PDA

View Full Version : Ability Score Generation Idea



bookguy
2013-08-01, 12:20 PM
My idea is a way for people to make their starting ability scores. Previously, I've allowed players to choose between the standard generation (4d6, pick highest 3, six times, arrange as you like) or starting with a very generous array (18, 16, 13, 12, 10, 7) to arrange as they wish. Of course, this is before applying racial modifiers. I've given these choices to my 3.5 and PF players.

My new idea is that scores have to add up to certain numbers. Scores are grouped into 3 categories: Str+Con, Dex+Wis, Int+Cha. The scores in one category must add to 29, another must add to 23, and the last add to 20. No score can be above 18 or below 7. As long as the scores fit these requirements, the players can choose the scores however they like. Again, this is before racial modifiers.

I like my new method because I think it develops more realistic characters, and I'm planning to use it for my next PF campaign (maybe as a choice with another method). However, I'm worried that it will make every character of the same class choose their scores exactly the same and might punish some MAD classes. I was also wondering if there was an easy tweak to help non-spellcasters. Any thoughts?

jedipilot24
2013-08-01, 12:55 PM
The DMG describes several alternate methods of ability score generation, one of them is the Organic method:
Organic Characters: Roll 4d6 six times, discarding the lowest each time. Place in order (Str, Dex, Con, Int, Wis, Cha) as rolled. Reroll any one ability of your choice, taking the higher score. Then switch any two ability scores.
This allows some control over which scores go where but prevents min-maxing and characters may end up with unusually high scores in what their class would usually consider a dump stat.

In my group, the DM gave us straight 4d6 but also allowed us to reroll 1s and 2s. All our scores were above average but that didn't make much difference because in every group of characters we made, there was only--at most--one full spellcaster, usually a healbot cleric (and usually me because no one else wanted to do it). The rest of the party was usually some mix of fighter, barbarian, ranger, rogue, with the occasional paladin.

The only 'easy tweak' that helps non-casters compete with casters is Tome of Battle; love it or hate it, that was its purpose.

Vadskye
2013-08-01, 01:12 PM
You're a White Wolf player, aren't you? Anyway, this is a bad idea in that it is fantastic for casters. Observe Minmax the Sorcerer:
7/10/16/11/10/18
He gets the equivalent of a 32 point buy, and he gets to place his stats almost perfectly.
Now look at what happens to a poor fighter:
18/14/11/13/9/7
He gets the equivalent of a 30 point buy, but his distribution is awful. Think I made a poor distribution? Try to improve it! He's forced to choose between Str and Con, which are his two primary stats, and has a terrible time trying to arrange his scores otherwise.

What does "more realistic" mean in this context? It's definitely not more intuitive, and I don't think it's more "realistic". You put similar stats together, which means that I have to specifically choose to be either strong or tough. In real life, let's say that you have two options for your leisure time: studying or exercising. One would naturally improve your mental stats, while the other would improve your physical stats. However, in this universe, I have to get weaker in order to get tougher, and vice versa. It doesn't make sense.

Razanir
2013-08-01, 02:08 PM
Str+Con is very, very bad for melee characters. Also, Dex+Int and Wis+Cha would be a better split. Legend and 4e already use them for Reflex and Will respectively

AttilaTheGeek
2013-08-01, 02:18 PM
My favorite stat generation gives each player one of three options. Yes, that means two players can choose different stat generations.

The first option is a mid-power point buy, either 22 points in Pathfinder or 28 points in 3.5. Mathematically, it's the least strong, but you know you'll be to do well and get whatever stats you need.

The second option is three sets of 4d6b3 in order, and you pick one of them. You'll almost definitely get some high stats, but you're not completely sure you'll get enough or get the right ones.

The third option is 3d12b2 six times, in order, no rerolls. Statistically, it's incredibly powerful, but it's also incredibly volatile. When picking this option, you'll definitely have solid stats for something, but you don't know what it will be.

They all work out to be equal because the player who picks the point buy can distribute their character's stats any way they like for maximum efficiency, but the 3d12b2 relies on sheer strength rather than efficiency and the 4d6b3 provides a happy medium between the two.

