Ability Score Generation (6d6, drop highest, drop 2 lowest)

My group has always used point buy or arrays to make characters, but with D&D next I wanted to have semi-random ability scores.

Just rolling 3d6 was a bit too random and the typical "4d6 drop lowest" was still a tad random and also resulted in strong PC's. After looking around at what must have been more than a hundred ways to generate random ability scores I decided on my own.

Roll 6d6, drop highest and 2 lowest. (I feel this would result in decent scores, but fewer really high and low results)

Players MUST reroll if:
     -Total ability modifier exceeds +8 (prevents overpowered characters)
     -Total ability score exceeds 80 (it's possible to get up to 82 and still have only a +8 total modifier; 16 16 14 10 10
       10 vs 17 17 15 11 11 11)

Players MAY reroll if:
     -Total ability modifier is less than +4 (prevents underpowered characters)
     -Total negative ability modifiers are less than -2 (a little negatives help shape a character, too many can
       ruin a character)
     -No ability modifier exceeds +1 (more protection from underpowered characters)


So, following these guidelines a character can have a total of 64 to 80 points. Falling into a situation where you need to reroll happens, but it's not super frequent.

For my group, I'm going to allow them to pick between assigning the highest ability score and two lowest or the lowest and the two highest. The other three ability scores will be randomly placed. This prevents someones primary or secondary ability from being a bad score, yet keeps some randomness to ability score assignment.

ANYWAY, I want to know what you guys think. What would you change? 
Also, if some one is good at probability some statistics would rock.
Like, what's the average total for a 6d6 (drop highest, drop 2 lowest) roll?

I don't like it. 

If you are going to do random then fine, but if it ends up in an exercise in just weighting the dice then what is the point? It smacks of power creep to be honest - and let's not forget that Race, Class and Level all add to Attribute levels (unlike previous editions) without any subtractions at all. 

I understand the frustration when you roll low, but my best solution is to provide a point spend model based on 70 points, say?  

Basically, I'd like all Abilities rolled on 3d6 (simply) with the follow-on option of points-spend if the player wants to reject his original rolls.
I understand not liking it, but it far from smacks of power creep. This system results in lower results than the 4d6 drop lowest method that most people use and isn't as likely to roll extremely low or high like plain 3d6. I want the uniqeness of random ability scores. Seeing the same point distributions over and over again on characters is boring.

Edit: I mostly DM, and when I do roll a character I prefer theme over functionality. Low ability scores aren't a problem as long as it isn't at the point of not being able to play the game. 
Well 6d6 with dropping the highest and dropping the lowest gives such a small divergence that it almost defeats the purpose of rolling at all to get random numbers.  4d6 dropping the lowest is optimal, the average stat will end up around 8-12 and if you allow players to distribute their score to any stat, thats all the optimization and options they need to make any concept they want. Important thing is that you will have freak divergence... an 18... a 4.  Thats what you want.

The purpose of using random attribute rolls is to create divergence and interesting characters players would ordinarly never create on their own.  Point Buy system really do lend themselves to optimization and ultimatly leads players to min maxing.  The end result is very one dimensional characters.  

Rolling attributes gives you considerable divergence which for D&D, a game of optimization, is a good thing.  You want people to have sub-optimal characters, it makes the game more challenging and more interesting for them.  They might resist you, players have a natural tendency to want to break the game, optimize and effective rationalize power gaming into character concepts.  I've never met a player who doesn't try to find a way.  In the end however they ruin their own experiance, I can't tell you how many times I had to kill off characters because they where over optimized and ruining the game.  Divergence is your friend.

"Edition wars like all debates exist because people like debates"

http://www.gamersdungeon.net/

4d6 minus lowest die means you can go from low scores to high scores.
6d6 minus one highest and 2 lowest means you can go from the low end of medium to high scores.
Personally I wouldn't enjoy playing a character who has at least +2 on every modifier. Unless I rolled it against all odds (4d6).

And I dislike rerolling. Unless the player is obviously in a massive disadvantage - like no positive modifiers, for example. Which technically should never happen.

"I am the original gangsta"

-Warlock lvl 5 at the start of the session, dead by the end.

Here's some arrays I rolled using different methods.

