Remember the "3d6 For Stats In Order" Thread? I'm doing it!

Ratskinner

Adventurer
EDIT: Its the Method @clearstream mentioned

At least for WotC-D&D, the best method for "fair" random stat generation is a method that I saw on this board. (Although I don't remember who thought of it, to give them credit.)

Make a deck of 18 cards from the following:
3-2's
4-3's
4-4's
4-5's
3-6's

Shuffle and then:

Easy: Deal them out into six piles of three cards each. Those are your stats, arrange to suit.

Rough: Same, except the stacks come in order.

Option to make it easier on folks: you may swap one pair of cards between two stacks. (Makes it very likely to get a 16+)

If you want to be super harsh, and replicate 3d6, use a deck of 18 that has three of each number from 1-6.
 
Last edited:

dnd4vr

Adventurer
This is precisely the opposite of my experience. From what I can tell, the AD&D exceptional strength bonus is a tremendous incentive for fighter-players to "cheat"* on stat rolls. I don't know that I've ever seen a single-class AD&D fighter without exceptional strength...which flies in the face of all this "tough guy" OSR rhetoric that I hear so much...and statistics.

*"cheat" here meaning everything from flat-out: "No, I rolled it straight up, honest!" To lobbying for more bizarre rolling methods to suicide by fighters without exceptional strength.
I suppose we simply had different experiences then. 18/xx STR in our 1E games was not rare, but uncommon. Of course for the XP bonus, 16 was often there, and many 17's. More commonly, Fighter's grasped up the Gauntlets of Ogre Power ASAP! The breakdown was probably about this:

15 or lower: 15%
16 or 17: 60%
18/xx: 25%
 

dnd4vr

Adventurer
EDIT: Its the Method @clearstream mentioned

At least for WotC-D&D, the best method for "fair" random stat generation is a method that I saw on this board. (Although I don't remember who thought of it, to give them credit.)

Make a deck of 18 cards from the following:
3-2's
4-3's
4-4's
4-5's
3-6's

Shuffle and then:

Easy: Deal them out into six piles of three cards each. Those are your stats, arrange to suit.

Rough: Same, except the stacks come in order.

Option to make it easier on folks: you may swap one pair of cards between two stacks. (Makes it very likely to get a 16+)

If you want to be super harsh, and replicate 3d6, use a deck of 18 that has three of each number from 1-6.
That was one of the arrays for cards I suggested... :D

When I was playing around with the Card method, I used more 3's and 4's than 2's and 5's (2:4, 3:5, 4:5, 5:4). But, the nice thing about it is you can tweak the deck (as mentioned). One variant I liked which gave slightly better scores (but nothing crazy) was 2:3, 3:5, 4:5, 5:5 (avg 11) or 2:3, 3:4, 4:4, 5:4, 6:3 (avg 12) if you want an 18 to be possible.
 

Ratskinner

Adventurer
More commonly, Fighter's grasped up the Gauntlets of Ogre Power ASAP!
Yeah, but that runs counter to your earlier claim:

The reason 3d6 worked okay in earlier editions, such as AD&D, was because ability scores weren't tied into things as invasive as they are since 3E and 5E in particular.<snip>
If Ability Scores weren't as important as they are now, then fighters wouldn't be as desperate for gauntlets or belts. (My experience matches yours, there.)
 
This is precisely the opposite of my experience. . I don't know that I've ever seen a single-class AD&D fighter without exceptional strength...
It is, in essence, a a significant class feature gated behind the highest stat requirement in the game.
You may have also noticed a lot of 16 Cons among non-fighters as only fighters got the hp bonus for 17 or 18.

But, it's also true that if you didn't have huge STR as a fighter you were probably first in line for the Gauntlet or Girdle that just /replaced/ your STR score. The more Monty Haul your DM, the less the point of investing in STR.