Network
2013-08-01, 02:25 PM
Str + Con does not make sense for obvious reasons, as mentioned. I suggest Str + Dex, which is really the closest to realism you can find, or Dex + Con, which is more balanced for spellcasters and not too bad for ranged characters.

jedipilot24
2013-08-01, 02:39 PM
The third option is 3d12b2 six times, in order, no rerolls. Statistically, it's incredibly powerful, but it's also incredibly volatile. When picking this option, you'll definitely have solid stats for something, but you don't know what it will be.

No kidding that's volatile; I just tried it out in an online die roller program and this is what I got:
Str 18, Dex 16, Con 4, Int 14, Wis 22, Cha 11

Razanir
2013-08-01, 02:43 PM
The third option is 3d12b2 six times, in order, no rerolls. Statistically, it's incredibly powerful, but it's also incredibly volatile. When picking this option, you'll definitely have solid stats for something, but you don't know what it will be.

3d12b2 sounds fun.


Str + Con does not make sense for obvious reasons, as mentioned. I suggest Str + Dex, which is really the closest to realism you can find, or Dex + Con, which is more balanced for spellcasters and not too bad for ranged characters.

Str + Dex: You're either a huge, slow muscle man or a nimble mousy street fighter.

Con + Int: You either did stamina training or fancy newfangled book learnin'

Wis + Cha: Charisma's debated often enough that I'm sure you could find an interpretation the flies counterpoint to wisdom.

SowZ
2013-08-01, 02:44 PM
My idea is a way for people to make their starting ability scores. Previously, I've allowed players to choose between the standard generation (4d6, pick highest 3, six times, arrange as you like) or starting with a very generous array (18, 16, 13, 12, 10, 7) to arrange as they wish. Of course, this is before applying racial modifiers. I've given these choices to my 3.5 and PF players.

My new idea is that scores have to add up to certain numbers. Scores are grouped into 3 categories: Str+Con, Dex+Wis, Int+Cha. The scores in one category must add to 29, another must add to 23, and the last add to 20. No score can be above 18 or below 7. As long as the scores fit these requirements, the players can choose the scores however they like. Again, this is before racial modifiers.

I like my new method because I think it develops more realistic characters, and I'm planning to use it for my next PF campaign (maybe as a choice with another method). However, I'm worried that it will make every character of the same class choose their scores exactly the same and might punish some MAD classes. I was also wondering if there was an easy tweak to help non-spellcasters. Any thoughts?

Here's why it is less realistic. In RL, people with high strength tend to have higher Con. People with high Int are more likely to have high Cha, (though there are definite and plentiful exceptions.) In this system, someone who is a bulging muscle strong man with 18 Str can only have a totally average 11 Con. (Would you expect to take down Hulk Hogan just as easily with a punch as your next door neighbor? Ignore this if your next door neighbor happens to be Hulk Hogan.)

Hanuman
2013-08-01, 03:16 PM
http://tools.digitalightbulb.com/pbcalc.html

bookguy
2013-08-01, 03:26 PM
You're a White Wolf player, aren't you?

I play mainly 3.5 and PF but I just started playing in a White Wolf campaign. I really like how character generation works there so I wanted to try a similar method for my upcoming PF campaign (which will probably be all WW players, incidentally. I'm moving to a new college where WW is very popular). My idea was that I'd have the scores paired with ones that are similar or go well together and so you'd have a lot of points to put into your strengths and only a few for your weaknesses. As you have pointed out, my method works out to about exactly the opposite of my intentions. :smallredface:

@AttilaTheGeek:

I like the idea with the three options (as I said in my first post, I have two options that I have typically given in the past, allowing players to either choose the more luck-based option or just take the array). Do your players who choose the more volatile options get to choose their class based on their stats, or do they have to pick their class first and hope they get good scores for whatever they're going with? Also, does point buy work differently between 3.5 and PF, or is there some other reason you're giving a different number of points?

Edit: ninja'd on the point buy question

Jormengand
2013-08-01, 03:30 PM
No kidding that's volatile; I just tried it out in an online die roller program and this is what I got:
Str 18, Dex 16, Con 4, Int 14, Wis 22, Cha 11

With that constitution and wisdom? Druid.