4d6, drop lowest
8, 11, 10, 7, 12, 16          ->       7,   8, 10, 11, 12, 16 (64)    ->     -2, -1,  0,  0,  1,  3     (+1 total, negatives at -3)
10, 9,10,12, 10, 9            ->       9,   9, 10, 10, 10, 12 (60)    ->     -1, -1,  0,  0,  0,  1     (-1 total, negatives at -2)
13, 14, 14, 12, 13, 11      ->     11, 12, 13, 13, 14, 14 (77)    ->      0,  1,  1,  1,  2,  2     (+7 total, negatives at 0)*
17, 15, 15, 13, 14, 10      ->     10, 13, 14, 15, 15, 17 (84)    ->      0,  1,  2,  2,  2,  3     (+10 total, negatives at 0)
9, 13, 8, 9, 9, 16              ->       8,   9,   9,   9, 13, 16 (64)    ->     -1, -1, -1, -1,  1,  3     (+0 total, negatives at -4)
10, 15, 15, 9, 13, 15        ->       9, 10, 13, 15, 15, 15 (77)    ->     -1,  0,  1,  2,  2,  2     (+6 total, negatives at -1)*
12, 13, 13, 11, 7, 12        ->       7, 11, 12, 12, 13, 13 (68)    ->     -2,  0,  1,  1,  1,  1     (+2 total, negatives at -2)
13, 12, 13, 12, 18, 17      ->     12, 12, 13, 13, 17, 18 (85)    ->      1,  1,  1,  1,  3,  4     (+11 total, negatives at 0)
13, 18, 6, 15, 14, 7          ->       6,   7, 13, 14, 15, 18 (72)    ->     -2, -2,  1,  2,  2,  4     (+5 total, negatives at -4)*
11, 11, 9, 7, 12, 7            ->       7,   9,   9, 11, 11, 12 (58)    ->     -2, -1, -1,  0,  0,  1     (-3 total, negatives at -4)

6d6, drop lowest 2, drop highest
13, 11, 13, 14, 14, 10      ->     10, 11, 13, 13, 14, 14 (75)    ->      0,  0,  1,  1,  2,  2     (+6 total, negatives at 0)
15, 10, 9, 11, 15, 16        ->       9, 10, 11, 15, 15, 16 (76)    ->     -1,  0,  0,  2,  2,  3     (+6 total, negatives at -1)
14, 8, 11, 10, 17, 10        ->       8, 10, 10, 11, 14, 17 (70)    ->     -1,  0,  0,  0,  2,  3     (+4 total, negatives at -1)
11, 11, 16, 14, 17, 6        ->       6, 11, 11, 14, 16, 17 (75)    ->     -2,  0,  0,  2,  3,  3     (+6 total, negatives at -2)
12, 8, 12, 15, 13, 10        ->       8, 10, 12, 12, 13, 15 (70)    ->     -1,  0,  1,  1,  1,  2     (+4 total, negatives at -1)
11, 12, 12, 11, 13, 16      ->     11, 11, 12, 12, 13, 16 (75)    ->      0,  0,  1,  1,  1,  3     (+6 total, negatives at 0)
11, 14, 15, 9, 8, 12          ->       8,   9, 11, 12, 14, 15 (69)    ->     -1, -1,  1,  1,  2,  2     (+4 total, negatives at -2)
12, 16, 15, 9, 12, 9          ->       9,   9, 12, 12, 15, 16 (73)    ->     -1, -1,  1,  1,  2,  3     (+5 total, negatives at -2)
14, 11, 18, 8, 12, 14        ->       8, 11, 12, 14, 14, 18 (77)    ->     -1,  0,  1,  2,  2,  4     (+8 total, negatives at -1)
15, 17, 11, 9, 15, 11        ->       9, 11, 11, 15, 15, 17 (78)    ->     -1,  0,  0,  2,  2,  3     (+6 total, negatives at -1) 

I had to reroll 9 time: 4 times due to total negative ability modifiers being less than -2, 3 times for a total being less than +4, and 2 times for a total exceeding +8. You can expect to roll about twice when making a character with this method.

You can see that the arrays for the 6d6 were varied, yet not too weak or strong; and despite rolling for 10 arrays not a single one was identical. The 4d6 method was also varied, yet only 3 of the arrays* fit into what the player handbook suggests as playable and one of those "playable" arrays has two stats sporting a -2 modifier.

I know this isn't a huge pool of data, but I'm not good at creating functions and I've been rolling dice for way too long lol.
Well 6d6 with dropping the highest and dropping the lowest gives such a small divergence that it almost defeats the purpose of rolling at all to get random numbers.  4d6 dropping the lowest is optimal, the average stat will end up around 8-12 and if you allow players to distribute their score to any stat, thats all the optimization and options they need to make any concept they want. Important thing is that you will have freak divergence... an 18... a 4.  Thats what you want.

The purpose of using random attribute rolls is to create divergence and interesting characters players would ordinarly never create on their own.  Point Buy system really do lend themselves to optimization and ultimatly leads players to min maxing.  The end result is very one dimensional characters.  

Rolling attributes gives you considerable divergence which for D&D, a game of optimization, is a good thing.  You want people to have sub-optimal characters, it makes the game more challenging and more interesting for them.  They might resist you, players have a natural tendency to want to break the game, optimize and effective rationalize power gaming into character concepts.  I've never met a player who doesn't try to find a way.  In the end however they ruin their own experiance, I can't tell you how many times I had to kill off characters because they where over optimized and ruining the game.  Divergence is your friend.



