Your Favorite Resolution Systems [Archive] - Giant in the Playground Forums

Telok

2022-07-14, 11:33 AM

Paranoia's d20 blackjack system: you have a "thing", like melee combat, nuclear engineering, boot licking, or ink squirting, with a 1-20 number. Roll that number or under, 1s are success with a drawback, the number exactly is success with a boost, 20s are comedic utter failure (because its Paranoia and possibly lethal because you have clones you need to use up). No math at the base rolls, opposed checks are dirt simple 'who rolled higher yet still under their number, you can do margins of success easily, modding the game tone or content is super easy. Bonus; its really fast, super newbie friendly, and you only need one die.

Also really fond of 3d6 vs 11 or less. All character stuff is a number (default 11) you need to roll under. Or you can flip it to emulate 3d6 + bonus vs 11+ if you have to, then all the character numbers are just low bonuses. Again, simple, flexible, easily moddable (you can go 4d6 drop lowest/highest if you don't want numerical modifiers), and its a normal probability distribution so you can't "roll too much" and bust the gameplay without also having to screw up somethings else like stacking penalties all over the place. Bonus; slots into AD&D really super well and makes the difference between say, strength 10 & 17 matter more than 1/6 times.

Generally I like 1-3 die fast & easy (theoretically the wotc d20 stuff should be but the exception stuff and constsntly changing numbers don't), or some sort of well mathed bell curve with 10 or less dice that doesn't break at the extremes of character math or too many/few rolls.

olskool

2022-07-14, 01:03 PM

My favorite resolution system is actually a pair of systems. The first is the roll-under percentile system as done by the Design Mechanism's MYTHRAS rpg. You add to characteristics scores from 1 to 20 together to get your base skill before training. You then roll under to succeed. However, there are varying degrees of Success. Rolling under your skill is a Success but we break down the level of Success as follows. Please note that I "homebrewed" the Degrees Of Success to avoid doing math during the game. The levels of Success are...

Simple = Roll under modified skill
Outstanding Success = Roll under 1/2 Skill (this doesn't exist in RAW)
Exceptional Success = Roll under 1/10th Skill rounding UP (This is 1/5th Skill in RAW)
Critical Success = Roll DOUBLES under Skill with 00 counting as "zero, zero" and NOT "100" (RAW is 5% of Skill value).
Fumbles = DOUBLES OVER Skill (RAW is 5% of the total over Skill Level counting down from 100).

My Other resolution system is basically a D20 roll UNDER system that mimics the percentile system above. It is a modified (by me) version of GDW's old V2.2 resolution system from TWILIGHT2000, MERC2000, and TRAVELLER: THE NEW ERA rpgs.

In that system, you combine a Skill Level rated from 1 to 10 with an Attribute Rating from 1 to 10. This combination of SKILL and ATTRIBUTE was known as an ASSET and was prerecorded with all the Skill Difficulty Levels on the PC's character sheet. So, if your PC needed to figure out what their Climbing Asset was, they would combine STR, AGL, and CON together and add that to their Climbing Skill. So if the PC above had an average of 6 for their three Attributes and a Climbing Skill of 5, that PC would succeed on an AVERAGE Task on a roll of 11 or less.
Tasks would be assigned a DIFFICULTY LEVEL and these were taken from TRAVELLER but then modified. The DIFFICULTY LEVELS were...

VERY EASY Task = (3 X Skill Asset) so a typically Average PC (Asset 10) would succeed on a roll of 30 or less (a 20 always fails).
EASY Task = (2 X Skill Asset) so a typical PC would succeed on a roll of 20 or less.
ROUTINE Task = (1.5 X Skill Asset) so a typical PC would succeed on a roll of 15 or less (75% chance).
AVERAGE Task = (Skill Asset) so a typical PC would succeed on a roll of 10 or less (50% chance).
FAIRLY DIFFICULT Task = (0.75 X Skill Asset) so a typical PC would succeed on a roll of 8 or less (40% chance).
DIFFICULT Task = (0.5 X Skill Asset) so a typical PC would succeed on a roll of 5 or less (25% chance).
FORMIDABLE Task = (0.25 X Skill Asset) so a typical PC would succeed on a roll of 3 or less (15% chance).
IMPOSSIBLE Task = (0.1 X Skill Asset) so a typical PC would succeed on a roll of 1 or less (5% chance).

Modifiers would be expressed in TWO ways. The first way is as a numerical Bonus or Penalty of from 1 to 3, which would SUBTRACT FROM (for a bonus) or ADD TO (for a penalty) the PC's DIE ROLL for the Task Check. This was done both to reduce any math needed and to allow bonuses to have a larger effect on harder skill tasks. A bonus of 1 on a VERY EASY task is probably not going to make a difference to the PC... BUT, on an IMPOSSIBLE Task, it would DOUBLE your chances of success!

Optionally, for larger changes, a DIFFICULTY SHIFT could be used. You bump the chances from AVERAGE to ROUTINE or from DIFFICULT to FORMIDABLE.

In addition, you still have Tiers of Success. On a roll that is equal to or less than your base Skill Rating (that 1 to 10 number) for an AVERAGE Task, you achieve an Outstanding Success (a roll of 5 for the PC described above). IF you roll under HALF of your Skill Rating (a 2 for the PC described above), you achieve an Exceptional Success. This does get modified by the Task's Difficulty Levels though, so a DIFFICULT Task would need a 3 or less for Outstanding Success and only a 1 would net an Exceptional Success.