Hanuman
2013-08-01, 04:44 PM
Allowing characters to buy their points means you are taking away that element. If you really insist on letting players keep their luck then remove 3 points from your pointbuy start and have them roll a 1d6 to determine their general power.

Otherwise I'm not sure what you're getting at.


With that constitution and wisdom? Druid.
Zen archery compound bow cleric imo.

AttilaTheGeek
2013-08-01, 08:12 PM
Allowing characters to buy their points means you are taking away that element. If you really insist on letting players keep their luck then remove 3 points from your pointbuy start and have them roll a 1d6 to determine their general power.

That's why the point buy option is weaker than normal. Most campaigns that use point buy do 32 in 3.5 or 25 in Pathfinder, compared to the 28 or 22 I suggested.

Vadskye
2013-08-02, 12:03 AM
My idea was that I'd have the scores paired with ones that are similar or go well together and so you'd have a lot of points to put into your strengths and only a few for your weaknesses. As you have pointed out, my method works out to about exactly the opposite of my intentions. :smallredface:
Unfortunately, it does the same thing in White Wolf; it's just a little more subtle about it. What about the point buy system doesn't work for you exactly? You say you want something more realistic - but what about point buy is unrealistic, exactly? (I'm not saying it's realistic - but I don't know how to address your concerns if I don't know what they are.)

bookguy
2013-08-02, 10:29 AM
My main issue with point buy is that players can prioritize stats however they want. I'm taking a look at organic now because that seems to solve that issue.

Deepbluediver
2013-08-02, 10:57 AM
Personally, I prefer less uber-specialized characters, so I tend to just have ability arrays (15, 14, 13, 12, 10, 8) and a racial system that is like pathfinders "+2 to two, -2 to one". This gives players a fairly diverse set of options depending on if they want to shore up weak points or min-max.

If a group really wants to do point-buy or can't decided, I give each player a choice between a 25-point buy or an array of 16, 14, 13, 12, 11, 10. That way you can pump one score if you want, but it comes at significant cost.


The third option is 3d12b2 six times, in order, no rerolls. Statistically, it's incredibly powerful, but it's also incredibly volatile. When picking this option, you'll definitely have solid stats for something, but you don't know what it will be.

That sounds interesting; I tried it a few times and this (http://www.giantitp.com/forums/showthread.php?p=15743265#post15743265) is what I got. Those distributions aren't bad, though it's only a small sample. It's also possible to end up with animal like intelligence or other nearly-crippling scores A wizard with a Con of 2 would basically be made of tissue paper...actually just about any class that was taking that kind of hit to HP would feel like they where made of tissue paper.

If you like it and are going for a high-power setting, I might decrease the variability somewhat by making it 3d10b2+4 instead.


Str + Dex: You're either a huge, slow muscle man or a nimble mousy street fighter.

Con + Int: You either did stamina training or fancy newfangled book learnin'

Wis + Cha: Charisma's debated often enough that I'm sure you could find an interpretation the flies counterpoint to wisdom.

I like you're explanations for Str, Dex, Con, and Int.

For Wis and Cha, how about ego vs. awareness? The more self-confident you are, the less easily you pick up on hints and signals around you. And the opposite being "a truly wise man knows he knows nothing". Or something like that.

AttilaTheGeek
2013-08-02, 12:03 PM
That sounds interesting; I tried it a few times and this (http://www.giantitp.com/forums/showthread.php?p=15743265#post15743265) is what I got. Those distributions aren't bad, though it's only a small sample. It's also possible to end up with animal like intelligence or other nearly-crippling scores A wizard with a Con of 2 would basically be made of tissue paper...actually just about any class that was taking that kind of hit to HP would feel like they where made of tissue paper.

If you like it and are going for a high-power setting, I might decrease the variability somewhat by making it 3d10b2+4 instead.

It's supposed to be volatile; the risk that your character could wind up with a 2 in CON is designed to stop it from being the obvious choice over a 4d6b3 array. And statistically speaking, the chance of getting a 2 in CON is less than one out of 1500.