If Ability Scores weren't as important as they are now, then fighters wouldn't be as desperate for gauntlets or belts. (My experience matches yours, there.)
Rolled ability scores weren't as important, because items could replace or boost them.

STR, as above, could be augmented by multiple items, an 18 DEX bestowed by one of two, books and wishes and random magical scenery could mess with your stats.



::sigh:: fun times
 
Last edited:

clearstream

Explorer
Well, I am glad to help and find it interesting. Point-buy is perfect for MAD classes as you can build a character with three good abilities and no penalties. You can draw a three score decent MAD character with cards, but because it is balanced out, you will have two penalties.
As I dig into this, I feel more sure race needs to be taken into account. Envision -

Cards-3 13, 13, 13, 11, 7, 6
Points-3 14, 14, 14, 10, 10, 10
Cards-2 15, 14, 10, 9, 9, 6

We choose monk for cards-3 and points-3, and fighter for cards-2, thus -

Cards-3 + Human 12, 14, 14, 7, 14, 8
Points + Wood elf 10, 16, 14, 10, 15, 10
Cards-2 + Mountain dwarf 17, 10, 16, 9, 9, 6 assuming fighter

Cards-2 has 1 better attack and damage, 1 more hit point per die, 4 better ac against cards-3.
Cards-2 has 1 more hit point per die, and 3 better ac against points-3.
Points-3 has 1 better ac, attack and damage over cards-3, and that is intentional: cards should be worse than points-buy, for my harder-difficulty campaign.

Cards-3 + Human + ASI 12, 16, 14, 7, 14, 8
Points + Wood elf + ASI 10, 18, 14, 10, 15, 10
Cards-2 + Mountain dwarf + ASI 17, 10, 16, 9, 9, 6

Adding an ASI closes the ac gaps by 1 (plate doesn't improve, the bonus from Dexterity does), otherwise the picture remains much the same.

I'm confident that the difference between cards and points-buy for monk equates to the intended harder-difficulty for my campaign. In choosing my card mix I set the baseline ability scores deliberately lower than points-buy.

I believe about 3 points of ac is intended class discrepancy (fighters are intended to have higher ac than monks). Cards-3's net modifiers are 3 better than cards-2, meaning they will be more versatile with better saving throws. On the other hand, they have taken a knock of 1 worse across salient combat abilities (their initiative is significantly better, but everything else is worse).

On balance, I'm very comfortable with where that lands. I expected MAD classes to be 1 worse and they are: this is well inside the bounds of playing style to cope with and any overshadowing will be down to that (and to other finesses the players might think of). I'm also comfortable that ideal MAD characters should be scarce: that will reflect the positioning of those classes in the world narrative.
 

clearstream

Explorer
Don't suppose you know where/who it came from originally. I first saw it several years ago.
I didn't originate it. I first heard of it from dnd4vr and others in a previous thread.

It seems likely it has been invented and reinvented more than once over the decades since RPG entered gaming culture, but I don't think anyone was claiming it :)
 

dnd4vr

Adventurer
I didn't originate it. I first heard of it from dnd4vr and others in a previous thread.

It seems likely it has been invented and reinvented more than once over the decades since RPG entered gaming culture, but I don't think anyone was claiming it :)
Yeah, using cards to replace dice has been around for decades. You just take Ace through Six for each suite, separate into piles, shuffle, and draw one from each pile.
 

dnd4vr

Adventurer
As I dig into this, I feel more sure race needs to be taken into account. Envision -

Cards-3 13, 13, 13, 11, 7, 6
Points-3 14, 14, 14, 10, 10, 10
Cards-2 15, 14, 10, 9, 9, 6

We choose monk for cards-3 and points-3, and fighter for cards-2, thus -

Cards-3 + Human 12, 14, 14, 7, 14, 8
Points + Wood elf 10, 16, 14, 10, 15, 10
Cards-2 + Mountain dwarf 17, 10, 16, 9, 9, 6 assuming fighter

Cards-2 has 1 better attack and damage, 1 more hit point per die, 4 better ac against cards-3.
Cards-2 has 1 more hit point per die, and 3 better ac against points-3.
Points-3 has 1 better ac, attack and damage over cards-3, and that is intentional: cards should be worse than points-buy, for my harder-difficulty campaign.