Finally, there was a version of ADVANTAGE & DISADVANTAGE in Twilight2013 that we adapted to our version of the game. Unskilled Task tests required you to roll TWO D20s and take the HIGHER SCORE (because this game is a roll-under system). Having an "Expert System" or other guidance gave you the ability to roll TWO D20s and take the LOWER ROLL (to represent the assistance on the Task).

The system was fast in play and allowed you to tailor tasks to fit any situation.

Vahnavoi

2022-07-14, 02:46 PM

One player is nominated as final authority for events in the game. Other players pitch possible courses of action to the chosen player. Chosen player decides based on whatever criteria they find reasonable which pitch to accept, or rejects them all in order to pose something else entirely. Any player disliking their choice may leave the game.

Satinavian

2022-07-14, 02:53 PM

My favorite one is from Splittermond :

Roll 2d10 + mod (usually skill points + 2 linked attributes, handily avoiding sad builds) against DC.
Per every 3 points above DC one gets a degree of success and pretty hard rules for what that means. Per every 3 points below one gets a degree of failure and usually also hard rules for that.

Now a roll of 1,1 or 1,2 means a fumble, which usually means 3 additional degrees of failure but might have special rules in some circumstances. Likewise aa roll of 9,10 or 10,10 means a triumph with 3 additional degrees of success.

So far the regular way to roll. But it is allowed to take a safe approach or a risky approach instead

Safe approach : the lower of the 2d10 is ignored. That significantly lowers the average, but makes fumbles impossible

Risky approach : Roll 4d10, take the best 2. But if any combination of dice is a 1,1 or 1,2 you have to take those and it counts as fumble. This means significant higher average, but not higher max possible results and higher fumble chance.

So nearly every roll has a tactical decision attached.

Black Jester

2022-07-14, 03:25 PM

My favorite resolution system is actually a pair of systems. The first is the roll-under percentile system as done by the Design Mechanism's MYTHRAS rpg. You add to characteristics scores from 1 to 20 together to get your base skill before training. You then roll under to succeed. However, there are varying degrees of Success. Rolling under your skill is a Success but we break down the level of Success as follows. Please note that I "homebrewed" the Degrees Of Success to avoid doing math during the game.

Hard to beat percentage based roll under for maximum transparency and ease of use. Mythras wins in this regard because it is such a well-written system and I like how clearly abilities and skill values interact. Definitely one of the best. The most fun of this are the combats: when every successful action offers the opportunity to manipulate your opponent, hurt him in creative ways or gain some sort of tactical advantage. There are so very little things in a combat that are as rewarding as punishing an attacker for daring to strike at you incompetently and let them pay the price.
Here is an example of a duel run in Mythras, action-by-action, accompanied by a neat little short film presenting the fight. (https://elruneblog.blogspot.com/2020/12/samurai-duel-combat-example-with-mythras.html)

However, my personal favourite is the old(ish) Cubicle 7 Doctor Who System (haven't played the new edition, can't say a thing about it): You roll 2d6, adding both a Stat and a Skill, both roughly ranging from 1 to 6, to the total result, trying to beat a target number, the higher the result the better. Again, super easy, very intuitive, and non-gimmicky (I don't like games trying to be clever and failing more or less spectacularly at reinventing the wheel). The fun part about the system is that the degree of success is well done, effectively adding the simple words "and" and "but" to the simple yes or no question of "did you succeed?", with a lot of free narrative space for the player to come up with these factors, not just for spectacular successes, but most importantly for setbacks.

Let's say, you try to repair the O2 recycling plant sabotaged by a mysterious foam monster on your moon base under siege. You roll on Ingenuity + Craft to jury-rig a solution, against a difficulty you might or you might not know. On a great success, you get a result of "yes, and...": Not only did you repair the recycling system, saving you and the cosmonauts from a grisly death by suffocation, you also found a clue about the foam monster's hideout!. A normal success ("yes") simply allows you to repair the unit. A marginal success, or "yes, but" would also allow you to repair the dang thing, but at a price: probably your loud banging has attracted the foam monster to your location - and know it hunts you through the dark corridors!
On the side of the success spectrum, you also get a marginal failure, or "no, but": While you couldn't repair the recycler (and the rest of the scenario will run under a ticking clock, until air will run out), you have at least located a weak spot: the foam monster is averse to great heat. Maybe you can use that in a later confrontation. Besides a normal, plain failure, however, you also have a critical failure, or "no, and..", indicating not only that you failed your task, but somehow made the whole situation worse: The O2 recycler might be still broken, but now also releases another gas into the station's atmosphere that makes everybody dizzy and intoxicated (fortunately for you, this might also affect the foam monster).

Xervous

2022-07-15, 12:02 PM

Xd6CS>4CF=1

Dicepool, 5s and 6s success, 1s are tallied for fumbles/complications.

Compare successes to target number or opposed pool. More than half your pool is 1s? Complication. Failed and complication? Fumble. Easily factors in degree of success/degree of failure.

And most importantly it’s a curve that stretches rather than moves. What did that mean? Failure gets less likely but never hits 0. Complications less likely but never 0. The value of adding extra dice to the pool never hits 0, but you often reach a point where you deem it’s good enough to go focus on other things.

