Percentile Based System Help\Question

[Zakharov]

Hey all. Long time reader, first time poster.

I've been working on a little, 'relatively' easy rpg system to play around the house as an introduction to the big boys. I have a large family, and as each family member has their own interests and friends, there's obviously a need to be simple to try to bring them in. Plus, I'm going to be heading into college soon, so I figured it would be something I could use later on, as well.

Basically, the system is based off of percentiles, or 1d100 (2d10). Player's have skills, abilities, and attributes. All of these are based off of percentiles. For instance, let's take Yuri the super spy:

Yuri's Fire Arms: Pistols skill is at 68%. The GM would calculate a difficulty modifier (GM: The agent you are trying to hit is hiding behind a crate, so I'll give you a 20% difficulty.) Doing the math, you would now have a (68% - 20% = 48%) 48% chance to succeed with your skill. Pretty simple, no?

Skills and Attributes go from 0 - 100 (or over, in some cases) and Abilities are "special powers" and are custom tailored at character creation.

The problem I'm having is with the Attributes. Now, the way I have it set up, is that all characters (NPCs and PCs) start with a 5 in every attribute. This is considered the norm, lower then 5 and you're considered weaker then your average human, the higher, the better. This increases both at character creation (which uses a point based system) and over the span of the game. The problem, however, is how to do what's normally called an Attribute Check. If I try to do it the same way as Skills, you end up with something like this:

Yuri has a perception attribute of 15 . The GM decides to have Yuri make a perception roll to spot someone out from a crowd. If the GM gives the player a difficulty of 20%, we now have a problem. The way skills work, you subtract the difficulty from the skill... subtracting 20 from 15 would give us -5. This could be considered an automatic failure for some people. I was thinking about putting a system in place of a Lowest Roll Possible kinda dealie, such as if you end up with a negative difficulty, you have to roll on a difficulty of 10%.

Let's try it the other way. If you subtract the attribute from the difficulty modifier, as I originally had thought of using when thinking of this so called 'system', you get an undesireable result that only works some of the time.

Let's return to Yuri's perception of 15 and the difficulty modifier of 20%. Using this method, we would get a result of (20% - 15 = 5%) 5% difficulty, which gets the desired result. However, the problem comes when we have higher attributes. The way it works now, you would actually be penalized for having a higher attribute.

If Yuri's perception was two points higher to make it 17, the result would now be (20% - 17 = 3%). Why would someone who obviously had a better sense of perception have a harder time making the roll?

And that's the pickle I'm in. I'll take any and all suggestions, but remember, I'm aiming for ease of play. I'd like to try to keep the percentile based system in, but I suppose if I have too, I can ditch the attribute system with percentiles for something better.

Thanks for the help!

[Jack Aidley]

Two thoughts:

1. Why are the stats so low, why not have higher stats?
2. Why is being impossible an undesirable result?

[Zakharov]

#1. I was using low stats for an example. A character, after creation, usually averages about 15 on all his stats at first level, and they increase each level as the character sees fit. Perhaps you are right, though, in that I should give them more starting points to fool around with, after all, if the scores averaged more around 30, there would be more room for leeway.

#2. It really depends on GMs. For me, I'm about 50/50. I dislike using "You just can't do that, you're not strong/smart/perceptive enough" as an excuse. I think there should always be a small percentage. But, I do understand that somethings should be an impossibility. If there was no impossibility, then a character with a 1 in strength could still have a 10% chance to move a mountain. Kinda silly. I should have said "A undesireable result if used constantly to prevent players from attempting something."

[MarktheAnimator]

Hello,
I use percentile dice in my game, Fantasy Imperium.

You can download a copy here:
http://www.fantasyimperium.com/playtesters

Here's how it works:

I have 10 stats (strength, endurance, dexterity, intuition, self discipline, reasoning, ego, awareness, presence, attractiveness).
Each stat is 1-100%, the higher the better.

To make a stat check, all you have to do is roll under your stat on percentile dice.  If it is difficult, you roll under half your stat.  If it's very difficult, you roll under one quarter of your stat.  If it's nearly impossible, you roll under one tenth of your stat.

There can also be situational mondiers determined by the Storyteller (plus or minus to the rolls).