Deepbluediver
2013-08-02, 12:47 PM
It's supposed to be volatile; the risk that your character could wind up with a 2 in CON is designed to stop it from being the obvious choice over a 4d6b3 array. And statistically speaking, the chance of getting a 2 in CON is less than one out of 1500.

Like I said, it's interesting, but it's a significantly higher powered variant than 4d6b3.
This calculator (http://rumkin.com/reference/dnd/diestats.php) gives it's average as just a hair below 16 and 17 as the most common value, compared to a mean and mode of 12.25 and 13, respectively, for the usual version.

I've done a bunch more rolls, and have yet to see any combo without at least one score higher than 20, and most have multiple 20+ values. Single digit scores have been rare.

Edit: Did a few more using this dice roller (http://www.brockjones.com/dieroller/dice.htm); finally got one with no 20+ values. It's still 19, 18, 15, 14, 14, 14 without racial modifiers.

The characters also don't seem very realistic or balanced, given that they can have levels of competence ranging from superhuman to "anemic toddler".

Seerow
2013-08-02, 01:52 PM
Like I said, it's interesting, but it's a significantly higher powered variant than 4d6b3.
This calculator (http://rumkin.com/reference/dnd/diestats.php) gives it's average as just a hair below 16 and 17 as the most common value, compared to a mean and mode of 12.25 and 13, respectively, for the usual version.

I've done a bunch more rolls, and have yet to see any combo without at least one score higher than 20, and most have multiple 20+ values. Single digit scores have been rare.

Edit: Did a few more using this dice roller (http://www.brockjones.com/dieroller/dice.htm); finally got one with no 20+ values. It's still 19, 18, 15, 14, 14, 14 without racial modifiers.

The characters also don't seem very realistic or balanced, given that they can have levels of competence ranging from superhuman to "anemic toddler".

Well if you really like d12 generation, you could do something like 6d12 take middle two.

http://anydice.com/program/2761

Gets you an average of 13, with a pretty nice bell curve. You have ~2% chance of getting above an 18, and the chance of getting above a 20 is practically nil.


Personally I'd stick with more traditional methods, but I could see that working.

Razanir
2013-08-02, 02:21 PM
Well if you really like d12 generation, you could do something like 6d12 take middle two.

http://anydice.com/program/2761

Gets you an average of 13, with a pretty nice bell curve. You have ~2% chance of getting above an 18, and the chance of getting above a 20 is practically nil.


Personally I'd stick with more traditional methods, but I could see that working.

That looks fun, but there are two problems.
1) Generation should have minimum 3+, maybe even 5+ to account for penalties. You know, for Int.
2) Who even cares about d12s? Seriously. All they're used for is barbarians, dragons, dragon barbarians, and greataxes

Deepbluediver
2013-08-02, 02:31 PM
Well if you really like d12 generation, you could do something like 6d12 take middle two.

http://anydice.com/program/2761

Gets you an average of 13, with a pretty nice bell curve. You have ~2% chance of getting above an 18, and the chance of getting above a 20 is practically nil.


Personally I'd stick with more traditional methods, but I could see that working.

That does form a very nice distribution, except for possibilities at the lower end, as Razanir mentioned. You could have a rule that you need to reroll if anything ends up as 4 or less, I guess; that shouldn't push it much higher.

Overall, my issue with most rolling-based systems is not the potential variation within any one character; I can adjust gameplay to accommodate for power levels. Its the variation between the entire group.

Unless you do dice rolls to control for a very narrow range, you're likely to end up with players having widely disparate ability scores more often than not.

Hanuman
2013-08-03, 01:27 PM
That's why the point buy option is weaker than normal. Most campaigns that use point buy do 32 in 3.5 or 25 in Pathfinder, compared to the 28 or 22 I suggested.

It's actually stronger, it just favors balanced characters.
You're going to have about a 14/14/14/14/13/12 array in pathfinder where as the 3d6 is going to put you down to about 10.5 average then about a +1.5 for the extra die swap, averaging 12 which is what the system is based around.

Assuming your average is 12, and that re-rolls are allowed below that point it's not weaker to have pointbuy, it's just lower risk lower reward to have a balanced character. However, if you are striving for a few 18's its probably better to go with 4d6-1d6, but you could easily get stuck with a weaker attribute array.