LecternOfJasper

2022-07-15, 01:50 PM

Black Jester

2022-07-15, 01:54 PM

That's pretty neat, is that tied to a particular system I can check out? I've been making my own d6 system and could use some inspiration.

It reads a lot like Shadowrun. However, that is a high complexity, very bloated rule system in most versions I know.

clash

2022-07-15, 03:16 PM

I'm rather fond of a system I played where it has dice pools that scale with scaling dice size.

So your strength attribute goes from 1-5 indicating how many dice you get to roll.

Skill attribute scales all your dice from d4 to d12 with 1 being a failure 2-5 being a success and 6 or above being 2 successes each.

tensai_oni

2022-07-15, 03:35 PM

Systems that aren't just binary you succeed/you fail. Doesn't matter if I roll 1d100, 2d10, x dice and keep y, if the result is still binary pass/fail then it's the same boring stuff for me.

Powered by the Apocalypse games have a good system because even a fail is supposed to progress the fiction, which means diagetically a failed roll can actually succeed, but it creates more complications than you started with. And that's good! It means failure is actually interesting and isn't just "well that was time/effort wasted".

At least that's the idea because for every well-designed PbtA game with defined but interesting results for each move's full hit, partial hit and miss, you get another, poorly designed game where the results are all "on a full hit you do it, on a partial hit you do it but with a complication" "what complication?" "dunno make something up lol".

False God

2022-07-16, 08:49 AM

Anything dice-pool, for the reasons Xervous said.

olskool

2022-07-16, 09:34 AM

That's pretty neat, is that tied to a particular system I can check out? I've been making my own d6 system and could use some inspiration.

FASA's Shadowrun used the Dice Pool mechanic. The newest STAR WARS is the most recent version. The White Wolf "worlds of Darkness" series (Vampire, The Maskerade) uses D10s for their dice pool systems.

Savage Worlds actually pushed a "Die Pool" system and Free League expanded upon it. In Savage Lands, you roll a single DIE for an Attribute or Skill but there are certain characters who get an additional die (D6) for being a "wildcard." During play, the die can step up to the next larger die size (ie a D4 to a D6) with bonuses or be reduced to the next lower die (ie a D8 to a D6) with penalties. any roll of 4 is a success and rolling 4 over that base chance of 4 (the game's universal target number) gets you additional successes.

Free League's system uses only D6 and above but they give you a Die for any Attribute plus a Die for any Skill (some game versions give a Die for equipment or magic too). These step up or down just like Savage Worlds but the target numbers are different. They start with 6 being a Success with each increasing die size adding Successes (D8 can have 2, D10 can have 3, D12 can have 4). Each die can have multiple Successes and penalties and bonuses step the dice you are rolling in tandem (you first step up a lower die to match the higher one) as well as stepping them down in tandem (the higher die is always reduced first).

We actually modified SHADOWRUN 1e/2e to a D10 dice pool method. Our version went like this...

You rolled a number of D10 for success equal to your Attribute or Skill. We DITCHED THE DICE POOLS. D10s could "explode." On a roll of 10, you would roll ANOTHER D10 for added successes. Thus a person with a lower level of skill COULD succeed at a harder Task through the use of "exploding" rolls of 10. A Catastrophic Failure occurred on a roll of all "1s." Thus, Skill was the gatekeeper here. Rolling one D10? You have a 10% of a Catastrophic Failure. Rolling 3 D10s? Not as likely to occur.

Target Numbers were set from 2 to 10. Any roll equal to or greater than the Target Number succeeds. We set the number of SUCCESSES needed based on the color codes for Decking security ratings. The chart below gives the number of Successes needed to complete a task...

GREEN = 1 Success needed
YELLOW = 2 Successes needed
ORANGE = 3 Successes needed
RED = 4 Successes needed
BLACK = 5 Successes needed

By using these color codes, we could quickly establish a Task's difficulty. For example, you might say: "Opening that door is a 5 GREEN Task." and I instantly know that I need to roll a 5 or more and get one Success. If you said: "hacking that terminal is a 4 RED Task." I'd immediately know that I need a 4 or higher but also need FOUR Successes! The system was a fast and intuitive hack of SHADOWRUN!

SimonMoon6

2022-07-16, 10:06 AM

I like Mayfair's DC Heroes RPG's system:

Every die roll you make (combat, skill, etc) involves the following:

There are four numbers that matter. In combat, the first two would be the DEX scores of both combatants (or a skill that replaces DEX or whatever) and the second two would be the STR (or other damage value) of the attacker and the BODY (plus modifiers possibly) of the defender. In skill checks, other numbers are used obviously, but there's always a "see if you hit/succeed" part and a "see how well you succeeded/how much damage you dealt" part.

There are tables for each of these two parts. The first table deals with "see if you hit/succeed". The dice you roll are 2d10 (with "exploding" dice, meaning roll again and add when you get doubles) with a result of 2 being an automatic failure. Against an equal opponent, you need to roll an average roll of 11 to hit/succeed. For lesser opponents, you need a smaller number; for more difficult opponents, you need a larger number. (These are all on the table.)

If you roll more than an 11 and also succeed, you may get "column shifts". This indicates that you have hit better than normal or you have succeeded better than normal. The higher you rolled, the more column shifts you get. These will "shift" your results on the second table to be more beneficial.