If you are competing against another person, you make a contest roll.
This is done by rolling percentile dice and adding your characteristic.  The higher roll wins.  For instance, if two ppl are arm wrestling, they roll percentile and add their strength.  The higher roll wins.

Skills are all percentile based too, and they work just like characteristic rolls.  So if you want to climb a tree, all you have to do is roll under your climing skill on percentile dice.

If you have no skill in something, there is still a chance of success.  Skills are all based on your characteristics.  
So all skills get a skill base percentage.  
This is equal to the characteristic it is based on divided by 5.  
So if you wanted to use try juggling, but had no skill, you would still have a chance of your dex divided by 5 (1-20%).  When you acquire skills, you simply add the percentage to the skill base, which is done only once.

There is free access to skills.  So anyone can learn anything.  However, there is a limit to how many skill points you can spend on a single skill while making a character.  If you want to develop your skill higher when making a character, you need to be in a profession that practices the skill.

So I have a list of professions you can choose from.  Each profession has a list of what skills you can concentrate on.
If you want to practice more than one profession, you simply pay some extra skill points.  
So you can get into any profession you desire, and can have many professions.  
It all depends on how you want to spend your skill points.

If you want to increase a skill or a characteristic, all you do is take an experience point and spend it on a roll.  To succeed, all you have to do is roll percentile dice over your current skill /cheracteristic.  The higher the skill is, the harder it is to increase.  If you succeed in the roll, you can increase it by 1d6%.  If you fail, you lose the x.p.

Simple and elegant.  It works quite well.  If you want to know how good your character is, you just look at his skills.  
So instead of "I'm a third level fighter," you would say, "I have a 74% chance to hit someone with two handed swords, etc.


So in your example, if you wanted to make skill checks or characteristic rolls difficult, simply cut the number in half, or one quarter.  

Instead of subtracting a percentile from a check, why not convert to a scale from 1-20?  Just divide the percentile penalty by 5.
So in your example, if you have a penalty of -20%, subtract 4 from the roll.
If you have a perception of 15, then the chance with a 20% penalty would be 11.  (20/5=4, 15-4=11).
Thats how I'd do it. :)


I hope this gives you some ideas. :)

[TonyLB]

I like the notion that Attributes offset difficulty modifiers.

If they offset each difficulty modifier then you can get a wonderfully cinematic system, I think.

[*]Firing two guns: -10%[*]Sliding down a bannister while firing: -10%[*]Firing while under fire by dozens of mooks: -10%[*]Total penalties for someone with Style 0: 30%[*]Total penalties for someone with Style 10: None.[/list:u]Particularly fun (IMHO) if the penalties come in pairs... one for you, one for your opposition.  Sure, sliding down the bannister is -10% for you to hit the mooks and only -5% for the mooks to hit you... but your attributes are high and theirs are low, so it's a relatively good deal.

And that's what I've got to say about that.  Sounds fun!  Tell us more!

[Zakharov]

MarktheAnimator: That's actually pretty interesting, I never thought of doing it that way. The way my skills work is the same way as far as a success being a roll under the skill percentage, but the 1/2, 1/4 and 1/10 thing is different. I'll look into it, thanks for the ideas! I'm also checking out your game as we speak. :)

TonyLB: Yeah, I do too, that's why I'm trying to use that. For your example though, are you saying that since your character would have a 10%, all -10%'s would be offset? Would a -20% become a -10%? That would be pretty interesting. Also, it would be fun if they did come in pairs. I'll have to do some testing for that. Thanks for the ideas!

You requested more info, so here it is:

I pretty much explained how skills work, though there is some things I've done to make it kind of interesting. You have broad skills, specialized skills and narrow skills. Imagine it like a prequisite tree. You only really need broad and specialized, but narrow skills can come in handy. For instance, if you had a hacker character, a skill might look like this:

Computers (60%): Hacking (5%): Linux (10%)

Computers is a broad skill. So any roll your character had to roll for that related to computers would be 60%. This is to simulate the fact of 'general  knowledge'. Someone who's a hacker will have an easier time figuring out something new involving computers then someone who hasn't been using them for very long. But, if it involves hacking, he would get the +5% bonus (becoming 65%) and if he was hacking Linux, then it would become (60% + 5% + 10% = 75%)

The way characters are created is a point based system. They have a certain number of points + 1d100 to spend. Originally, it was 300, but I may be upping it to 500. The reason you need so many is because it works off of 'Costs to purchase' for each attribute. For instance, a Human would have a cost of 5 for all attributes. To raise his Strength by 10, he would have to spend (5 * 10= 50) 50 points. As I said, all characters start with 5's in the attributes, so after purchasing 10 points of strength, the character would effectively have 15. I might have to do some changes as far as that goes,  as Jack Aidley pointed out.

