PDA

View Full Version : Critique my dice mechanic!



Feddlefew
2017-04-04, 03:08 AM
Edit: Now on version 0.2! (http://www.giantitp.com/forums/showsinglepost.php?p=21887096&postcount=7)

Background: I'm making a quick and dirty paranormal espionage RPG for a community game-jam like... thing...

I looked at a lot of different dice systems over the past few days, and I really didn't find one that fit my needs. So I hashed together this one.

Goals:
1) Whatever dice mechanic is used needs to scale decently as characters grow in power.
2) Experts should be more reliable and use less resources than novices.
3) This will probably be the first Table Top RPG for a lot of people, so the basic mechanics need to be easy to get the hang of.
4) Mechanics need to capture the tension of having limited resources, and encourage players to manage risks.
5) The game should be play-by-post friendly, so rolling a lot of dice once and then manipulating the results is better than making a series of rolls.

What I've come up with is a hybrid system that combines aspects of a few different mechanics. The previous sentence may be setting off alarm bells for some of you, which is completely reasonable. I've thought about this for awhile and I think it works out okay mathwise. The basic idea is as follows:

At the start of a mission/event, each character has a largish pool of dice they can draw on. The dice pool represents abstract stuff like time, material resources, and the character's remaining mental and physical stamina. When your character needs to make a check, you do the following:

1) Pick a number of dice from the appropriate pool(s) to wager.
2) Discard dice which roll over your skill. Skills range from 1 (worse than an average novice) to 9 (pinnacle of human achievement).
3) Total remaining dice. If the total exceeds the the target number, success!

Exceeding the target number by a certain amount may have additional effects, depending on what the roll was for. However, all dice wagered are removed from the character's pool. Run out of dice, and bad stuff(tm) happens, but will be some limited ways to regain dice.

Things which I need to fine tune:
-What size dice to use? (I'm probably going to go with D10s unless the community insists on using D6s.)
-How do you determine initial pool size?
-Does supernatural power get its own pool?
-How do we treat resources like wealth, or political connections?
-Are there limiting factors on how many dice a person can use for one roll besides the number of dice remaining in their pool(s)?
-Are there situations which allow dice to be rerolled before discarding or totaling?

Gray Mage
2017-04-04, 10:53 AM
I think I'd only remove from the pool the dice that rolls over the skill, as removing all seems too punitive. By only removing the discarded ones you ensure that they can keep doing what their character is good at and that the initial pool doesn't need to be too big. Even if someone has max points that should happen somewhat often. I feel initial pool size should also related with skill.

If skills got to 9 I'd say a d10 is a good size. Minimum 10% to lose from the pool is good, as that's the best of the best.

Also, I'd maybe make it so that instead of not considering discarded dice, it simply means that it's limited by the character's skill. So if someone has a skill of 4 and rolls a 6 it counts as a 4. That way someone with less skill may be able to do something big, but might need to empty almost all his pool so it's a one time thing and if someone is really, really bad (say 3 or 4 out of 10) he doesn't need to fear rolling and getting a bunch of zeroes.

For finite resources like favours and wealth, maybe you could have a separate, smaller, pool that has a number of automatic sucesses (maybe with the skill rating or fraction of it, as your discretion), but this resource doesn't automatically refresh, instead it's gained by quest rewards/actions and it's always discarded when using.

erikun
2017-04-04, 12:58 PM
I've not run into a system such as this, so I can't really say how much it would work. I do notice a few similarities to existing systems, which I can point out.

I would also note that the best answer to "How well will this work" is to simply give it a try and see what works and what doesn't. We could theorycraft all day, and arguably get some progress in doing so, but you will ultimately want to give the system a try to see what works.


Your limited dice pool per day reminds me of the Warriors Adventure Game. (http://www.warriorcats.com/games-and-extras/games/adventure-game) (based off the Warriors cat books by Erin Hunter) In this system, a player is given chips equal to each of their three main abilities. Normally you just compare Skill + Ability to determine success, but you can spend a chip of the appropriate Ability in order to gain a +1 bonus. You can also spend chips on some Knacks, or special abilities. As such, the majority of the game does not involve spending these limited resources: It is assumed that most challenges will either be overcome or find another way around them. Spending the chips is reserved for important decisions or doing something special.