On the second table, you compare your STR (or whatever) to your opponent's BODY (or whatever). There will be a number on the table that represents how much damage you deal (or how well you succeeded), but then the column shifts will move your result to a column with bigger numbers.

And the best part is that, for tasks with unclear difficulty numbers, there is a table that lists appropriate numbers depending on how difficult the GM thinks the task should be, ranging from "easy" to "Herculean".

Climbing a wall with handholds? Easy. Climbing a wall without handholds? Difficult. Climbing a sheer glass wall covered in slippery oil? Herculean.

PhoenixPhyre

2022-07-16, 11:13 AM

I'll be honest. I care very little about things like bell curves, fidelity, etc. What I care about is speed. Optimize the hot path, and the core resolution mechanic is as hot as it gets. The more operations (especially table lookups) you need to accomplish the core "did thing succeed" meta-task, the fewer actions you can take in a fixed-length session. Which means more time sitting around waiting for things to resolve and less doing things.

NdX roll under? Fine. As long as any potential modifiers are written down and (at the timescale of most operations) fixed. d20 + mods vs static TN? Fine, with the same criteria. NdX vs fixed TN? Probably ok. Dice pools can work (because counting successes is fairly fast), but if every attack (to take one example) needs 4 rolls and decisions about how to split your dice pool (how many to block, how many to hold for another attack, etc), no thanks. Conditional bonuses, conditional rerolls (including exploding dice)? I'd rather avoid those. When the resolution system is invoked, all the decisions should have been made and it should just resolve. I care less about piloting the rules and more about what's going on in the game's fiction itself. The resolution system is there as a tool to be used to resolve uncertainty, not the primary draw of the system.

And anything involving looking results up on tables, especially conditionally (where deciding which tables to reference is influenced by results from a previous table lookup) is just flat out. Table lookups either require massive delays (orders of magnitude slower than other operations, similar to having to reach out to the network instead of looking in the processor cache) or holding tables in memory (which has major mental overhead and often requires falling back to physical tables because memory is faulty). Tables are fine for rare or asynchronous things (things you can do ahead of time, like random loot). But not on the core resolution path.

LibraryOgre

2022-07-16, 11:26 AM

I like Savage World's dx v. 4 (usually), with important characters getting a d6 that is considered separately.

So, if I have a d4 in Fighting, I roll a d4 and a d6, and take the better of the two. My opponent, who isn't a main character, might be rolling a d6, and might have a higher parry, but I have that little bit better chance to succeed because I am Dramatically Important.

Cluedrew

2022-07-16, 04:46 PM

I think I have to go with the Powered by the Apocalypse 2d6+stat vs 7-9. It is fast, easy to use, easy to remember and has multiple tiers of success. And that is just the dice results themselves; although the choices of how the miss/weak hit/strong hit results are much more controversial but I enjoy the entire "success with consequences".

It does have downsides, for instance you cannot us it in a numbers-go-up system, but none of them bother me too much.

oxybe

2022-07-17, 01:04 AM

Depends.

I'm pretty sure everyone can agree that no one resolution system fits all cases.

My group's recently been playing a lot of the Alien TTRPG. system is pretty simple: attribute+skill dice pool of d6's, target number is 6. one 6 in the pool is a success, extra successess can be spent on various skill tricks. the thing about the system is that there is also a stress system. stress adds extra d6's that can add to your success overall, but the stress dice can also effectively crit fail, causing your character to spend all the ammo in their weapon, blue screen on the spot for a turn, run away or get more stressed out. take too much stress and your character can develop neurosis.

You want some stress for the extra dice it gives, but too high and if stress procs you may get one of the more debilitating effects, and a high stress means a higher chance of it proc'ing.

Would I want a system like that for D&D? nah. but it works in the Alien TTRPG.

Shockwave

2022-07-17, 04:56 AM

Tanarii

2022-07-17, 05:42 PM

Faily

2022-07-17, 07:51 PM

I am very fond of FFG's Star Wars-system (Edge of the Empire/Age of Rebellion/Force and Destiny). Not one I thought I would like so much, but I've really come to like how it is much more creative in handling outcomes and resolutions from a dice-roll.

A successful roll can come with complications, a failed roll can come with positives, and Triumph/Despair really creates standout moments. I think it works best with a playgroup that is creative and quick to offer suggestions to these rolls, because it can become a bit too much for a GM sometimes to determine the creative outcome of every roll, and I've had great experiences with it in Play-by-Post (where you often have a bit more time to really think on the dice results).

Overall one of my favorite systems for interesting and dynamic results.

Xervous

2022-07-18, 08:00 AM

I agree with your overall point, but replace d6 to d10 and have adjustable TN's Basically world of darkness.

Adjustable TNs drastically change the math behind everything. SR3 had adjustable TNs while SR4 had a static TN, it’s an order of magnitude more complex to deal with moving TNs and variable dicepools to the point that the value of player options are ambiguous. Players can understand larger and smaller pools without being able to quote the curve distributions, but can you pick the better roll out of 5d10TN7 and 7d10TN8 without a calculator?

Black Jester

2022-07-18, 10:01 AM

I'd actually love to see a system where there were 2 parallel resolutions systems.

One simple one for combat and other situations where success is measured as the combination of many actions each with its own resolution check, basically either d20 or d100 plus simple and not often changing mods vs fairly straightforward TN.