A Barbarian (as a race) has a lower Strength cost then a Human. It only costs 3 points per Strength to purchase. So to increase their Strength by 10, as a Barbarian you would only spec (3 * 10 = 30) 30 points. I used this system to promote off shoot characters. A Barbarian could be a mage of a sort, sure, but he would still have higher physical attributes then most mages.

------ ------ ------

I don't know though, I would like to get some more ideas on how to resolve the Attribute Check problem (no offense to MarktheAnimator, but the more the merrier!). Give me your best shot!

[Zoetrope10]

Dick Brown, in Space Gamer Fantasy Gamer #7 (1993, p. 68) offered the following, quite simple, %-based task resolution system. Sounds like it could meet your requirement.

Apply the relevant task difficulty modifier to the player's roll:

Simple: no adjustment
Difficult: +20
Tasking: +40
Limit: +80

A result less than or equal to the player's % rating is a success.
A result exceeding the % rating by up to 20 is a mixed result.
A result exceeding the % rating by 21 to 60 is a failure.
A result 60 or more above the % rating is a mishap.

A mixed result is a success, but with an additional, unwanted outcome. A mishap is a failure, with an additional, unwanted outcome.

A roll of 1 always succeeds; a roll of 100 always fails.

In your first example, Yuri has a "Fire Arms: Pistols" skill of 68%. Shooting at an agent behind a crate is a difficult task (+20). Yuri rolls the dice and adds 20 to the result. If the result is less than or equal to 68, he succeeds.

In your second example, Yuri (perception 15) tries to spot someone in a crowd, also a difficult task (+20). Unless Yuri rolls a 1, the best he will be able to get is a mixed result.

René

[Zakharov]

That's a really neat way of approaching the problem. I've always liked the idea of variable levels of success. Also, it makes having lower attributes not really a problem. Thanks for the post!

[MarktheAnimator]

Offended? Me?

I like ideas too! :)

So do you need to have a prerequisite to get the more detailed levels of skills?  (you have to buy computer in order to buy hacker?).

Specialties are an interesting idea.  I've seen something like it in HERO system, where they have general skills with many different kinds of subcategories.  For instance, a knowledge skill could be about anything, and what the particular skill is, is chosen when you get the skill.  (e.g. Area Knowledge, Washington D.C.)....

But specializing a skill my fix some design problems I'm having with my sci-fi game.... hmmmm...

So why wouldn't a person just buy a general skill (computing) instead of getting more specific skills?  Why get computer 60% and hacking 5% when you can just buy computer 65% and get bonuses to everything?

[Zakharov]

Ideas are good!

The basic mechanic is that you pay as much as the skill. If you know you're doing a lot of hacking, it's easier to pay the 5 points for 5% of hacking then to pay 61, 62, 63, 64 and 65 for broad computers. This is to simulate the fact that one usually starts off learning broad knowledge, but when you get too a point, you start to specialize, wether you know it or not. A musician first learns how to play his instrument (broad) and then a type of music (specialized) and possibly a specific area in that genre (narrow).

I'm still working on a lot of the ideas for the system, the ability check thing has just bugged me for a while. The idea above isn't completed, it just gives you an example of what I mean.

[Bob Bell]

Hi.

I wrote a basic system like you are working on ages ago and just typed it up and posted it, check what I have so far if you like  at www.bobsgames.net

I have a question--your skill system is based on d100 so why is your stat system based on a different scale? This causes issues in other games where, for instance, Hit Points go to the moon but weapon damage is 1d6, 1d8, or some such low number. I would suggest the easiest and most logical thing to do is make the stat system on the same 100 point scale.

Doing this means your character can roll against a Stat like Yuri's Perception just the same as using his Skill with Pistols. No problem.

The only tweaking you have to do is if you are going to use Stat + Skill, in which case your difficulty modifiers have to go up, but again, that is a very minor issue and just means increase the difficulty modifier!