One other aspect this system has is that the adventures need to be fairly well structured. If you look at the practice mission from above, you'll note that it is fairly linear (frequent split paths, but you just overcome the challenge or take a different route) and there are scripted rest points for recovering chips. The chip recovery method makes it a lot harder to just throw the players at a random scenario and hope they do well; there needs to be some method, designed into maps (until players can freely transport wherever they way) to allow for it to happen.


Roll-under and then total is not a system I've specifically seen before. Most games use a roll-under (or roll-over) and then count the number of successes. Burning Wheel, World of Darkness, and so on. The biggest concern with the roll-under idea you have is that it can result in some wildly different values. A character, even a very skilled one, can end up rolling 5, or even 0, on their collective dice rolls. That same character can end up rolling 25 or even 40 with the same dice setup. The result is a massively swingy system, and one which I think would be hard to nail down the probability of certain events happening. Some tricky dice manipulation in anydice.com (http://anydice.com/) might be able to show off just what the odds are, but they certainly do not seem intuitive.

I will note that wagering large amounts of dice and then discarding them will end up becoming quite costly. The above example required skill level 8 and five dice, for a .001% chance of getting that 40 value. A player would need to toss away far more dice to reliably hit that high of a number, which means that either you want your target values far lower or be far more generous with your system. In fact, my biggest concern is what the balance point with this idea would be: exactly how many dice can you hand out to allow players to reasonably hit those high values, while not providing so many that they could just breeze through hundreds of lesser challenges?



3) This will probably be the first Table Top RPG for a lot of people, so the basic mechanics need to be easy to get the hang of.
4) Mechanics need to capture the tension of having limited resources, and encourage players to manage risks.
This is a tricky one, and one of the big concerns I would have. People who are not familiar with tabletop roleplaying at all are not going to be familiar with RPG game systems or how mechanics work. They are not necessarily going to know how to assess risks or manage resources, because that is fairly strange in many cases. As such, the best RPG systems for brand new players are likely going to be ones where the GM handles the majority of the legwork and just interacts with players on a basic concept level.

Unfortunately, that conflicts with the idea of limiting player resoruces with the amount of dice and rolls they can make. If you try to manage a system like that by the GM, then you start running into situations where the GM tells the players they can't do things because "not enough game resources", which is going to sound meaningless and arbritrary. But if you dump this information onto brand new players themselves, then they aren't going to understand what's going on and won't be able to manage the resources in a meaningful way.

Now, you can have a system which allow the (in-character, roleplay) resources to be managed and handled easily. If the system has an easy way to keep track of bullets, food, stamina, etc. then it would work as a specialty resource management system: players will understand that they are low on bullets, or exhausted, or whatever. But they won't necessarily know that they have 12 dice left in their dice polls and that going "all out" is not going to allow them to drive a car afterwards, because of how the dice pools work.


I think you might need to make some priorities between desires, in case some situations like this come up. Is the limited dice pool and roll-under-and-total essential, to where you absolutely want to use those? They you'd want to take a look at probabilities and distribution and go through a lot of trials, in order to see what number ranges are reasonable and what the system can represent. Do you want to make it open and accessable to new players, possibly a large number of different people new to RPGs? Then you'd want to focus on a system which can easily be run when only one person at the table understand it, regardless of the mechanics used.

Right now, it sounds like you just have a handful of ideas and goals, without much priority in how to accomplish them.

Airk
2017-04-05, 04:13 PM
This is just a sideways twist on the dice system in Lady Blackbird and its forebears. Specifically, to quote that game:



Rolling the Dice
When you try to overcome an obstacle, you roll dice. Start with one die. Add a die if you have a trait that can help you. If that trait has any tags that apply, add another die for each tag. Finally, add any number of dice from your personal pool of dice (your pool starts with 7 dice).
Roll all the dice you’ve gathered. Each die that shows 4 or higher is a hit.
You need hits equal to the difficulty level (usually 3) to pass the obstacle.