Sir.  Gary Gygax could have written this himself.  It is a fact that any DM has to learn early that what players want will often not be what is best for their enjoyment of the game.   They generally appear to want things easy but are far happier when they have to work for it.   


My Blog which includes my Hobby Award Winning articles.

Well 6d6 with dropping the highest and dropping the lowest gives such a small divergence that it almost defeats the purpose of rolling at all to get random numbers.  4d6 dropping the lowest is optimal, the average stat will end up around 8-12 and if you allow players to distribute their score to any stat, thats all the optimization and options they need to make any concept they want. Important thing is that you will have freak divergence... an 18... a 4.  Thats what you want.

The purpose of using random attribute rolls is to create divergence and interesting characters players would ordinarly never create on their own.  Point Buy system really do lend themselves to optimization and ultimatly leads players to min maxing.  The end result is very one dimensional characters.  

Rolling attributes gives you considerable divergence which for D&D, a game of optimization, is a good thing.  You want people to have sub-optimal characters, it makes the game more challenging and more interesting for them.  They might resist you, players have a natural tendency to want to break the game, optimize and effective rationalize power gaming into character concepts.  I've never met a player who doesn't try to find a way.  In the end however they ruin their own experiance, I can't tell you how many times I had to kill off characters because they where over optimized and ruining the game.  Divergence is your friend.

I have some friends who don't care about optimizing their characters. As long as they can do the cool things the descriptions promised them.

When we write something that sound like an universal truth, we can be sure that it will turn into a lie.
Then I agree with Emerikol, Gary Gygax could have written this himself.
(yes, that also sounds like an unversal truth, which is utterly funny !)

If you think my english is bad, just wait until you see my spanish and my italian. Defiling languages is an art.

Yep, the whole min max thing is a bit of a perception issue really.

I prefer points build because I prefer to simply prioritise what I'm good at, and what I'm not so good at, and move on. I don't get coaught up in the numbers particularly, and if everybody works on the same amount of points it's really somebody's  own problem if they want to min/max. On a 70 point basis (which has scores averaging just below 12), I'd distribute thus: 

16, 13, 13, 10, 10, 8. (i.e. One thing I am 'great' at, two things I am 'good' at and one thing I am 'poor' at).

When the effecting scores ultimately just range between -1 and +3, then why do you need to spend more time on them?  

With bonuses due to Race and Class still to come. (In my view, the Races should be have penalties for nominated Abilities too!)

Now, if somebody prefers random chance then great, but honestly if you are prepared to roll the bones then you should be happy to accept whay you roll, without making up elaborate ways of weighting the dice roll. 

Well 6d6 with dropping the highest and dropping the lowest gives such a small divergence that it almost defeats the purpose of rolling at all to get random numbers.  4d6 dropping the lowest is optimal, the average stat will end up around 8-12 and if you allow players to distribute their score to any stat, thats all the optimization and options they need to make any concept they want. Important thing is that you will have freak divergence... an 18... a 4.  Thats what you want.

The purpose of using random attribute rolls is to create divergence and interesting characters players would ordinarly never create on their own.  Point Buy system really do lend themselves to optimization and ultimatly leads players to min maxing.  The end result is very one dimensional characters.  

Rolling attributes gives you considerable divergence which for D&D, a game of optimization, is a good thing.  You want people to have sub-optimal characters, it makes the game more challenging and more interesting for them.  They might resist you, players have a natural tendency to want to break the game, optimize and effective rationalize power gaming into character concepts.  I've never met a player who doesn't try to find a way.  In the end however they ruin their own experiance, I can't tell you how many times I had to kill off characters because they where over optimized and ruining the game.  Divergence is your friend.

I have some friends who don't care about optimizing their characters. As long as they can do the cool things the descriptions promised them.

When we write something that sound like an universal truth, we can be sure that it will turn into a lie.
Then I agree with Emerikol, Gary Gygax could have written this himself.
(yes, that also sounds like an unversal truth, which is utterly funny !)



I guess my point is that the point buy system or any die rolling system that "rigs" artificially the end result that the player wants to achieve is pointless, may as well just ask them what they want and give it to them.  Why bother with all the hocus pocus.  Players will always do three things given the oppertunity.

1. Create a powerful character
2. Have dump stats
3. optimize their attributes with powers to ensure whatever powers they use will be as maxed as they can get them.

There is nothing wrong with that until you connect the fact that the point buy system ensures that their character will always have a 17 or 18 in their primary stat, which will bolster all the powers he will use to their max or near max.  While simultanously they will "dump stat" attributes for things they don't care about.  

