I generally prefer point buy or array, but if random rolling: 4d6, drop lowest, arrange.

There was a really odd one I saw once, I think in Dragon Magazine, which gave you a pool of dice -- 20 or 30 or something. You'd assign a number of dice to each stat, and then roll and keep the best 3 (since you can't exceed the 18 attribute cap). Let's say your DM gives you 24d6 to work with, and you're a fighter. So you'd do this:

STR: 5d6
DEX: 4d6
CON: 5d6
INT: 3d6
WIS: 3d6
CHA: 4d6

Not sure if I like it, but it is interesting.

Quote from: Ghostmaker on July 01, 2021, 12:13:41 PM
I generally prefer point buy or array, but if random rolling: 4d6, drop lowest, arrange.

There was a really odd one I saw once, I think in Dragon Magazine, which gave you a pool of dice -- 20 or 30 or something. You'd assign a number of dice to each stat, and then roll and keep the best 3 (since you can't exceed the 18 attribute cap). Let's say your DM gives you 24d6 to work with, and you're a fighter. So you'd do this:

STR: 5d6
DEX: 4d6
CON: 5d6
INT: 3d6
WIS: 3d6
CHA: 4d6

Not sure if I like it, but it is interesting.

Definitely interesting, but ultimately all the alternate methods comes down to "how do we increase the odds of giving a player the character they actually want to play without alienating the players who want to feel special because they were lucky with the RNG?"

It's basically trying to have the cake and eat it too while still having the problem that there's enough wiggle room that if you suspect a player is a dice fudger you're still going to want them to roll the character up in front of you.

Which is one of other nice things about arrays/point buy; you don't need to take the player's word for it... the numbers either add up or they don't. Which in turn means that CharGen can be homework instead of taking up table time. Session Zero can be brainstorming ideas for PCs with actual associations and reasons for adventuring both generally and with each other and collaborating with the GM on how they fit into the GM's campaign world.

It's NOT OSR-style play, but not everything has to be.

Quote from: Zalman on July 01, 2021, 10:28:55 AM
I am noticing a lot of conflation in this thread. "3d6 down the line" has two elements to it: its stat range, and its order. It seems to me that most people who dislike this method argue against one or the other only, and yet prefer an alternative that dispenses with both. I think each element has its own advantages and disadvantages, and that they are severable.

I suspect they're being conflated because that's the usage pattern. There are some gamers (mostly old-schoolers) who use 3d6 in order; and some old-schoolers and some new-schoolers use best-3-of-4d6 and arrange (that's Method I in the original DMG).

As far as I know, I don't think anyone uses best-3-of-4d6 down the line; or roll 3d6 and arrange. Though I'd be curious if anyone has experience with those.

Quote from: Chris24601 on July 01, 2021, 02:13:53 PM

Quote from: Ghostmaker on July 01, 2021, 12:13:41 PM
I generally prefer point buy or array, but if random rolling: 4d6, drop lowest, arrange.

There was a really odd one I saw once, I think in Dragon Magazine, which gave you a pool of dice -- 20 or 30 or something. You'd assign a number of dice to each stat, and then roll and keep the best 3 (since you can't exceed the 18 attribute cap). Let's say your DM gives you 24d6 to work with, and you're a fighter. So you'd do this:

STR: 5d6
DEX: 4d6
CON: 5d6
INT: 3d6
WIS: 3d6
CHA: 4d6

Not sure if I like it, but it is interesting.

Definitely interesting, but ultimately all the alternate methods comes down to "how do we increase the odds of giving a player the character they actually want to play without alienating the players who want to feel special because they were lucky with the RNG?"

It's basically trying to have the cake and eat it too while still having the problem that there's enough wiggle room that if you suspect a player is a dice fudger you're still going to want them to roll the character up in front of you.

Which is one of other nice things about arrays/point buy; you don't need to take the player's word for it... the numbers either add up or they don't. Which in turn means that CharGen can be homework instead of taking up table time. Session Zero can be brainstorming ideas for PCs with actual associations and reasons for adventuring both generally and with each other and collaborating with the GM on how they fit into the GM's campaign world.

It's NOT OSR-style play, but not everything has to be.

Completely agree. PF2E's character generation (where every option you pick as you build your PC bumps one stat or another) is also neat.

OSR is nice, but it's not like the Ten Commandments, carved into stone and brought down the mountain by Gygax and Arneson

Quote from: jhkim on July 01, 2021, 02:37:25 PM
As far as I know, I don't think anyone uses best-3-of-4d6 down the line; or roll 3d6 and arrange. Though I'd be curious if anyone has experience with those.

I have done both  of those and run a game with them, along with too many options to list.  Though I'm not sure that exactly counts, because that was with a group that would try something different every weekend, roll up characters, get them killed, and reset the deck for trying something new next week.

Having tried nearly everything at one time or another, I'm largely in the camp of wanting one of the following:

- The characters are practically throwaway, you go through them fast, until you latch onto one that you like.  Lots of rolls spread over many characters.  The fun is dealing with it best you can.
- The characters are designed to be long term from the get go and abilities are relatively set at the beginning.  The exact method doesn't matter that much, as long as it tends to not produce a lot of outliers.  The fun is getting to do what you want.
- The characters may or may not be long term (depending on events in the game). Abilities have some way of improving that tends to even them out somewhat if the game goes long enough.  The fun is that abilities scores are just another way in which your character advances.

For the same reason, in D&D I prefer either fixed hit points OR rerolling the whole set at each level.  (At 2nd level, roll both dice. If total greater than current hit points, take the new roll.)