One complex system, preferably bell curve based, that strongly reflected character(s) skill and lots of fiddly modifiers, for times when a whole host of player decisions is ending up being abstracted into a single roll for success. Also with degrees of failure and degrees of success.

The German RPG The Dark Eye handles things that way - combat actions are solved with a realtively simple, D20-roll under system, while skill tests are used with 3 d20, compared to three stats. However, last time I checked, that was a big, very cumbersome system with a lot of moving parts, options and modifiers.

Players can understand larger and smaller pools without being able to quote the curve distributions, but can you pick the better roll out of 5d10TN7 and 7d10TN8 without a calculator?

The much more important question is: do players need to know? If transparancy of success rate is relevant, anything but a percentage-based system is going to be opaque to someone. Does it matter, for the actual fun in the game to know the absolute chance, or is it sufficient, to get a rough impression of a) how good your character is supposed to be) and b) how difficult a task is?

Shockwave

2022-07-18, 11:42 AM

Adjustable TNs drastically change the math behind everything. SR3 had adjustable TNs while SR4 had a static TN, it’s an order of magnitude more complex to deal with moving TNs and variable dicepools to the point that the value of player options are ambiguous. Players can understand larger and smaller pools without being able to quote the curve distributions, but can you pick the better roll out of 5d10TN7 and 7d10TN8 without a calculator?

I understand precisely what you are saying, I don't think it's a bad thing. SR3 is probably my favourite setting + System combo

The much more important question is: do players need to know? If transparancy of success rate is relevant, anything but a percentage-based system is going to be opaque to someone. Does it matter, for the actual fun in the game to know the absolute chance, or is it sufficient, to get a rough impression of a) how good your character is supposed to be) and b) how difficult a task is?
This pretty much, why should we always know the best method to get a desired result?

Xervous

2022-07-18, 12:03 PM

This pretty much, why should we always know the best method to get a desired result?

The main benefit from having such a detailed, layered resolution system is in the minute differences it can quantify for players to choose among. If opposing choices are ambiguous in play the system isn’t giving the players much more meaningful agency than “what’s behind door 2?” game shows. It’s just wasted fanfare. Ambiguity from a character creation standpoint isn’t anywhere near a concern as those who really care about the singular decisions can run the math on their own time and make those choices.

In other words, there shouldn’t ever be a play situation where a player has to weigh moving the TN against adjusting the dice pool.

Telok

2022-07-18, 03:59 PM

The main benefit from having such a detailed, layered resolution system is in the minute differences it can quantify for players to choose among. If opposing choices are ambiguous in play the system isnÂ’t giving the players much more meaningful agency than Â“whatÂ’s behind door 2?Â” game shows. ItÂ’s just wasted fanfare. Ambiguity from a character creation standpoint isnÂ’t anywhere near a concern as those who really care about the singular decisions can run the math on their own time and make those choices.

In other words, there shouldnÂ’t ever be a play situation where a player has to weigh moving the TN against adjusting the dice pool.

True, but what does that have to do with most of the variable TN dice pool systems? Generally you're looking at something like "task A is TN X, you can use your pool of 7 dice or your pool of 10 dice" as the complexity, meaning there's probably no actual complexity because more dice is better.

Now with a slightly more complex system where your point might matter is something like DtD40k7e with it's roll Xd10, keep Y of them vs TN Z. So you could try for something like Int+Tech-Use at 7k3 (roll 7 & keep 3) or Dex+Larceny at 5k4 (roll 5 & keep 4) to open an electronic lock. But the TN is the same and the books spell out that one kept die is worth two rolled dice meaning that particular choice is basically a wash (the 20%/50%/70% numbers are within +/-1 point of each other for those rolls).

As long as the writers of the system can do the math and communicate he rules of thumb for the odds there shouldn't be any issue. If there is an issue because the writers failed either of those tasks (ref: D&D 4e first couple of skill challenge versions) then it likely won't matter what dice system gets used, people are having problems because of the writers not because of a dice mechanic.

Satinavian

2022-07-19, 02:21 AM

Players can understand larger and smaller pools without being able to quote the curve distributions, but can you pick the better roll out of 5d10TN7 and 7d10TN8 without a calculator?Of course. That is still pretty trivial and stuff people should have had in school.

I mean, when we still played shadowrun 3, it was pretty abvious most people at the table were aware of the expected value of sucesses even in fast paced combat.

It with some more complications, like "want at least 1 success but more successes are preferrable" or "as many successes as possible but no more than one 1" or similar, I could still do it without calculator, but it would be kinda tedious and not practical to do on the fly in game. And i know other players who are not good at math would not be able to do that even with calculators.

This pretty much, why should we always know the best method to get a desired result?
Because presumably the PCs know their own abilities and can, if they are not amateurs, chose the best way to solve a problem if there are meaningfull differences.

Interesting choices are not between using a good approach and a bad approach. Interesting choices might be risk vs reward, like in the Splittermond example above, where you could go for "higher average but higher fumble chance/normal test/ lower average without fumble chance". Or using consumables to get a bonus or try it without. Or having a pool you can distribute to several tasks.

These are actually interesting, engaging choices. Those, where the players just don't understand their chances and pick at random are not.