If you pass, discard all the dice you rolled (including any pool dice you used). Don’t worry, you can get your pool dice back.
If you don’t pass, you don’t yet achieve your goal. But, you get to keep the pool dice you rolled and add another die to your pool. The GM will escalate the situation in some way and you might be able to try again.


There's some rearrangement of stuff here, but fundamentally, the Lady Blackbird system is simpler and achieves the same things as what you want, I think. I find the whole "Pick some random number of dice that will then maybe get discarded, and total up the survivors" to be substantially more complicated and tougher to grasp, and it doesn't seem to have any serious advantages - it's actually doubly punitive to people with low skill, because they will both be discarding more dice AND the dice they are allowed to keep will have lower values. Additionally, the random spread in your game is very large for skilled characters, because while they will keep more dice, those dice are extremely unreliable in terms of value.

Also, depending on the target number range you settle on, you could easily get into the territory where people will be slowing down for arithmetic on the "totaling" phase.

Cluedrew
2017-04-05, 04:30 PM
Things which I need to fine tune:My opinion on each.

-What size dice to use? (I'm probably going to go with D10s unless the community insists on using D6s.)
As long as you have enough it probably doesn't matter. Maybe try d10s and if you find you aren't using the whole range (or just that a skill 5 doesn't feel different from a 4) go to d6.

-How do you determine initial pool size?
I would say keep it fixed, or fixed for each type of pool. You already have one axis to represent ability, I don't see why you should add enough in the simple game you are aiming for.

-Does supernatural power get its own pool?
-How do we treat resources like wealth, or political connections?
Fold both into the main, pool, as they both seem to be part of the existing generic pool.

-Are there limiting factors on how many dice a person can use for one roll besides the number of dice remaining in their pool(s)?
Yes. In fact I would almost lay out a few levels of resources that you can invest. Because 1-N dice is to many decision points however 3 (a brief attempt, a serious attempt, a committed attempt) with some narrative description makes things much clearer by reducing the complexity of the problem and putting it in more concrete terms.

Knaight
2017-04-05, 04:32 PM
This is just a sideways twist on the dice system in Lady Blackbird and its forebears. Specifically, to quote that game:

Not quite. Here, you always drain from your pool, it isn't a reserve for hard rolls. That sets up a situation where the way skill is determined is in making draining from the pool efficient at what you're good at because it gets more results.


I find the whole "Pick some random number of dice that will then maybe get discarded, and total up the survivors" to be substantially more complicated and tougher to grasp, and it doesn't seem to have any serious advantages - it's actually doubly punitive to people with low skill, because they will both be discarding more dice AND the dice they are allowed to keep will have lower values. Additionally, the random spread in your game is very large for skilled characters, because while they will keep more dice, those dice are extremely unreliable in terms of value.
It's not random though, it's a calculated decision. That's not a problem. The random spread for skilled characters is where this gets much more iffy, and I probably would go with a more conventional dicepool system (where rolling under the skill could be maintained as a success, thus altering the average ratio of successes:dice).

Feddlefew
2017-04-05, 11:05 PM
I've had some time to think about what you guys have said. I also talked to a few of my fellow game jam-thing participants.

General concerns:

-I agree that the swingyness is pretty bad, and conflicts with the goal of having experts do constantly better than novices. I've been playing around with AnyDice and some other statistical software, and I think I've come up with a few major tweaks that will make the system easier to use while still preserving the overall feel.

-I did lock down the dice size at d10.

-I'm aiming for a more RP-heavy game, so needing to roll will happen less often than I think some of you are expecting. Instead of having to roll multiple times to sneak through a facility as you reach each guard or whatever, there would be one moderately difficult roll. You wouldn't need to make another stealth check if you, say, came across a locked door you needed to get past, but if the alarm was raised or you got into a fight you might need to make another, higher difficulty check.

