• The Infractions Forum is available for public view. Please note that if you have been suspended you will need to open a private/incognito browser window to view it.

The Inverted Dice Pool (Yet another core mechanic)

fheredin

Registered User
Banned
Validated User
The inverted dice pool is something I've been toying with for some time as a way to cross rules-light and crunchy gameplay. I would appreciate comments and criticisms on the mechanic itself or if I've explained it properly. There are also a couple glitches I will discuss at the end.

An inverted dice pool is the logical inverse of a traditional dice pool. Where most dice pool systems add dice and look to roll high, this system has a set pool of dice, it improves your roll by shrinking your step dice a size, and you are looking to roll low.

  • You start by collecting three step dice representing attributes or skills immediately involved in the action. Smaller dice are better.
  • You roll them and count any dice which rolled 4 or lower as a success.
  • If you roll a 1, you may explode the die, or reroll it to see if it provides you with another success. You may keep rerolling it as long as it keeps rolling 1.
When none of your dice read a 1, your dice have cooled off. Certain mechanics--such as another player assisting you, spending metagame currency, or spending extra AP in combat--can give you a force, which restarts a cooled off die. You should always spend your forces on the smallest die you can to get the best chance of success. However, you can only force a die to restart once. After the die has cooled off a second time it is iced, and you'll have to move down to the next die until you run out of dice.

After the roll, you spend successes to power outcomes in the narrative in a one-to-one fashion. Say, for example, you are picking a difficult lock and roll three successes. You have to spend one success to overcome the difficult part, one success to successfully pick the lock, and then you can freely choose what to spend that last success on. It could be to pick the lock quickly, quietly, or you could spend it so no connected traps trigger. But you only have one extra success to play with, so you only get one critical effect. If you don't get enough successes, this logic can also work in reverse; failing to pay a success can result in a partial success with an appropriate narrative consequence for what you skipped.

Discussion

There are two reasons for the wonky roll-under format. It lets exploding dice become a lot more potent at high level, but it also means you can use both d20s and d4s as step dice. Historically, step dice systems struggle to get these dice to cooperate. Flipping things upside down makes the big gap between d20 and d12 less noticeable and the progression follows a better curve. I also think it's nifty how the specific combination of dice you are using combine with forces to make a sort of supply and demand curve of sorts. The value of adding another force scales in value to the next die it would reroll.

However, there are a few problems I want to discuss. For starters, the designer-end math behind this thing is a nightmare on wheels, so I've had to resort to intuitive balancing. The second is that this system is really, really crit-happy and extra successes can and do cause analysis paralysis. Finally, the granularity is...really low. It has a ton of levers and knobs you can manipulate to make the system feel crunchy, but it isn't so good at being crunchy. However, if I adjust the numbers to increase granularity, the analysis paralysis on crits will only get worse.

Comments? Criticisms?
 

Knaight

Registered User
Validated User
Want me to hook you up with some Octave scripts for designer end math? It's my go to when I get in over my head in Anydice.

I'll also say that this generally seems interesting, and that I particularly like the unexploded success pattern of 1/5, 1/3, 1/2, 2/3, 1. That pattern doesn't necessarily play well with small pools though, and as you said it's crit heavy. That sort of long tailed distribution isn't necessarily a problem though.
 

Naeddyr

two
Validated User
If you think the granularity is too low, have you thought about using the same kind of die for your static sized pool, and change the target number to be rolled?
 

fheredin

Registered User
Banned
Validated User
Want me to hook you up with some Octave scripts for designer end math? It's my go to when I get in over my head in Anydice.

I'll also say that this generally seems interesting, and that I particularly like the unexploded success pattern of 1/5, 1/3, 1/2, 2/3, 1. That pattern doesn't necessarily play well with small pools though, and as you said it's crit heavy. That sort of long tailed distribution isn't necessarily a problem though.
If you can write the scripts, please. I can't, and would very much appreciate it. I can't really balance high end play correctly beyond guess and check.

EDIT: And I actually didn't notice the unexploded probabilities followed the Fibbinacci sequence until you pointed it out. Huh. /edit

At the moment the worst part is the extra successes causing analysis paralysis, and I'm looking for ways to cope with it. At the moment I am looking at omitting damage rolls in favor of a success siphoning mechanic. For example, if you attack with a taser and roll 2 extra successes, you could split them between dealing critical damage and a stun siphon. Additionally, I'm thinking of a called shot mechanic to help bleed off even more successes.

