, even under these rules. That will only help on some rolls; for example, I stopped allowing aid on Perception a while back (PCs who've already spotted something can point it out to ones who haven't, but that's not the same thing). But on attempts to forage I agree that it seems it should be possible. You might flavour it (no pun intended) as the low-skilled characters picking berries or doing similar relatively easy tasks, while the more skilled ones hunt or fish.

(In my own die-pool based system, the two "main" characters would roll their usual dice, everyone else would pick one of them to help out and roll one die, and the chosen character basically treats that die as part of his/her roll. 3rd and 4th generation D&D have Aid Another, which works similarly. 5th is more coarse-grained in this respect, but there's no reason you couldn't grandfather in the older version of helping if you like it better.)
 
Last edited:

jeffh

Explorer
I think these premises need consideration.

When must the whole party complete a task? If only half of them can cross an underground river, well, either the successful ones lower the drawbridge from the other side, or the non-swimmers find another route. If this is just a way to address splitting the party, you will hopefully agree that splitting the party occasionally is not a bad thing.
Even in these cases, it's preferable if everyone can get across at once, no? I don't see how the (in any case situational) existence of alternatives invalidates the general point.
 

DMMike

Game Masticator
Even in these cases, it's preferable if everyone can get across at once, no? I don't see how the (in any case situational) existence of alternatives invalidates the general point.
You're right - the point isn't invalidated. There are times when the PCs and GM want the entire party to be in the same situation, and it's too time consuming to wait for everyone to roll.

Bear in mind: rules have agendas. You can't completely separate the rules from the story. So the Two Roll rule will likely have an effect of making the party "sticky," like the Attacks of Opportunity made some D&D duels sticky. Not a bad thing, but urging gameplay in one direction might lead it away from another.

Personally, I usually don't find it preferable for the party to act as a whole. When the PCs have to compromise, sacrifice, and even go solo...that's when you get some good drama. My PCs just spent an entire session together in a playhouse, just to get mostly separated by the end of this last session. Now, they're shaking in their boots because they feel their lifelines pulled taut, in all directions!
 
You're right - the point isn't invalidated. There are times when the PCs and GM want the entire party to be in the same situation, and it's too time consuming to wait for everyone to roll.

Bear in mind: rules have agendas. You can't completely separate the rules from the story. So the Two Roll rule will likely have an effect of making the party "sticky," like the Attacks of Opportunity made some D&D duels sticky. Not a bad thing, but urging gameplay in one direction might lead it away from another.
That is a good and interesting point. I'm probably influenced by the fact that my players are not just willing but often eager to split the party, either for RP reasons or to maximize freedom of action (as opposed to survivability). So party stickiness isn't a rules feature that I need to prioritize.
 

jeffh

Explorer
I meant "preferable" tactically. But I must say I see much less conflict between gameplay and story here than some of you appear to. If I'm trying to think in character (which, as a player, I sometimes do and sometimes don't), I am if anything even harder pressed to think of a good reason why my top priority should be something other than survival than if I'm thinking consciously of the tactical game. Unless I've more or less consciously created a character with a literal death wish, it's very hard to justify not having self-preservation be a major motivating factor - not the only one by any means, but a big one, that wins nearly any conflict. People who talk like it's good role-playing to ignore tactical considerations puzzle me.
 

Advertisement

Top