Cards-3 + Human + ASI 12, 16, 14, 7, 14, 8
Points + Wood elf + ASI 10, 18, 14, 10, 15, 10
Cards-2 + Mountain dwarf + ASI 17, 10, 16, 9, 9, 6

Adding an ASI closes the ac gaps by 1 (plate doesn't improve, the bonus from Dexterity does), otherwise the picture remains much the same.

I'm confident that the difference between cards and points-buy for monk equates to the intended harder-difficulty for my campaign. In choosing my card mix I set the baseline ability scores deliberately lower than points-buy.

I believe about 3 points of ac is intended class discrepancy (fighters are intended to have higher ac than monks). Cards-3's net modifiers are 3 better than cards-2, meaning they will be more versatile with better saving throws. On the other hand, they have taken a knock of 1 worse across salient combat abilities (their initiative is significantly better, but everything else is worse).

On balance, I'm very comfortable with where that lands. I expected MAD classes to be 1 worse and they are: this is well inside the bounds of playing style to cope with and any overshadowing will be down to that (and to other finesses the players might think of). I'm also comfortable that ideal MAD characters should be scarce: that will reflect the positioning of those classes in the world narrative.
The problem with your analysis is both Cards sets have a 6. That is not likely really, but needed if you want to have higher scores elsewhere. Since it is without replacement, having a 13 in one score makes it less likely to have a 13 in another. A MAD character in the cards method with high stats will have low ones as well. That is the difference between cards, and point-buy/4d6k3. The other systems don't have to have low stats to have high ones. What you call the ideal MAD character wouldn't be ideal to many tables cause they necessitate low scores (Cards-3 has a 7 and 8, Card-2 has 9, 9, and 6).

Either way, as long as you are happy with the system, it is all moot. :) Since any class can be played with any stats, you can make a Paladin with {9, 11, 10, 13, 9, 11}, even if it might better suited to a Wizard.
 

clearstream

Explorer
The problem with your analysis is both Cards sets have a 6. That is not likely really, but needed if you want to have higher scores elsewhere. Since it is without replacement, having a 13 in one score makes it less likely to have a 13 in another. A MAD character in the cards method with high stats will have low ones as well. That is the difference between cards, and point-buy/4d6k3. The other systems don't have to have low stats to have high ones. What you call the ideal MAD character wouldn't be ideal to many tables cause they necessitate low scores (Cards-3 has a 7 and 8, Card-2 has 9, 9, and 6).

Either way, as long as you are happy with the system, it is all moot. :) Since any class can be played with any stats, you can make a Paladin with {9, 11, 10, 13, 9, 11}, even if it might better suited to a Wizard.
On Int and Cha? As dump stats those are typically not too problematic. At high-levels there are some nasty Int-save based spells, and Cha-save effects that the cards monk will fear until they get diamond soul. The cards fighter will share that fear. Optimisers who get to allocate at will traditionally dump some abilities because they don't have much mechanical relevance.

This is where I find other analysis is failing me a bit. Mechanically, an awful lot of the time a 15 looks like a 14. And a 6 only looks different from a 10 if a character needs to make a check against it. I changed my mind about points-buy based on your analysis, but I believe you are overstating the consequential differences between the arrays, after race. Perhaps it's just that our experiences of play meaningfully differ in that regard?
 

clearstream

Explorer
Either way, as long as you are happy with the system, it is all moot. :) Since any class can be played with any stats, you can make a Paladin with {9, 11, 10, 13, 9, 11}, even if it might better suited to a Wizard.
This also is true (and can get a bit lost in the optimisation discussion). It seems to me that the very definition of ideal only holds meaning within a mechanical context.
 

dnd4vr

Adventurer
On Int and Cha? As dump stats those are typically not too problematic. At high-levels there are some nasty Int-save based spells, and Cha-save effects that the cards monk will fear until they get diamond soul. The cards fighter will share that fear. Optimisers who get to allocate at will traditionally dump some abilities because they don't have much mechanical relevance.

