D&D General Ability score generation: The Snake

Andvari

Explorer
Like many others, I like rolling for ability scores while simultaneously disliking how it can generate overpowered characters, underpowered characters and very uneven adventuring parties. So I often think about generation methods that mitigate or eliminate the undesirable outcomes the process can create. Here's one such method.


The Snake
This method turns ability score generation into a group activity, where generating outlier characters is less likely and the ability score totals for each party member are more likely to be close. It can create characters with low or high scores just like the standard 4d6 method.

1. Each player makes one roll of 4d6, drop lowest, in turn, going clockwise around the table. Record the results in a single column from top to bottom.
2. Repeat above until there are six columns in total, each with one number per player. The use of columns is simply to make it easy to see when you're done.

Here's an example of what the result might look like with a group of 4 players. Player 1 rolls 8, player 2 rolls 15, player 3 rolls 16, player 4 rolls 15, player 1 rolls 12, player 2 rolls 16 etc.

8 12 18 14 10 15 15 16 13 9 17 13 16 8 8 11 14 12 15 10 12 14 11 11

3. Sort all numbers from highest to lowest.

Using the example above, we have the following order: 18, 17, 16, 16, 15, 15, 15, 14, 14, 14, 13, 13, 12, 12, 12, 11, 11, 11, 10, 10, 9, 8, 8 and 8.

4. Add each number in order, into a new set of six columns in the winding, snake-like order shown below.

1 8 9 16 17 24 2 7 10 15 18 23 3 6 11 14 19 22 4 5 12 13 20 21

Using the example above, we end up with the following six columns.

18 14 14 11 11 8 17 15 14 12 11 8 16 15 13 12 10 8 16 15 13 12 10 9

Each row is now an ability score array that can be assigned randomly to each player. Due to the snake pattern, each row is balanced by being on opposite sides of the vertical number ordering. For example, the first row has the best number in rows 1, 3 and 5, but the worst of 2, 4 and 6.

In steps 1-2, the players could in theory just roll 6 numbers each as normal and sort them into a common set. The reason for taking turns is to make it feel more like a team effort where they are all contributing, rather than rolling up separate characters and "losing" them to the common pool.

There's also the question of how to handle new characters after the initial party is created. Perhaps the players simply roll up a new set of rows, with new characters taking each set in order or randomly as they join the game.
 
Last edited:

log in or register to remove this ad



DND_Reborn

The High Aldwin
He means you zigzag vertically. First column is going from the biggest value down, then it steps to the right at the bottom and goes from bottom upwards, then it steps to the right etc
Going from this to this:
1664308464264.png

Even with your additional info makes no sense to me. Sorry. :(
 



DND_Reborn

The High Aldwin
I think it would be simpler to do this:

Group the first with the last (green)
Group the second with next to last (red)
Group the third with the third from last (blue)
Group the fourth with the fourth from last (purple)
and so on...
1664311440747.png

Then you end up with the following arrays:
1664311380692.png

With totals that are even closer.
 









Aurel Guthrie

They/Them
It seems pretty balanced, but when putting so much work into rolling for stats it's hard to justify not simply using point-buy. To me the point of rolling for stats is.. rolling your stats. Here you're rolling a stat that goes on a pool of stats to be distributed fairly, it's cool on paper but I can't see it being fun in practice because the stats you're getting are not your stats, the ones you rolled originally. If that makes sense.

In other words, to me the point of rolling for stats is the instant hit of dopamine you get out of rolling and writing the result on your sheet, which is missing here.
 

My suggestion above would have a table like this:
View attachment 262536
The problem is, this ensures that the first array always gets the best result of a given block of 4, and the fourth array always gets the worst result in that block. The ox-plow-like alternation is an effort to ensure fairness.

If you really wanted to ensure fairness you'd look for a way to distribute these things via the Thue-Morse sequence. That would give you a table like this:


---Roll #NRoll #NRoll #NRoll #NRoll #NRoll #N
Array A1811141724
Array B2512151821
Array C369161922
Array D4710132023

It's imperfect, because we don't have a clean power of 4 number of elements. But it's the best that can be done with four arrays. You'd need to do a total of six arrays if you wanted it to work out neatly (because then we'd use the first nontrivial instance.) That would look like this (skipping the labels above to make it fit in a smaller space):

11217222732
2718232833
3813242934
4914193035
51015202536
61116212631

The bottom row kinda gets shafted here 'cause it takes so long for it to be the "first" choice, but it's not the worst thing ever. The top row will likely end up with only one amazing stat (plausibly 16-18) but several middle-of-the-road stats, while I would expect the second or third array to deliver the best overall performance.

Edit: On reflection, probably not worth it. You'd need many more 'rounds' of sharing to make it actually shift toward fairness, and with stats being a highly uneven spread, the 5th and 6th arrays are almost guaranteed to be shafted, while the first three arrays are almost guaranteed to be excellent. Alas. I find the Thue-Morse sequence fascinating but it really only works for binary splits, anything bigger starts to get unwieldy.
 
Last edited:

It seems pretty balanced, but when putting so much work into rolling for stats it's hard to justify not simply using point-buy. To me the point of rolling for stats is.. rolling your stats. Here you're rolling a stat that goes on a pool of stats to be distributed fairly, it's cool on paper but I can't see it being fun in practice because the stats you're getting are not your stats, the ones you rolled originally. If that makes sense.

In other words, to me the point of rolling for stats is the instant hit of dopamine you get out of rolling and writing the result on your sheet, which is missing here.
Perhaps that explains my antipathy for rolled stats. I don't experience a dopamine rush like that. Instead I mostly feel dread and then either frustration (because my stats are way better than anyone else's and I feel like I'm going to deny them participation) or bitter disappointment (because literally the only stats I'm allowed to roll are AMAZING or absolute garbage, with nothing in-between.)
 


I feel existential angst at point-buy/standard-array, because the stats might as well be predetermined by class then.
I mean, feelings are feelings, so there's not a whole lot to discuss. But I find that very strange. With point-buy, you can choose whether to play to your strengths or diversify--and I have often chosen the latter rather than the former. For example, I like Paladins with a broad set of decent scores in 4e D&D, so I tend to dump Dex and take only 16s in my main stats, as opposed to 18s as is "standard," so I can have solid 14s in my secondary stats and at least some kind of positive modifier in Intelligence--averting the standard "Paladunce" expectation.

You can't do that with fixed stats by class. Point-buy offers freedom. Be what you want to be.
 

An Advertisement

Advertisement4

Top