My 2 cents,
bob bell

[GregS]

Cool discussion and ideas.

I definately agree with Bob...I would strongly encourage you to universalize the stat/skill/roll system.  Heck, if you wanted, you could even do a 0-50% range for each, thus giving you a stat + skill total of up to 100.  That makes applying modifiers simple and doesn't leave you in the lurch for skill webbing because they always have the stat to fall back on.  

Now, granted, that's a departure...but I just thought I'd throw it out there.

[MarktheAnimator]

50% for each?

You will have to decide just how much a stat will contribute to a skill roll.

50/50?
10/90?
30/70?

I use 1-20 for stat and then let them add percentage to this number for a combined skill roll.  I suppose the higher the skill goes, the less the stat enters into it... hmmm...

you could also convert the other way.

instead of percentile, you could stick with 1-20 ... just by dividing by 5.

[M. J. Young]

Let me call your attention to Multiverser, because it uses d100 for resolution of all skill-based activities, which are the most common in the game. It does not use d100 for attribute checks, but I'll get to that in a moment.

In your system, you really are going to have to seriously rethink the relationship between attribute values and the resolution mechanic. Let's take your example. The basic attribute is 5%; it's probably going to be about 15% when the character is ready to play. Let's forget the difficulty modifier for the moment--this is the ordinary task, the unmodified situation. Your ordinary basic attribute of 5% is going to pass his perception check one time in twenty tries--which means that nineteen out of twenty times he's going to fail it. Your standard 15% attribute is going to succeed thrice as often, three times in twenty--but he's still failing 17 out of every twenty rolls, 85% of the time.

Why?

[*]It makes sense if attribute checks only happen in near impossible situations--when the character really doesn't have much chance of success. Thus if an ordinary person would see whatever this is at least 10% of the time, he's assumed to have seen it. We're not going to bog the game down with rolls for things that the character might be able to do; we're going to assume that he's got incredible superhuman luck, and if he might succeed (at attribute-based tasks) he does, and only use attribute checks when no one believes he could do it.[*]It makes sense if attribute checks are only used as fallback for the absence of skill--if a character with no skill at shooting a gun is going to attempt to shoot a gun at someone, so we're going to use his dexterity because the odds of success are going to be so far below those he would have if he had the skill. This implies some silly things--like you need Break Door skill to break down a door, because Strength is unlikely to do it if you don't know how--but if combined with the first statement (that attribute-based tasks succeed automatically if the odds are not overwhelmingly against the character) it could work.[*]It makes sense if you want to drive players away from attributes into skills that cover those areas. If my best hope for perception is that I'm going to see things one shot in five (an attribute of 20) when the roll is unmodified, I'm going to attempt to find a skill that lets me spot danger, or allows me to respond quickly when surprised, because I know my perception rolls will fail me regularly.[/list:u]
If none of those are your objective (and you can't think of one that I missed), you've probably got to rethink the attributes area rather seriously.

If you're committed to keeping the percentile system, the advice already given is key. To design a resolution mechanic, one of the critical questions to ask yourself is how often you want the characters to fail the roll. Think of it in terms of ordinary people doing ordinary tasks, then swap out for extraordinary people and extraordinary tasks. That is, what chance has an ordinary person to do an extraordinary task, or an extraordinary person to do an ordinary task, or an extraordinary person to do an extraordinary task? Figure out how to get your mechanics to represent those odds. Maybe the answer is that all attribute checks start with +40 or +50 on the attribute to push the numbers up to the middle of the range before modifying them. Maybe the answer is that all characters get a starting attribute of 45 on which to build. Maybe you need to give them more points, but declare minimum attributes--all players must buy all attributes up to at least 30.

In the alternative, you should consider abandoning the commitment to percentile dice, and instead examine where your character attributes are likely to be, thus modifying your die rolls to produce the desired outcomes. For example, you've indicated an "ordinary" range of attributes from about 5 to 15, and suggested that the "extraordinary" range goes up to 100, but that it's extremely difficult to get above perhaps 25. Let's take that as a base on which to experiment. Now you could say that for any simple task the player rolls d10; clearly any character with an attribute of 10 or greater need not roll for simple tasks. For difficult tasks d20 is rolled, which means that until your attribute reaches 20 you've got a chance of failure. You could go with a d30, and there are simple ways of getting linear rolls to 40, 50, 60, 80, 100, and 120, if you think they're warranted within the game.