PhoenixPhyre

2022-07-19, 09:33 AM

Beware of false precision. The idea that you need to be able to precisely calculate the probability of success (especially when degrees of success are involved) to have agency/make meaningful decisions is, to me, very strange. Because in real life, where we seem to make meaningful decisions all the time[1], we not only don't know the exact probability of success of anything we do, there are reasons to believe that we can't know that probability beyond the very most high-level "X is more difficult than Y" sense. No, not even for simple operations you do all the time.

In fact, assigning probabilities to events (in the frequentist sense) can really only be done successfully for things without agency. You can define the probabilities of two subatomic particles interacting in particular ways (my field of research as a PhD student) with quite rigorous precision. You can't assign probabilities for two people interacting in particular ways with any kind of precision whatsoever. In fact, it's difficult if not impossible to chart out even a substantial section of the possible interaction pathways for any two arbitrary people. You can do slightly better for large groups of people interacting, but only slightly. Hari Seldon's psychohistory is not a real thing.

So in a game, all that's really needed is "Task A is harder than Task B" with "you can do that without enough risk to worry about[2]" or "that task is impossible for you" as third options. The whole bell curve/fine-grained detail thing? Yeah, it may be nice math, but it's not really modeling anything fundamental, especially at the scale of most campaigns.

[1] not getting into the whole free will debate here.
[2] which is not no risk, there's always risk. But the risk is low enough or not interesting enough that we're going to ignore it for game purposes.

LibraryOgre

2022-07-19, 01:21 PM

One simple one for combat and other situations where success is measured as the combination of many actions each with its own resolution check, basically either d20 or d100 plus simple and not often changing mods vs fairly straightforward TN.

One complex system, preferably bell curve based, that strongly reflected character(s) skill and lots of fiddly modifiers, for times when a whole host of player decisions is ending up being abstracted into a single roll for success. Also with degrees of failure and degrees of success.

Hackmaster does something like this in skill checks

Unopposed tests are d100% under your skill percentage; average tasks have a -40% on the roll, while difficult tasks are straight roll (there's more modifiers, but those give you some idea).
Opposed tests are d100+skill modifier v. d100+skill modifier of the opponent.

For example, if someone is sneaking past a relatively inattentive guard, the GM might call for an unopposed test, or they might call for an opposed test. Making a basket or asking for information is likely to be unopposed; interrogation or picking pockets is more likely to be opposed.

Satinavian

2022-07-20, 01:42 AM

So in a game, all that's really needed is "Task A is harder than Task B" with "you can do that without enough risk to worry about[2]" or "that task is impossible for you" as third options. The whole bell curve/fine-grained detail thing? Yeah, it may be nice math, but it's not really modeling anything fundamental, especially at the scale of most campaigns.
It is true that you don't need exact numbers and a rough extimate works quite well.

However that is even more reason to go for bell-shaped resolution distributions. Because with those you can plan around likely results being roughly in the middle and extreme cases being rare, so the average already gives you a good idea what to expect.
Furthermore the main disadvantage of the bell shaped resolutions is that some people find those to complicated to calculate in the head, but if you don't need the excat numbers, that drwback disappears.

Ignimortis

2022-07-20, 02:07 AM

Telok

2022-07-20, 11:48 AM

Dicepools. Single die systems, in my experience, either turn everything into a circus if you cannot outscale the die with bonuses to the point where rolling the minimum value is still some sort of success, or, as consequence of letting you scale, lock people out of trying even simple tasks eventually.

Meanwhile dicepools, usually constructed of things like stat+skill or stat+stat or skill+skill, tend to perform somewhat more reliably. Take a pile of d6, put the TN at 4, count up 1s for the "uh oh" factor.

While speed of resolution is important, it's 2022. Dice rollers exist and mitigate the most issues with dicepools, while they will never fix issues with single die distribution.

I feel ya. I track some stats on my current D&D character and... last two sessions had six combats totaling 25 turns (not rounds, just my character's turns) of which 18 involved eldrich blast x3, of those 50+ d20 rolls there was one 20. Chap sitting next to me plays a barbarian, missed one session and spent a 5 round combat perma stunned (int save, natch plus missed a turn last fight from a dropped haste spell), got three 20s in two combats of not always raging and only being hasted for about 3 turns.

Its why I like the flat d20 or d100 for comedy & horror. In Paranoia you play hot potato with a grenade because its supposed to chunky salsa someone and the flat math is fast & lol-random no matter how good or bad your character is supposed to be. CoC your 90% dodge & axe skills aren't supposed to let you tank a dozen ghouls even though you're a super-human axe ninja at that rank.

Then you get D&D with Thog the 20th level fighter wearing a belt of giant strength getting pantsed by Bo-Bo the geriatric poo flinging gibbon because d20+13 vs d20-2 gives you 1-3 vs 19-20 = the poo flinger is your world champion wrestler today. Sure its usually a "rare event", but D&D has so many rolls it basically comes up every sesson or two. Makes for great comedy, but not very heroic when some old monkeys take down your "extra strong" fighter.

Ignimortis

2022-07-20, 02:17 PM

Its why I like the flat d20 or d100 for comedy & horror. In Paranoia you play hot potato with a grenade because its supposed to chunky salsa someone and the flat math is fast & lol-random no matter how good or bad your character is supposed to be. CoC your 90% dodge & axe skills aren't supposed to let you tank a dozen ghouls even though you're a super-human axe ninja at that rank.

