Character creation idea-rolling the dice [Archive] - Wizards Community

Character creation idea-rolling the dice [Archive] - Wizards Community

Post/Author/DateTime Post
Psionx
10-03-06, 02:19 AM roll 24d6 times and record each number that has been rolled and how many times each die was rolled. The creator may then distribute those numbers as they choose for an ability which almost guarantees an 18 in something, but at a cost to another score and the dice pool. For higher up games, 30d6 may be rolled and the dice results are pooled and distributed.

How does this idea sound?
Psionx
10-03-06, 01:17 PM :bump: 20 views and no replys? Is my proposition truly so unholy?
Toloran
10-03-06, 01:50 PM Sorry, but WAY too complicated (that and there doesn't seem to be any particular benefit of it to make the extra work worth it).
Psionx
10-03-06, 02:12 PM Ah but there's an easy way
1- follow the link: http://d20srd.org/diceBag.htm
2- set # of d6 to 30
3- click on the d6
4- copy the numbers into notepad
5- systematically delete all the numbers and tally each: delete all the 6's and count while you do it
6- record the numbers
7- distribute them as you wish

Is this technique for character stat generation balanced?
Paxar
10-03-06, 02:19 PM Just to try it out, I used your method and these are my stat results:

18
17
17
17
15
12

That's using 30d6
Psionx
10-03-06, 02:21 PM Remember thought that that is for a high level campaign and is the equivalent of the 5d6 (minus the reroll of course)
so do those stats work for a high powered campaign type character?
Toloran
10-03-06, 02:22 PM Ah but there's an easy way
1- follow the link: http://d20srd.org/diceBag.htm
2- set # of d6 to 30
3- click on the d6
4- copy the numbers into notepad
5- systematically delete all the numbers and tally each: delete all the 6's and count while you do it
6- record the numbers
7- distribute them as you wish

Is this technique for character stat generation balanced?

Thats still way to complicated for what its worth. Most of the time, if players are going to be rolling their stats, they aren't going be using a computer to do it: They'll be using dice. Granted, your system may (I'm too lazy to do that math) be more balanced but it will take far too long even using the computer. You are better off just using the "normal" rolling systems or point buy.
Psionx
10-03-06, 02:25 PM If you have 30 d6's at home you could put em' all in a sock and dump em out then pair up the results lolz! Of course you need to have taken the lord of the dice feat at 1st level:D .
KShaver
10-03-06, 03:30 PM If you have 30 d6's at home you could put em' all in a sock and dump em out then pair up the results lolz! Of course you need to have taken the lord of the dice feat at 1st level:D .

Or just proficiency in Warhammer or Shadowrun.
pinkbunny
10-03-06, 05:55 PM I would give 20-22d6, as you have no chance of low rolls coinciding on one set. if you roll the same number of dice, you are effectively removing all low rolls. If you roll less it give you maximum customization, with similar results
Amnarii
10-04-06, 12:07 AM I think for this method, a standard 18d6 would be best. Trying out that method 10 times, to get a bit of an 'average' result, yielded the following 'rolls' (always taking the the highest 3 dice left for each 'roll')

4,2,5,6,4,4,4,1,1,5,6,5,2,6,4,5,3,3 = 18, 15, 14, 12, 8, 4
5,6,6,6,5,5,3,2,5,3,5,4,6,4,2,6,1,6 = 18, 18, 15, 14, 10, 5
5,2,1,1,1,2,6,3,1,6,4,2,5,3,4,2,4,2 = 17, 13, 10, 6, 5, 3
2,4,6,6,4,5,1,1,2,3,2,1,1,2,6,6,6,4 = 18, 17, 12, 7, 5, 3
1,6,3,6,6,4,3,6,5,3,3,3,6,2,1,4,2,2 = 18, 17, 11, 9, 7, 4
1,1,3,2,1,6,3,6,3,6,6,3,1,5,4,4,4,4 = 18, 15, 12, 9, 6, 3
2,4,6,6,2,6,5,4,4,6,1,6,5,4,3,6,4,5 = 18, 18, 15, 12, 11, 5
5,2,3,2,3,3,6,3,1,6,3,4,3,6,4,3,5,2 = 18, 14, 10, 9, 8, 5
3,3,3,3,6,4,3,6,4,3,5,3,5,5,6,3,6,3 = 18, 16, 14, 9, 9, 9
2,5,2,3,2,5,4,6,1,6,1,5,5,6,6,4,4,2 = 18, 16, 14, 11, 6, 4

So out of 10 sets, there were 11 18's; an average of 1.1 per set. There were 3 3's; an average of 0.3 per set.

Try it in 'ranges'; from 3-6 there were 14 or 1.4 per set.
From 7 to 9 there were 10, or 1.0 per set.
From 10 to 12, there were 10, or 1.0 per set.
Counting 13 to 15 we have 10, again 1.0 per set.
Moving to the final 16 to 18, there are 16, or 1.6 per set.

This indicates a fairly average distribution overall; I am sure it would even out a little more over a higher sample set - something like an even 1.5 low, 1.5 high, and 1.0 for all the middle. That meas you have to suck up some really low scores for the really high scores. Or, take one really high score and just balance the rest; you could end up with very tats under 8 if you juggled those rolls right.