-There will be ways to regain dice. The most popular one so far has been allowing players to add a complication to the mission to regain dice, like a fire breaking out or a new objective popping up.

-I'm using the depletion of the dice pool as the main failure state instead of character death, so completing the mission but not being able to maker your getaway is a reasonable outcome. There's still other ways your character can get disabled, like if they get beaten up, but that would be a temporary set back instead of a mission (or game ending) situation. A character that runs out of dice mid-mission can be helped by other characters, and any event which gives all characters more dice would still give that character more dice.

Specific concerns:

@GrayMage:
-As much as I like the idea of players being able to keep successful dice, I'm using the wagering system to discourage everyone, including experts, from throwing a huge number dice at problems to ensure success. I think I'm probably going to cap the number of dice that can be rolled at once.

@Erikun:
-The main reason I haven't prioritize my goals is because I am still very early in the design phase, and a lot is still dependent on community feedback once the project formally announced. For instance, we haven't actually decided if characters are going to have hit points, or how combat will be handled.

- For now I'm sticking with roll-under for now because I think it's the most intuitive way to handle skill level in a system where you can pick the number of dice you roll. I'm also trying to avoid players having to stop and look up what modifiers they need to add every time they roll, which in my experience has been the hardest thing for new players to handle. I also just happen to like the idea of a blackjack-style system.

- I think I will go with having wealth and political resources use a completely separate, possibly point based, system since they'd be used between missions.

@Airk & @Knaight:
-Knaight's got the right idea about how the system is supposed to work.
-However, Airk is right about the totaling phase taking too much time.

@Cluedrow
-I'm thinking mission difficulty might determine pool size. Considering what the pool represents, I agree that I probably should lump paranormal powers and mundane ability into the same pool. How paranormal abilities are handled is going to be decided at a later time.
-I think I'm going to cap the number of dice that can be rolled at once at 10, baring the use of special abilities or equipment. I do like the idea of laying out a few different levels of dice wagering for the players, even if just to provide context for roleplaying.

Tweaks:

For the new version I'm moving to a more traditional success system. Rolling under your skill counts as one success, and getting at least the target number of successes means that you succeed at your task. However, to determine the magnitude of success you total the n highest successful dice, where (for now) n=2 under normal circumstances. Since I dislike critical failure, failing just means you've A) wasted dice and B) you have to deal with the logical consequences of failing that check.

The good thing about doing it this way is that it's pretty easy to get a feel for how many dice you need to roll. Experts will also consistently do better at novices, and magnitude of results directly correlates to both skill level and how much is invested into the roll.

I've also decided that, for now, we can consider a normal character to be a novice at most skills, which means that their skill level starts at three. I guess the best comparison is skills in D&D which can be used untrained vs. skills that can't be, although this system let's you try anyway.

Borrowing a little from Dwarf fortress's skill system, the skill levels are:
1) Terrible- You have been handicapped or taken a flaw at character creation. Basically unusable.
2) Dabbling- This an entirely novel situation; you have no idea what to do but are going to try anyway!
3) Novice- You have an idea of what to do, but haven't really put any practice in.
4) Adequate -(description pending)
5) Skilled- (description pending)
6) Professional- (description pending)
7) Expert- Few tasks are difficult for you.
8) Master- The highest most characters can have; tasks which would be difficult for others are trivial.
9) Legendary- The pinnacle of human achievement.

The new rolling protocol:
1) Pick a number of dice to wager.
2) Roll dice. Discard dice that roll over your skill.
3) Compare the number of the remaining dice to the target number. If you meet or exceed the target, you are succeed at the task.
4) If you succeed, add the two highest remaining dice together- this is the degree of your success.

What degree of success means in this case is up for debate, but generally I think it would mean getting additional benefits out of a roll, such as reducing the difficulty if you need to repeat the task later, making something of a higher quality, or just plain doing or getting more of something.