However, I usually don't like "carefully crafted to make everything fit exactly towards some mechanical ideal" characters, either.  So that means some kind of randomness, somewhere. Or when I do want that, I want more options than D&D will typically provide.  Might as well play Fantasy Hero or GURPS.

Quote from: jhkim on July 01, 2021, 02:37:25 PM

Quote from: Zalman on July 01, 2021, 10:28:55 AM
I am noticing a lot of conflation in this thread. "3d6 down the line" has two elements to it: its stat range, and its order. It seems to me that most people who dislike this method argue against one or the other only, and yet prefer an alternative that dispenses with both. I think each element has its own advantages and disadvantages, and that they are severable.

I suspect they're being conflated because that's the usage pattern. There are some gamers (mostly old-schoolers) who use 3d6 in order; and some old-schoolers and some new-schoolers use best-3-of-4d6 and arrange (that's Method I in the original DMG).

As far as I know, I don't think anyone uses best-3-of-4d6 down the line; or roll 3d6 and arrange. Though I'd be curious if anyone has experience with those.

Pretty much seen the same thing. I use 3d6 in order for my OD&D games, but have used 3d6 arrange for BX/LOTFP. However, after moving over to the playtest rules for lamentations, I went back to 3d6 in order. In general though, it seems 3d6 in order are married more often than not. Same with 4d6 arrange.

Quote from: Shasarak on July 01, 2021, 02:30:54 AM
I thought we agreed that 4e sucked because of the art direction? 

What part? The PHB I had a look through looked fine? Aside from some of the dragonborn looking like someone dropped an anvil on their heads... ow...

3e or 3.5 on the other hand... MM was pretty bad.

Can we all at least agree that the 12d0/1 +8 in order method is the best? 

I liked 3d6 in order, but I also like Method V.

Quote from: Zalman on July 01, 2021, 10:28:55 AM
I am noticing a lot of conflation in this thread. "3d6 down the line" has two elements to it: its stat range, and its order. It seems to me that most people who dislike this method argue against one or the other only, and yet prefer an alternative that dispenses with both. I think each element has its own advantages and disadvantages, and that they are severable.

Put me down for neither 3d6 nor down the line.

No body got time for that shit.

Quote from: Omega on July 01, 2021, 03:22:23 PM

Quote from: Shasarak on July 01, 2021, 02:30:54 AM
I thought we agreed that 4e sucked because of the art direction? 

What part?

Quote from: jhkim on July 01, 2021, 02:37:25 PM

Quote from: Zalman on July 01, 2021, 10:28:55 AM
I am noticing a lot of conflation in this thread. "3d6 down the line" has two elements to it: its stat range, and its order. It seems to me that most people who dislike this method argue against one or the other only, and yet prefer an alternative that dispenses with both. I think each element has its own advantages and disadvantages, and that they are severable.

I suspect they're being conflated because that's the usage pattern. There are some gamers (mostly old-schoolers) who use 3d6 in order; and some old-schoolers and some new-schoolers use best-3-of-4d6 and arrange (that's Method I in the original DMG).

As far as I know, I don't think anyone uses best-3-of-4d6 down the line; or roll 3d6 and arrange. Though I'd be curious if anyone has experience with those.

I primarily use Best 3 of 4d6, Straight Down the Line.  I love it.  Using it right now, by the way.  16 S, 12 D, 12 C, 9 Int, 13 W, 11 Cha.  Hello Mr. Fighter.

Quote from: oggsmash on July 01, 2021, 11:27:47 AMThe Roman army did not recruit based on the pay and retirement plan.  They sold the fact that winning armies get to LOOT!!  But, they also had some requirements for stature and build.  they were not so interested in the dude who was sickly and weak.

Yes, this is why a fighter must have at least Str 9 and Con 7 in AD&D1e.

It's not until you hit Con 6 that you get a malus to hit points - which is about when someone might perceive you as "sickly."

Strength likewise has no malus until 6, but the army would also be concerned about your being able to move equipment and your encumbrance, too. Remember that you can do a strict overhead press of 10lb for each point of Strength (with some more complex rules for percentile) which means a minimum press of 90lb. It's that sort of level you'd need to be able to put a heavy pack on your back, help move around tent poles and fortification stakes, and so on.

Quote from: Jam The MF on June 29, 2021, 07:43:08 PMYes, to an extent; but why would someone with 11, 8, 9, 7, 9, 6 ever leave the farm to go place their life in constant peril?  I understand why someone with 17, 14, 15, 11, 12, 10 would.

Obviously both of these stats are possible with just 3d6 (if somewhat unlikely.) The chances of rolling no stat (out of 6) greater than 11 are the same as rolling no stat lower than a 10. And those chances are about 6%.

On the other hand, the chances of rolling a character better than 17 14 15 11 12 10 in at least one stat with just 3d6 is about 80%, believe it or not.

Quote from: Chris24601 on July 01, 2021, 10:13:30 AMAnother thing that doesn't sit right about the randomized stats is the degree to which training plays a part. I'll never be a top tier athlete, but if I started weight and endurance training plus practicing particular sports I guarantee my Str, Dex and Con stats would be vastly better a year from now than they are presently.

In Basic, this was achieved by exchanging ability score points 2 for 1 in your prime stat. Although you could not lower a stat below 9 and you could not lower Dex, Cha, or Con.

Quote from: BasicThis is like practicing hard to
learn your Class, but at the cost of not
developing another Ability at the same
time.