Rare things you will see in D&D are Wis fighters, stupid Wizards, slow thieves ..etc... Which is the case even if you use 4d6 as people still place their best rolls to whatever primary attributes they will have for their character.  The only difference is that they won't be optimized.

Note too that this not stop anyone doing anything, nor does it break concepts.  It simply brings the game to a greater average where exceptions are just that.. exceptions... not the rule.  In a typical Pathfinder or 4th edition game, whatever classes the players create are always going to be at the peek of performance of that class.  If you like that sort of thing... why bother with point buy or limitations.. or creating some sort of die roll system that creates that effect most of the time?  Just give players the attributes they are asking for or fit their concept.  I mean .. random attributes are intentionally random.  Your players have to be on board and enjoy the divergence, if they are not, if they insist on min maxing or customizing.. just let them write down what attributes they want.  

Do an experiment next time.  Tell people they can have any stats they want, just write them down and make sure they match your concept.  Then compare it to the point buy system.  Most players will go over budget. 

"Edition wars like all debates exist because people like debates"

http://www.gamersdungeon.net/

Here is why XGuild

Every player wants a character that is minimally good enough.  But he could care less about any dump stats.  He will never put much in a dump stat if he can put it in something else.

My issue is that universal class dump stats is boring the tenth time you play it.  Why not have a wise or charismatic fighter?  

My own system is as follows (at least for now )
1.  Roll 4d6 drop lowest IN ORDER
2.  If total falls below 80 reroll the lowest stat.
3.  If total is still below 80 then go back to step 2.
4.  Now exchange any two stats ONE TIME.

This will guarantee a player can have a decent stat in his primary attribute.  But it also leaves open the possibility that he could have some non-traditional stat that is great.  It also means that his secondary preferred stats won't always be great.

I am not against good scores.  I'm against bad scores everywhere else.  Because I think roleplaying opportunities are lost.   

My Blog which includes my Hobby Award Winning articles.

Here is why XGuild

Every player wants a character that is minimally good enough.  But he could care less about any dump stats.  He will never put much in a dump stat if he can put it in something else.

My issue is that universal class dump stats is boring the tenth time you play it.  Why not have a wise or charismatic fighter?  

My own system is as follows (at least for now )
1.  Roll 4d6 drop lowest IN ORDER
2.  If total falls below 80 reroll the lowest stat.
3.  If total is still below 80 then go back to step 2.
4.  Now exchange any two stats ONE TIME.

This will guarantee a player can have a decent stat in his primary attribute.  But it also leaves open the possibility that he could have some non-traditional stat that is great.  It also means that his secondary preferred stats won't always be great.

I am not against good scores.  I'm against bad scores everywhere else.  Because I think roleplaying opportunities are lost.   

I actually think the issue lies more with stereotyped Classes than it does with points spend. 

Why have Fighters with high Wisdom or Charisma for example?

Answer: because they could make for interesting characters!

The trick is allowing Classes the customisability to be able to take skills, feats and specialisations that allows them to play these interesting characters.

A Fighter with a high Charisma could be an inspirational leader....if there were some abilities in offer that he could choose to accentuate this idea. A Fighter with a high Wisdom could be a seasoned veteran, etc.
Here is why XGuild

Every player wants a character that is minimally good enough.  But he could care less about any dump stats.  He will never put much in a dump stat if he can put it in something else.

My issue is that universal class dump stats is boring the tenth time you play it.  Why not have a wise or charismatic fighter?  

My own system is as follows (at least for now )
1.  Roll 4d6 drop lowest IN ORDER
2.  If total falls below 80 reroll the lowest stat.
3.  If total is still below 80 then go back to step 2.
4.  Now exchange any two stats ONE TIME.

This will guarantee a player can have a decent stat in his primary attribute.  But it also leaves open the possibility that he could have some non-traditional stat that is great.  It also means that his secondary preferred stats won't always be great.

I am not against good scores.  I'm against bad scores everywhere else.  Because I think roleplaying opportunities are lost.   



Its a fine system, I mean I don't have anything against good stats either but for me as a DM and as a player, the best role-playing oppertunities and the best characters are the ones you did not expect to have.  

Its not just about stats for individual characters, but characters as a group.  I mean having a character with a 4 Charisma, or 5 Dex in the party creates problems for everyone.  Its these divergence which create unusual role-playing situations and make the game a challenge.  The 4 CON mage is mostly unaffected in terms of his own abilities and powers, but when its time to make a run for, suddenly its a problem for the whole group.

To me fantasy role-playing adventures are always the most interesting when characters have major disabilities as well as powerful abilities.

"Edition wars like all debates exist because people like debates"

http://www.gamersdungeon.net/



Why have Fighters with high Wisdom or Charisma for example?

Answer: because they could make for interesting characters!

The trick is allowing Classes the customisability to be able to take skills, feats and specialisations that allows them to play these interesting characters.