This is where I find other analysis is failing me a bit. Mechanically, an awful lot of the time a 15 looks like a 14. And a 6 only looks different from a 10 if a character needs to make a check against it. I changed my mind about points-buy based on your analysis, but I believe you are overstating the consequential differences between the arrays, after race. Perhaps it's just that our experiences of play meaningfully differ in that regard?
Of course dump stats are only problems if you are are "attacked" in their regard, otherwise they are inconsequential, especially if you don't have skills that are penalized under them.

There really isn't much difference between a +1 mod and +2 or -2 and 0. Even when making a check against a 6, the -2 only means an additional 10% of failure. Respectively, it can be a lot! If you need a 18, but the -2 penalty makes it so you need a 20, your odds of failure increase by 200% (15% vs only 5%). That is 6.67:1 becomes 20:1.

So, the consequential difference involves a lot of factors and a CHA 6 might never hurt the character at all. In that respect, the cards system can work as well as any other provided the player (and character) can live with the potential weakness.
 

dnd4vr

Adventurer
Don't suppose you know where/who it came from originally. I first saw it several years ago.
Well, of course I wouldn't be surprised if you saw it someplace else at some time in the past, but it was an array I came up with--I didn't see it anywhere before but I am not surprised it is not "originally" by me LOL!
 

FrogReaver

Adventurer
Why I like rolling: It allows me to create new and interesting characters that aren't really possible to make under point buy and array style methods.

What I dislike about rolling is it can cause some pretty big swings in character capabilities from one PC to the next.

So my question about the cards method is: Compared with rolling how many interesting stat combinations does it take away. Also compared with rolling how consistently strong are the characters it produces.
 

clearstream

Explorer
Why I like rolling: It allows me to create new and interesting characters that aren't really possible to make under point buy and array style methods.

What I dislike about rolling is it can cause some pretty big swings in character capabilities from one PC to the next.

So my question about the cards method is: Compared with rolling how many interesting stat combinations does it take away. Also compared with rolling how consistently strong are the characters it produces.
That's determined by the card mix. Some features of my mix are
  • ability scores range 6-15
  • only one 15 is possible, and only one 6
  • only two 14s are possible, and only two 7s
  • the sum of scores is always 63 matching (but not the same as) the 10.5 average on 3d6
  • every score and combination within the range and summing to 63 is possible
It would be easy to tailor the mix to suit what you want. Another approach I thought of this morning was to have 54 cards, each naming one ability (so 9 "strength" cards, 9 "dexterity" etc). Set each initial ability at a score of 6, and then shuffle and draw (without replacement) 27 cards. Each card drawn for an ability increases it by 1.

Some features of that system
  • the maximum is 15 and the minimum is 6
  • three 15s are possible, but then the other scores will be 6
  • the average draw should increase each ability by 4.5 i.e. they become 10.5
  • every score and combination within the range and summing to 63 is possible (and more possible than in my current mix)
One could make the baseline 8 and use a deck of 42 cards, 7 for each ability, and drawing 21 of them. That would look a lot like points-buy. That pushes scores on average higher, to 11.5. One nice thing about this approach is that it is easier to understand the probabilities. A possible downside is it requires more cards.

The motive for using cards without replacement is similar to that of points-buy - a high score in one place entails a low score in another, or a character might have all scores at whatever is average for the system. (Cards with replacement are essentially custom dice, it is without replacement that matters.) Unlike points-buy, it takes players down unexpected paths, not wholly of their own choosing :)
 

Advertisement

Top