Multiverser scales attributes and skills on an 11 to 40 range; ordinary human attribute is 15, and these rarely exceed 30 (considered the best that can be achieved by humans without extraordinary means), while  skills range upwards from that. The two are added, along with bias (an even number between 2 and 30 which is significant in the game), for the chance of success, with other modifiers added or subtracted after that. Attribute checks against the 11 to 40 scale are made on 2d20 for simple checks and d30+10 for difficult ones. If an attribute check is to be modified, the adustment is made to the attribute itself, creating an "effective attribute" for the task.[/list:u]I hope this helps.

--M. J. Young

[Bob Bell]

Zakharov:

'relatively' easy rpg system to play around the house as an introduction to the big boys

MJ Young:

Attribute checks against the 11 to 40 scale are made on 2d20 for simple checks and d30+10 for difficult ones. If an attribute check is to be modified, the adustment is made to the attribute itself, creating an "effective attribute" for the task.

These comments are meant to focus on Zakharov's stated goal of a simple introduction and do not reflect my agreement with MJ's comments only because they mostly add complexity, especially the Multiverser mechanic which I will address at the end of this, and so are not appropriate for an "easy introduction".

I would keep a single dice type, linear rolls (a single dice), a single range of values for all actions and attributes, and a single action modifier made consistently either before or after the action roll for each participant in a contest if you really want an easy introductory rpg mechanic. I would personally use d100 or 2d10, but as long as you are consistent in scale you could use whatever sided die you want.

For example, you want to introduce your mom to rpg action, so you have two choices before you (I recommend you actually try this even if your mom is a die-hard analytical statistician or has been gaming for years):

    1) Explain to mom that her character has a 50% chance to miss when punching the goon in front of her and she rolls dice that will give her a range of from 1-100, so she needs to roll 50 or more to hit. Tell her to hit the goon's boss, who is a better fighter, she has a bigger chance of missing (70%) and so she has to roll 70 or more. Extrapolate and do your perception check against a 100 pt scale. Try adding an explicit modifier after she gets the basic action roll down (the boss number has an implicit modifier, but don't get into that, keep it simple!) Always apply any modifiers before or after the roll, for consistency for your new player.

    2) Explain to mom that her character needs to roll 2d20 and add her skill + her appropriate stat and get a 30 or more to avoid missing the goon in front of her because a 30 on 2d20 with average stat+skill over an 11-40 point range is "average" difficulty for a low-level character. If she wants to hit the goon leader she will have to roll a d30 and add 10 for some reason, plus her stat and skill because that is a different difficulty and can't be explained the same way. Mom won't even want to touch a scale of attributes like perception that doesn't match the scale of skills in this system and may have modifiers explained or added differently.

Mom will understand method 1 pretty well, even if she has questions she will know what her chances of actually doing something are. Mom will not have any idea what her chances of success are relative to one another and will be 100% dependent on you to tell her what to roll with Method 2, which is really 3 or more methods tossed together.

Even an experienced gamer might be misled by the Multiverser system: I am not sure if the rule quote by MJ is correct because I don't know the Multiverser game, but if it uses the dice system as written, it is flawed in my opinion because it is easier to do great at difficult actions than easy ones. Why? Because 2d20 make a bell-curve whereas 1d30 is a linear chance. Simple tasks have a 1/400 chance of a 40 but hard tasks have a 1/30 chance of a 40. Which would you rather roll? I also don't think it matters what you apply an action's modifier to other than confusing new players if you don't apply it consistently before the roll or after the dice roll. The result is the same no matter when a +/- modifier is used or what component it is added to, just do it consistently before or after the roll.

Better-still, for an introductory game just tell the player what their percent chance of failure is and have them roll over that every time. Simple, Logical, and quick to explain in complete detail: "your dice roll is always against your character's chance to fail at a vital action, whatever factors go into that number just remember that you can look at the situation as a whole and say your character is X% likely to fail, and that is the number you have to beat." Leave complications like dice averaging, attempts at scaling over obsequious ranges, and varying systems of probability within the same game for the post-introductory play, if at all.

Bob