A Fighter with a high Charisma could be an inspirational leader....if there were some abilities in offer that he could choose to accentuate this idea. A Fighter with a high Wisdom could be a seasoned veteran, etc.



We already have that... Its called 1st edition.

Just saying.

EDIT:  I would also say that 4th edition did this very well too, though, its benefit was always combat driven.  So an inspirational leader with high Charisma would give a combat bonus in some way rather then say something more related to what the attribute represents.  Traditionally in some way these things have always existed in D&D, but to me 1st edition represented it better because it formulated a purpose for each attribute in narrative terms rather "bonuses" to some mechanical aspect of the game.  Or perhaps better to say it did both.

In the end though dumps stats have always existed in some form or another for every class, its a rather natural thing, but I think having a stat have an impact narratively means that if you dump stat in something, you'll have to live with some consequence, some drawback and this is how you get to those "we aren't perfect" narratives which really make the game far more interesting than your typical "we are awsome at everything" games. 

"Edition wars like all debates exist because people like debates"

http://www.gamersdungeon.net/

You all make interesting points, but there are already a hundred threads covering those topics. I'd like to get the discussion back to the mechanical aspect of rolling 6d6 drop highest and two lowest. 
Someone made the point that rolling that many dice is overcomplicated, which may be true for your goals. I'm trying to get Semi-Random ability scores, with outliers less common and enough player choice that if they had a class in mind their primary attribute won't be poor. I feel rolling in this manner accomplishes that, but I want to know if others have another way of accomplishing this and if anyone is capable of statistics regarding 4d6 (drop the lowest) and 6d6 (drop the highest and the lowest 2).

Let's also keep this in the realm of 4e and Next 
A long time ago, when I was rolling ability scores, we had a system.

we rolled three times : 6 + 2d6
and three times : 4d6 drop the highest.

If you think my english is bad, just wait until you see my spanish and my italian. Defiling languages is an art.

A long time ago, when I was rolling ability scores, we had a system.

we rolled three times : 6 + 2d6
and three times : 4d6 drop the highest.

That could result in some harsh arrays. I like it >:D
In my expreince the biggest issue with randomly rolled stats is the disparity between players.  If one person's highest is a 14 and the next guy's lowest is a 14 you have tension. 

The last time we did random rolls we tried something new.  Each player and the DM rolled one stat (4d6 drop the lowest) and everyone used the same stats, placed how they chose. We got the fun and tension of random rolls, without any one person getting shafted.

I love D&D more than I could ever love a human child.

In my expreince the biggest issue with randomly rolled stats is the disparity between players.  If one person's highest is a 14 and the next guy's lowest is a 14 you have tension. 

The last time we did random rolls we tried something new.  Each player and the DM rolled one stat (4d6 drop the lowest) and everyone used the same stats, placed how they chose. We got the fun and tension of random rolls, without any one person getting shafted.


Rolled Scores are not for every group. I was in a 3E campaign that ended before the 1st session because someone rolled AWESOME (lowest score was a single 15) and an uber-competitive player couldn't accept being obviously weaker than another PC. If you have players like that, Point Buy or Array is by far the best option.

As a player, I'm not a fan of rolled Scores, only because I suck at rolling. In 3E I rolled the worst possible character that was considered playable. I would like some reasonable guidelines as to what is and is not considered playable in the final product.

As a DM, I like rolled Scores. It allows for each character to become unique, especially if some method is used to keep a player from placing them optimally. I want someone to have a great Score in their primary Ability, and a good one in their secondary, but I also like a few oddities. A high Int Fighter, or a high Str Wizard come to mind.

A method I've used before is to have a player choose their Class first (most players do this anyway). Each class has a Primary and a Secondary Score. Roll 4d6 as normal, and assign the highest roll to the Primary and the 2nd highest to the Secondary. Then assign the remaining scores randomly (1d4, 1d3, then 1d2). This allows for some interesting characters, but still allows players to focus on what their character class is designed around. The only complaint I've heard about it is when Con is the lowest score, because on a bad roll that can greatly reduce survivability. I have (on occasion) allowed the player to swap Con for their 2nd lowest Score.
A long time ago, when I was rolling ability scores, we had a system.

we rolled three times : 6 + 2d6
and three times : 4d6 drop the highest.

Back in the day, we rolled one set of stats with 4d6, drop lowest and reroll ones then rolled a second set, 5d6 drop 2 lowest and picked the set we liked. It allowed for a random feel without a big chance that someone would feel they sucked. And we NEVER played that you couldn't pick where your stats go. I've never understood forcing someone to play something they don't want (IE the player that REALLY wants to play a wizard rolls a 4 int... well too bad for you, here is a barbarian sheet...).