That would obviously increase the crunch factor, but it may at least reduce the analysis paralysis problem during combat.

If you think the granularity is too low, have you thought about using the same kind of die for your static sized pool, and change the target number to be rolled?
I have. In fact an early iteration was a d12 pool. Changing TN just doesn't work as smoothly as a static TN. It's not bad, but players never seem to drop into rhythm with the system. As a secondary problem, it can't quite capture as much nuance from the players' stats.

In general, I think that granularity is an overrated metric; players can seldom discern things to within 5% accuracy, so there's no real need to go below that. Which was why I experimented with this in the first place. However, the inverted dice pool has the opposite problem; with successes ranging in the 0-5 range, the granularity is about 17% per success. That's...big. Not unplayable, but big.
 

thorya

Statistical out-liar
Validated User
So I love this sort of mechanic. It's very similar to my default one-shot rules and my current project, so I might be bias. But people get it fast, resolution is fast, and it generally worked, so I think you're on a good track. Roll low solves so many problems with stepping dice. Granted I mostly used it for quick fast and loose games, but I never found the granularity to be a problem.

D4, D6, D8, D10, D12, D20

Is six steps, which is plenty for most things and because you've got three dice all varied independently you've got lots of places to make adjustments. You can add a D5 for another step.

When I have used it in the past for one shots it was two dice, not three, and I used variable TN from 2 to 6ish (managed on the GM side of the screen), that gave me the following results.

2 successes with total still less than equal TN- critical success
2 successes- success
1 success- success at cost
0 successes- failure
Lowest die twice TN- critical failure

You might be able to solve the analysis paralysis problem with something similar and avoid the time sink of an exploding die and people needing to assign successes to things.

Maybe-

3 Successes- critical success
2 successes- success
1 success- success at cost
0 successes- failure

Then there's no meta analysis to cause paralysis. You rolled well, you did the thing well. No trying to guess if the lock has a trap or you'll just be wasting the success or mechanics to bleed off successes.

Also, I think a variable TN will be easier than variable required successes, which is what you're doing now even if you don't call it out directly.
 

SladeWeston

Registered User
Validated User
I also have a good number of articles about exploding dice math HERE and HERE.
As for complexity, analysis paralysis, I would suggest handling how Star Wars RPG handles it. Their Advantage/Threat system is very similar to what you are suggesting with spending crits. The way they got around too much complexity is to provide a relief valve of simple yet always beneficial options for spending their advantage, in addition to tricky, situational or most complex ways of spending it.
Example, the player rolls and gets a handful of advantage. They use some of it to activate the knockdown trait on their weapon but have 2 remaining and don't really have a good situational use left for them so they decide to use them to give an extra dice to the next ally.
The way this will often work in practice is that players get familiar with the options they always have and when they start to feel like their AP is getting the better of them, they will just fall back on that list, which is full of generically useful options. This lets them load tons of crunch into talents and class features, but if a player wants to avoid those options they can, knowing they can fall back on the generic options they will have at their disposal. Definitely worth taking a look at that system if you haven't.
 

fheredin

Registered User
Banned
Validated User
You might be able to solve the analysis paralysis problem with something similar and avoid the time sink of an exploding die and people needing to assign successes to things.

Maybe-

3 Successes- critical success
2 successes- success
1 success- success at cost
0 successes- failure

Then there's no meta analysis to cause paralysis. You rolled well, you did the thing well. No trying to guess if the lock has a trap or you'll just be wasting the success or mechanics to bleed off successes.

Also, I think a variable TN will be easier than variable required successes, which is what you're doing now even if you don't call it out directly.
Thank you for your comments and encouragement. Hmm, that's an interesting thought. I'm not particularly keen on abandoning the narrative and mechanical power that spending successes brings, but I might be able to "trapdoor" it out of less important checks by implementing something similar to that. I already have a diceless resolution model which just equates the 6 step dice to a letter grade and the better letter grade wins. I use it for stuff like passive perception checks and search checks where you don't necessarily want to tip players off or interrupt game flow with dice.