Then you get D&D with Thog the 20th level fighter wearing a belt of giant strength getting pantsed by Bo-Bo the geriatric poo flinging gibbon because d20+13 vs d20-2 gives you 1-3 vs 19-20 = the poo flinger is your world champion wrestler today.

Pretty much, yes. If you want consistent competence (and I tend to play systems where that suits the mood best), dicepools work best. If you want to always be unsure of the roll for either comedy or suspense, single-die might be a lot more conductive to getting the right atmosphere.

PhoenixPhyre

2022-07-20, 02:42 PM

See, I don't notice that. But that's, I think, because I treat the resolution system as only applying when there is meaningful uncertainty about the outcome. I'd never expect a resolution system to handle everything--if the fiction dictates that only one option (or only a small set of options) make sense, I'm not going to ask for a random result at all. Because there isn't uncertainty.

Resolution systems work best when they focus on neutrally picking between acceptable results. If some of the possibly chosen results aren't acceptable fictionally, they should be filtered out before you even get to the mechanical rolling dice level. Dice should be neutral, people don't have to be.

But then I'm very accepting (and welcoming, in fact) of active DM involvement at every stage, including task resolution. What I don't want is the rules demanding that they have priority over the people.

Notafish

2022-07-20, 02:49 PM

For generic TTRPG resolution, I like "roll the Target Number or higher" with either a d20 or 3d6, depending on how swingy you want things to be. It's easy to teach and to comprehend (especially if D&D3e+ was your introduction to the hobby), and can be spiced up with crits, fumbles, and degrees of success without needing to introduce a completely new system. When I'm playing or running an RPG, I usually don't want to think too hard about dice probabilities, so familiarity and ease of use is nice for me.

However, my favorite resolution system is Blood Bowl block dice, which is sort of like a dice pool, but with asymmetric dice. Blood Bowl is not really a TTRPG - it's a Games Workshop game that asks, "what if the NFL, but with orks?" The dice determine the outcome of an attempt to hit or throw a block on an opposing character. Depending on match-up (which character is stronger/has more support from teammates), either the aggressor or the defender will choose one die from a pool of between 1 and 3 six-sided dice. In general, the dice have a 1/6 chance of displaying an outcome that is bad for the aggressor, 3/6 chance of a neutral outcome, and a 2/6 chance of the aggressor succeeding. However, character traits can modify the way these outcomes are adjudicated (for example, a character with the "Frenzy" trait treats a roll of 3 or 4 as "push the defender and roll again" while other characters will only push the defender and end their turn). It's a very focused system that would take a lot of work to adapt to other settings, but it works very well in the game for which it was designed, and the rules interactions complicate the game in a lot of fun ways.

Telok

2022-07-21, 12:24 AM

See, I don't notice that. But that's, I think, because I treat the resolution system as only applying when there is meaningful uncertainty about the outcome. I'd never expect a resolution system to handle everything--if the fiction dictates that only one option (or only a small set of options) make sense, I'm not going to ask for a random result at all. Because there isn't uncertainty.

Really in D&D its only mostly noticable when you track actual data instead of relying on memory & chatter, or with less experienced DMs who think the rules should be used. Having played with DMs who truely had bad math (Mr. "three 30% chances equal one 90% certainty" was a memorable one), didn't know/think they knew better than the devs, or aren't sure enough of their storytelling ability to ignore the rules, its pretty ugly when they trust the game to just work and it keeps giving them jank.

Personally tho, if I'm running something that I need to regularly override the rules to keep it working I'll start wondering why I'm not using something less failure prone. Maybe something light & fast to the point of infalliability like Lasers & Feelings or such. Of course if I could find enough Zelazny fans in one place I'd have to shell out for a copy of Amber Diceless too.

Firest Kathon

2022-07-21, 09:45 AM

The German RPG The Dark Eye handles things that way - combat actions are solved with a realtively simple, D20-roll under system, while skill tests are used with 3 d20, compared to three stats. However, last time I checked, that was a big, very cumbersome system with a lot of moving parts, options and modifiers.

The 5th version of The Dark Eye simplifies things a lot. Rules are available or free in German (https://www.ulisses-regelwiki.de/start.html) and in English (https://www.ulisses-regelwiki.de/home.html). The resolution rules are available here (https://www.ulisses-regelwiki.de/checks.html). Abbreviated: Attributes range usually from 5-18 (more extreme values are possible), you roll 3d20 under three attributes. Ability points are used to compensate for rolling over the target number. 0 or more points remaining mean success, number of remaining points determines the quality if needed.

What I really like about that system is that each check depends on (up to) three attributes. So even if my character is not particularly strong, I could still be good a athletics checks because they also depend on Courage and Dexterity in addition to Strength. And you can easily replace some attributes to run e.g. a check for knowledge about athletics with Int, Int, Sag instead.

PhoenixPhyre

2022-07-21, 09:55 AM

Really in D&D its only mostly noticable when you track actual data instead of relying on memory & chatter, or with less experienced DMs who think the rules should be used. Having played with DMs who truely had bad math (Mr. "three 30% chances equal one 90% certainty" was a memorable one), didn't know/think they knew better than the devs, or aren't sure enough of their storytelling ability to ignore the rules, its pretty ugly when they trust the game to just work and it keeps giving them jank.