Shiroiken's way is sort of ok with me but I'd rather I picked which way my character was. While he MAY end up unique with a crappy score, if it's in something not in keeping with the character I'd like to play to really makes it less fun for me. For example, if I'd make a charismatic ladies man fighter, I'd put as good a stat in CHA as I could without gimpimg my combat but with a forced pick it's those picks that determine the character I play. I roll a 4 in the CHA slot and out goes the character I want to play. I know some figure out their character after stats and that's cool, but I make a character and then roll stats and place them to fit the character I want to play.
Well, if we wanted to encourage high-Wisdom fighters we could give them some sort of class feature, perhaps at 1st level, to let them add their Wisdom bonus to attack and damage on opportunity attacks.  Since that's situational we could add some other benefit, like halting movement with opportunity attacks, allowing them a subtle way to demonstrate their superiority at the fundamentals of combat.  A kind of "Combat Superiority" feature, if you will.  We can work out the details later.

Though personally I'd say it would be best to move away entirely from the 3-18 model, which is much more complicated than it needs to be, and move to a 1-5 model, with 1 being a weak average human and 2 being a strong average human.  A stereotypical nerd has Strength 1 and Intelligence 2, a stereotypical high school jock has Strength 2 and Intelligence 1.  3 is Olympic athlete or Mensa member, 4 and up are superhuman.
*snip*



Oooh, an INTELLIGENT idea.  You realize it's never going to happen, right?
Oh, of course, it's not traditional at all.  But I can dream.
To me rolling dice for attributes is strictly a D&D thing, I've never practiced it in any other game.  In general I think the reason behind doing it in D&D, at least as it was in 1st edition, was just part of the sort of atmosphere and quirk of the system.  If you wanted to get that experiance, you follow the footsteps of the classic method.  I don't think I could come up with a particularly convincing reason to do it other than that.

That said, if your gonna do it, may as well do it in the classic way.  Its why I say "rigging" it to get some sort of results you want is kind of pointless.  If you want slightly random attribute scores with some semblence of control, dice is not the way to go as the chaotic nature of it will always result in big upsets or huge winners no matter how you ultimatly do it.


The  6d6 method your talking about will probobly yield roughly the same results as the 4d6 method I think, I mean Im not statistics expert but I rolled some dice for a few minutes as Im sitting here and the divergence was pretty much the same.  In both cases I ended up with a couple of 5's and a couple of 17's and 18's.  I think your fine to use it but I would be prepared for some upsets none the less.

"Edition wars like all debates exist because people like debates"

http://www.gamersdungeon.net/

I've used dice-rolled stat allocation in most RPGs that I've played, from the various iterations of Gamma World, Boot Hill, all iterations of D&D, all iterations of Call of Cthulhu, Star Wars, and scores of others. At this point, it's second nature to go into an RPG with the expectation of rolling stats. The idea of point-buy was very foeign to me the first time I saw it. I remember thinking at the time that point-buy had to be some form of cheating. Of course, time went on, and eventually I warmed up to it and found that newer generations of players typically preferred point-buy to die-rolled. I really don't think there is a wrong way to do it...whatever you and your table feels is the most fun method.
"The world is indeed comic, but the joke is on mankind." - H.P. Lovecraft
Oh, of course, it's not traditional at all.  But I can dream.


It IS traditional.
Just in other games
That's how White Wolf products have always worked

Not enough granularity for my taste

I'd rather have it on both sides of the scale with 0 being the average (-5 to +5)
It would basicaly be the same, except with the modifiers themselves as the ability scores 
(isn't it how Dragon age RPG works also?) 
Try radiance RPG. A complete D20 game that supports fantasy and steampunk. Download the FREE PDF here: http://www.radiancerpg.com
We prefer 3d6 reroll 1's as the method for rolling ability scores. It ranges between 6-18, which avoids having scores too low while raising the average just a little within the bell curve.
Its a fine system, I mean I don't have anything against good stats either but for me as a DM and as a player, the best role-playing oppertunities and the best characters are the ones you did not expect to have.


I agree about the "best [...] are the ones you did not expect to have" bit, though I think it gets tempered by a lot of folks (myself included) feeling that the worst role-playing opportunities and worst characters are the ones you do not want to have.

I would really love to see more characters (all characters, even) "work" to a level that most players can enjoy regardless of stats.