Perhaps an intermediate mechanic where you just have difficulty tiers as 1, 2, or 3 successes? That's pretty simple, almost certainly faster and lighter, and sits pretty easily right between the diceless streamline and the full mechanic. And then any moment you want, the GM or the player can turn on the success spending rules to do something specific. Nothing about the dice or the total number of successes needs to change. If one of the players turns on the mechanic, they will already know what they want to spend successes on and there'll be virtually zero analysis paralysis. The decision to spend the success and to activate the spending success rules happened at the same time, and that moment probably happened before the player even reached for dice.

This is weird. I've seen all sorts of systems with multiple core mechanics, but generally those are for apples and oranges like older versions of D&D with attacks and saving throws. I don't think I've ever seen an RPG where parts of the core mechanics fold out of the way if you aren't using them quite like we're talking about here. It would probably involve some serious player and GM learning curve...but it also might be totally worth it.


I also have a good number of articles about exploding dice math HERE and HERE.
As for complexity, analysis paralysis, I would suggest handling how Star Wars RPG handles it. Their Advantage/Threat system is very similar to what you are suggesting with spending crits. The way they got around too much complexity is to provide a relief valve of simple yet always beneficial options for spending their advantage, in addition to tricky, situational or most complex ways of spending it.
Example, the player rolls and gets a handful of advantage. They use some of it to activate the knockdown trait on their weapon but have 2 remaining and don't really have a good situational use left for them so they decide to use them to give an extra dice to the next ally.
The way this will often work in practice is that players get familiar with the options they always have and when they start to feel like their AP is getting the better of them, they will just fall back on that list, which is full of generically useful options. This lets them load tons of crunch into talents and class features, but if a player wants to avoid those options they can, knowing they can fall back on the generic options they will have at their disposal. Definitely worth taking a look at that system if you haven't.
Hmm. I have in the past dismissed Star Wars because I don't like systems with custom dice. That may have been premature. I certainly need a few good ways to spend successes, and I will totally mine it for ideas.

Also, thanks for the articles. I've already read the Anydice article when I was deciding whether or not to include exploding dice. It mirrored my own experience with Savage Worlds that players tend to like d4s because they get a lot of explosion rushes, but that doesn't actually translate to success in the check. Exploding dice are very good at crossing player's brains like that.

Including good ways to spend successes is obviously ideal, but what do you think of folding the whole mechanic away if you don't need it like above?
 

thorya

Statistical out-liar
Validated User
I think your fold out of the way mechanic sounds like a good solution. Definitely fits my Golden rule of only spending time on something if it matters.

I have seen this sort of solution in games for equipment rules. If it matters, you can get nitty gritty on what equipment gives X or Y bonus/penalty, but generally it's just you have the equipment to use your skill and you resolve as normal. Seems to work well, especially for bridging the Gap for groups that want varied levels of crunch. Though players that choose to engage with spending successes every time will probably gain a slight advantage in play.
 

SladeWeston

Registered User
Validated User
Generally speaking, I'm a huge fan of non-binary resolution mechanics and for crunchy systems, they are one of the first things I look for. I think dropping them would remove a lot of what I found interesting about your system. At least imo. Also, hit me up with a private message if you'd like access to SWRPG or Genesys pdfs.
 

fheredin

Registered User
Banned
Validated User
I'm torn on that. I already have those systems, I just hadn't made the connection to mine them for crit mechanics. And...oh dear, they love those dice symbols. And Link, when Navi speaks, use ⬆ to listen well to her words of wisdom.

I mean goodness gracious.

The thing is that a trap door mechanic should in theory allow a group or GM to pick the depth they want to use. I agree with thorya thorya that this will translate to an advantage in play, but an optimization in how you play as opposed to the character you made may trigger constructive envy.

To make this clear, I'm not talking about removing the spending success mechanic outright. I'm talking about making three parallel CRMs which are (mostly) compatible.

  • A diceless resolution for invisible checks,
  • A basic pass/ fail mechanic for when you just want a yes or no answer, and
  • The full success-powering shebang.

If I were playing this hopefully soon-not-hypothetical system for a hypothetically ideal group, we'd always use the success powering mechanics and not the intermediate CRM. I suspect you would, too. However, not everything in this universe is ideal and I know a number of players who would cause lag if that was their only choice. Besides, one of my key complaints with RPGs out there is that they are such specialists at doing one bloody thing that they tend to turn slight differences in player capacity and tastes into full group schisms. I hate that part of the hobby.

However, adding more ways to spend successes never hurts. Well, until you add too many, anyway.
 
Top Bottom