Personally tho, if I'm running something that I need to regularly override the rules to keep it working I'll start wondering why I'm not using something less failure prone. Maybe something light & fast to the point of infalliability like Lasers & Feelings or such. Of course if I could find enough Zelazny fans in one place I'd have to shell out for a copy of Amber Diceless too.

It's not overriding the rules to use them exactly when they say they're supposed to be used and not when they're not. And those DMs? Weren't actually using the rules. Because the rules explicitly point out that you shouldn't ask them to do things they weren't designed for. Which is what they were doing in rolling for everything.

Personally, I don't see rules as some kind of simulation engine (given the current state, what is the next state). Rules are there as tools to be asked specific questions smaller than the totality of the next state given a state of equilibrium (of options A and B, both of which are acceptable, which one happens?). So I'm really only interested in the places of the probability curve where things are basically flat anyway--the ends of the curve aren't the job of the rules at all. Rules are not the masters, where overriding them is a special thing. They're passive tools that wait until they're called on. And knowing when not to call on rules is just as important.

If I wanted to play a game where the rules were in the driving seat, well, there are lots of board games and computer games that scratch that itch. I play TTRPGs for the human element, for the human to be in the drivers' seat. For me, free-form is the default and ideal, and rules are there to help lessen the load of free-form (at the cost of narrowing the field when they're used).

Telok

2022-07-21, 10:56 AM

It's not overriding the rules to use them exactly when they say they're supposed to be used and not when they're not. And those DMs? Weren't actually using the rules. Because the rules explicitly point out that you shouldn't ask them to do things they weren't designed for. Which is what they were doing in rolling for everything.

See, you keep saying that all the 5e DMs I've ever seen are playing the game wrong by trying to follow the rules as they understand them. Either you or they are: 1. Reading different sets of books, or 2. Getting totally different messages from them. Although, now that I think about it, you could be referencing information sources & advice that they don't (they just stuck to the books without hitting up the net & non-WotC publications), and you certainly seem to have a better intuitive grasp of iterative probability calculations.

But either way it doesn't change my perception or experience that dice systems of basically straight dice with flat probabilities and binary results are best suited to comedy & horror genera based on the style of outcomes where the randomness of the die overpowers player choices. That informs my preference for bell curve result dice systems for games ouside of comedy, horror, & ultralight. Your comment isn't about the flat d20 plus assorted other random effect dice being a better or faster dice system, but about people not playing D&D like you want them to.

Easy e

2022-07-22, 12:19 PM

I am a big fan of dice pools, but especially depleting dice pools, where you have to decide how many to use at X time, or be caught out without enough dice in a pool later.

Plus, they help make the game more "success" focused and less failure prone.

Marcloure

2022-07-30, 12:24 AM

I really like Ironsworn's dice system: you roll d6+stat and compare with 2d10. If your d6+stat roll is above both d10 results, it's a success. Between the results is a partial success, below them is a failure. Lastly, if both d10 rolled the same number, then that action is a critical or fumble depending on whether you succeed or fail.

Tanarii

2022-07-30, 11:47 AM

See, you keep saying that all the 5e DMs I've ever seen are playing the game wrong by trying to follow the rules as they understand them. Either you or they are: 1. Reading different sets of books, or 2. Getting totally different messages from them. Although, now that I think about it, you could be referencing information sources & advice that they don't (they just stuck to the books without hitting up the net & non-WotC publications), and you certainly seem to have a better intuitive grasp of iterative probability calculations.
My experience has been is many if not most 5e DMs don't read Chapter 8 of the DMG, which very clearly states you shouldn't roll for everything right at the beginning of the section on ability checks, that you should ask if something cannot fail or succeed before calling for a roll.

As far as I'm concerned that chapter is required reading for anyone wanting to be a 5e DM and should have been Chapter 1. Because I've experienced far too many DMs trying to run the game 3e style, roll for everything even when they think it should automatically succeed or fail. Especially in the early days of 5e.

Telok

2022-07-30, 07:17 PM

My experience has been is many if not most 5e DMs don't read Chapter 8 of the DMG, which very clearly states you shouldn't roll for everything right at the beginning of the section on ability checks, that you should ask if something cannot fail or succeed before calling for a roll.

As far as I'm concerned that chapter is required reading for anyone wanting to be a 5e DM and should have been Chapter 1. Because I've experienced far too many DMs trying to run the game 3e style, roll for everything even when they think it should automatically succeed or fail. Especially in the early days of 5e.

I'm sure that's nice for DMs who know they want PCs to autosucceed or autofail at something. The newer D&D 5e DMs I've seen have read the whole DMG and have not DMed D&D 3.x or 4e. Its not that they can't read, but that they don't know if they should fiat the PCs succeeding or failing practically everything out of combat. After all, rolling for everything and following DCs printed in the adventures works great in combat so it should be fine out of combat too, right? Or are they getting a different message from what they read than you did?

I think D&D 5e is really two different games in & out of combat. But the books don't really seem to say that and the big, high visibility, character abilities like spells & checks work the same on both sides. So the DMs get big chunks of time, rules, and mind space given over to the "roll for everything each time" combat game, then they try to use that experience & knowledge in the rest of the game. But that's not about the resolution or dice system, its more about the game having two modes that play super differently and not reaaly spelling it out in detail.