Stats as interesting character traits and attributes, I like.
Stats as optimization tools, I really don't.
Feedback Disclaimer
Yes, I am expressing my opinions (even complaints - le gasp!) about the current iteration of the play-test that we actually have in front of us. No, I'm not going to wait for you to tell me when it's okay to start expressing my concerns (unless you are WotC). (And no, my comments on this forum are not of the same tone or quality as my actual survey feedback.)
A Psion for Next (Playable Draft) A Barbarian for Next (Brainstorming Still)
Although I prefer my players play using the Standard Array method, we've played with various methods, probably the most popular random method we ever used was 4d6, minus the lowest, reroll 1's, however, if you got 4 6's, you got a 19, and if you got 4 1's, you got a 20 (but of course, you can try to reroll your 1's to get that 19, but there's no way to get that 20 except on the 1st roll of each stat) We've played this with both place your stats, and put them in order, and then 1 free swap. We've also done it with place in order, but you can subtract from stats to add to others, at a cost of 2 for 1 pt, and only 1 stat can be increased (although multiples can be decreased in this method)

there are dozens of ways to generate stats, I say as long as you and your group are happy with what your using, that's all that matters! 
Want continued support for 4e, check this out, 4e Lives and Breaths

Check out MY eZine, Random Encounters Seuss (lordseussmd on YM)
To get rid of dump stats, you have to make sure no stats are dumpy. If you don't want fighters to dump CHA and wizards to dump STR then make sure there are reprecussions to either decision. For instance, if low CHA meant you could not exert your will over too many magic items (Max magic items = (Level / 5, round up) + CHA), then fighters might be inclined to go for CHA 12+. Not every fighter would need to, but those who definitely want magic items would.

Same for wizards and STR. I liked melee touch attack spells or spells that boost your sword swinging, since a beefy fighter/wizard would benefit from those types of spells. Just make it an option and some wizards might want to pursue that. If every wizard spell is based off INT, then you're going to see INT maxed and everything else ignored, every single time.
If you care about balance and making sure players aren't over/underpowered, then point buy is for you.

If the objection is "point buy is min-max and results in ugly stat distributions that don't look organic," then the problem is the underlying attribute system itself. 4E had this problem because every single character had one stat they used 80% of the time, another stat 15% of the time, and the other 4 stats less than 5% of the time. They're dump stats because they're dumpy. Fix the system (see above post) and you won't see ugly stat distributions in point buy.
When I first started playing we used exclusively 4d6 drop lowest. My DM at the time loved it too much. With this group we used all kind of different stat generation- array,point buy,4d6 drop lowest, 5d6 drop lowest 18 max and variations of everything above. In my opinion it really depends on what kind of game do you plan to run/play. We used 5d6 drop lowest when we played high powerd gestalt characters on epic quest for instance. This 6d6 drop highest and 2 lowest is interesting but i can't really find a game type to use it where it would be better than other methods. And about dump stats. We mostly play in a way where there are no dump stats,especialy now with DDN. You want to have all stats at least with no penalty cause even if you're fighter you'll roll CHA now and then.

IMAGE(http://www.nodiatis.com/pub/20.jpg)

Rare things you will see in D&D are Wis fighters, stupid Wizards, slow thieves ..etc... Which is the case even if you use 4d6 as people still place their best rolls to whatever primary attributes they will have for their character.  The only difference is that they won't be optimized.

This is a problem with how any edition of D&D rewards/penalizes you for that. You could play a Wizard with 9 Int, but you're not going to be casting many spells that way.

The mechanics of the system do not support these concepts. (Which I think is a flaw of D&D)

In my expreince the biggest issue with randomly rolled stats is the disparity between players.  If one person's highest is a 14 and the next guy's lowest is a 14 you have tension. 

The last time we did random rolls we tried something new.  Each player and the DM rolled one stat (4d6 drop the lowest) and everyone used the same stats, placed how they chose. We got the fun and tension of random rolls, without any one person getting shafted.


I like this idea. In my experience the biggest problem with random rolling is indeed the differences between players that can and will be created.

And to answer the OP: Decide what you want, randomness or predictability. Your proposal seems to be aimed at trying to achieve both, while that is impossible.

Predictable with a little randomness can be achieved by picking an array and then adding 1d2 or 1d3 to every ability.

Otherwise you're just going to have to be happy with the predictability of an array or point buy, or be content that some in the party will have 18,17,16,16,15,14 and others 11,9,9,8,5,4 like in one of our 2E games. (after which our group decided to go point buy)

Star Wars Saga Edition is by far the best RPG that Wizards of the Coast has ever published.

 

I once asked the question (in D&D 3.5) "Does a Druid4/Wizard3/ArcaneHierophant1 have Wildshape?". Jesse Decker and Andy Collins: Yes and the text is clear and can't be interpreted differently. Rich Redman and Ed Stark: No and the text is clear and can't be interpreted differently. Skip Williams: Lol, it's worded ambiguously and entirely not how I intended it. (Cust. Serv. Reference# 050815-000323)

This is NOT about having a poor primary stat.  The primary stat needs to be decent and no one is arguing that point.  What I am arguing is that the other stats should vary.  The ones that are not essential to your character but rather can be helpful in different ways are where some of the roleplaying comes in.  Of course the wizard will cast spells and the fighter swing a sword.  But those odd scores that don't matter directly are still sources of roleplaying ideas.

My Blog which includes my Hobby Award Winning articles.

Why can't it be about the primary stat?

"You can only be an adventurer if you have (at least) one high attribute" is pretty arbitrary in the first place.


That aside: if you must have "primary stats", I actually like the Gamma World (4e D&D version) take on it: everyone gets a specific "high" score in their primary stat, and secondary stat (or a higher score if they have only one stat to worry about), and then just straight 3d6 for the rest.
Feedback Disclaimer
Yes, I am expressing my opinions (even complaints - le gasp!) about the current iteration of the play-test that we actually have in front of us. No, I'm not going to wait for you to tell me when it's okay to start expressing my concerns (unless you are WotC). (And no, my comments on this forum are not of the same tone or quality as my actual survey feedback.)
A Psion for Next (Playable Draft) A Barbarian for Next (Brainstorming Still)
This is NOT about having a poor primary stat.  The primary stat needs to be decent and no one is arguing that point.  What I am arguing is that the other stats should vary.  The ones that are not essential to your character but rather can be helpful in different ways are where some of the roleplaying comes in.  Of course the wizard will cast spells and the fighter swing a sword.  But those odd scores that don't matter directly are still sources of roleplaying ideas.


It's only going to make a real roleplaying difference if the stat differences can be significant. Nobody is going to care much about a 13 or 14 in Charisma, but 4 or 18 is probably going to make a real difference.

It really depends on if you want players to have a concept for a character first and then have the attributes molded to fit, or if the attributes are set first and then the concept follows from that.

That's a choice a DM and a group of players has to make.


Star Wars Saga Edition is by far the best RPG that Wizards of the Coast has ever published.

 

I once asked the question (in D&D 3.5) "Does a Druid4/Wizard3/ArcaneHierophant1 have Wildshape?". Jesse Decker and Andy Collins: Yes and the text is clear and can't be interpreted differently. Rich Redman and Ed Stark: No and the text is clear and can't be interpreted differently. Skip Williams: Lol, it's worded ambiguously and entirely not how I intended it. (Cust. Serv. Reference# 050815-000323)


It's only going to make a real roleplaying difference if the stat differences can be significant. Nobody is going to care much about a 13 or 14 in Charisma, but 4 or 18 is probably going to make a real difference.



Not true.  A 14 intelligence fighter will be played vastly different from an 8 intelligence fighter in my campaign.  

My Blog which includes my Hobby Award Winning articles.

If you want 'normalized random', roll three numbers, then use the 21's compliment (subtract the roll from 21) of those rolls for the other three.

It's only going to make a real roleplaying difference if the stat differences can be significant. Nobody is going to care much about a 13 or 14 in Charisma, but 4 or 18 is probably going to make a real difference.



Not true.  A 14 intelligence fighter will be played vastly different from an 8 intelligence fighter in my campaign.


While not as extreme as 4 and 18, that is exactly the point I'm trying to make.

Either you are going to have low variance in your rolling and then most stats will be 12-16, or you're going to have a more unpredictable method but then you also have the risk of high differences between characters.

The only way to avoid that and still roll stats, is what someone else suggested, roll once and give those stats to everyone.

As a sidenote, where would you start making a difference for example on Charisma: 13 vs 14, 12 vs 14, 11 vs 14, 10 vs 14, 9 vs 14? Apparently 8 vs 14 is a big enough gap for you. Is that the same as 12 vs 18, or is it only because the 8 will give a negative modifier and you percieve it as a "bad stat"? How about 6 vs 10, etc.

When to make stats have a roleplaying effect is often arbitrary and varies from person to person and group to group.

Your example of 8 vs 14 is a gap of 6 on a scale of 3-18, it's the difference between rolling 5,5,4 and 1,2,5 for example. More typical of 3d6 than of other methods with lower variance.

Star Wars Saga Edition is by far the best RPG that Wizards of the Coast has ever published.

 

I once asked the question (in D&D 3.5) "Does a Druid4/Wizard3/ArcaneHierophant1 have Wildshape?". Jesse Decker and Andy Collins: Yes and the text is clear and can't be interpreted differently. Rich Redman and Ed Stark: No and the text is clear and can't be interpreted differently. Skip Williams: Lol, it's worded ambiguously and entirely not how I intended it. (Cust. Serv. Reference# 050815-000323)

Just tell players to roll random, and if they don't like the results spend points instead. Creates a 'safety net' of players not being forced to play with below average players is all - and to me that is all you need.   
I don't think some score variation between different players is any big deal.  It really isn't a competitive game.  I think being obsessed over a modifier here or there is a sign of someone thats maybe not a good fit for my campaign.

My Blog which includes my Hobby Award Winning articles.

Sign In to post comments