I think it works fine, and if you and your group are willing to spend an extra minute (or ten, for some people), it is fine to use.
Psionx
10-04-06, 02:35 AM thx, I feel that this system allows for alot of flexibility as well as luck so thats why I suggested it, just wasn't sure exactly how many to roll.
Amnarii
10-10-06, 10:59 PM Just for fun and out of sheer boredom, I went ahead and ran the test (above) several more times. I made 1,000 sets of 6 numbers; each set of 6 numbers derived from 18d6 rolls and using the top 3 dice for each number...

And, I came up with the following (roughly balanced) averages:

3/4: 811 (.8 per set)
5/6: 738 (.7 per set)
7/8: 747 (.7 per set)
9/10: 803 (.8 per set)
11/12: 731 (.7 per set)
13/14: 697 (.7 per set)
15/16: 666 (.7 per set)
17/18: 807 (.8 per set)

(note that I made a slight error in balancing the above groupings, as the first group was 4 numbers [3-6] and each following group was 3 numbers [4-9, 10-12, 13-15, 16-18]...that is why I am using groups of 2 now)

I started to try the same using a few extra dice; first 20d6 and then 22d6; I was planning on moving up to 24d6. Unfortunately I lost the spreadsheet I was using in the midst of this and only managed to save 250 of the 20d6 and 250 of the 22d6, none of the 24d6. Anyway, the averages seem to jump disproportionately as you add 2d6; dropping out the lower (3-6) and focusing more on the upper (15-18). I would say 20d6 would work for a high powered game, 22d6 is a bit much and 24d6 is probably right out for any game (unless you want your players to have 3 18's and nothing under a 10).
Mitchell
10-10-06, 11:39 PM I used something like this (a full 24d6 were too much, imo, given the extra flexibility), but I won't do it again.
My thought was noble, I believed. let my players have a good, solid, high number or two. They immediately decided that a 3 cha was okay, a 6 wisdom wouldn't hurt, etc.
If they wanted a fighter type, it was all about the highest possible physical numbers, and everything else didn't matter. if they were casters, then it was caster stat+. The charismas were so low, I had to wonder why they would group up with each other... All I could think of is that no one else would hang out with them.
They weren't characters anymore, they were caricatures. The strong, tough, somewhat nimble fighter with an IQ comparable to their unintelligent weapon. No thanks.
The only way I would do this again would be to install floors on stats, forcing them to use some mid-range numbers in what they consider to be non-essential stats.
Jo_Pistachio
10-10-06, 11:51 PM So make minimum scores of 8 or so, like in points buy.

Course, the problem comes with people like me, since I can roll 25 1s with 30 dice. :rolleyes:
Sorti
10-11-06, 01:19 PM In a high-powered campaign I DM, we used the following method:
Roll 9d6 and write every result. Every stat starts at 9.
Assign 1 dice to each of 3 stats and 2 dices to each of the other 3 stats. Results above 18 are 18.
It works quite fine.
zioth
10-11-06, 05:22 PM I wrote up a quick program to calculate the results of your method. Here they are:

Average Avg # Avg of
Dice Attribute of 18s 3 lowest stats
18 10.52 2.02 6.31
20 11.29 2.33 5.88
22 11.98 2.68 5.07
24 12.61 3.01 5.21
26 13.11 3.34 4.97
28 13.55 3.68 4.74
30 13.90 3.99 4.53
32 14.22 4.31 4.37
34 14.53 4.65 4.25
36 14.79 5.01 4.09
38 15.01 5.30 3.98
40 15.24 5.65 3.87

So, while stats aren't ridiculously high on average, the number of potential 18s is. Using the 30d6 method, min-maxers will have an average of four 18s. Using your low-end value of 24d6, you can have three 18s. Those same min-maxers could have three stats in the 5-6 range, making their characters even more silly.
Max Octavious Xavier
10-11-06, 05:52 PM roll 24d6 times and record each number that has been rolled and how many times each die was rolled. The creator may then distribute those numbers as they choose for an ability which almost guarantees an 18 in something, but at a cost to another score and the dice pool. For higher up games, 30d6 may be rolled and the dice results are pooled and distributed.

How does this idea sound?
24d6 is too many, do you really think 4d6 drop zero is a good idea? That is an average of 14 per stat!
Try:
roll 18d6 (for high power, try 20d6 drop 2 low)
divide them into 3d6 per stat

Also, this sort of idea gets posted every few weeks. That may be why it didn't get a quick response. To bad there isn't a good search function.

Post/Author/DateTime	Post
Psionx 10-03-06, 02:19 AM	roll 24d6 times and record each number that has been rolled and how many times each die was rolled. The creator may then distribute those numbers as they choose for an ability which almost guarantees an 18 in something, but at a cost to another score and the dice pool. For higher up games, 30d6 may be rolled and the dice results are pooled and distributed. How does this idea sound?
Psionx 10-03-06, 01:17 PM	:bump: 20 views and no replys? Is my proposition truly so unholy?
Toloran 10-03-06, 01:50 PM	Sorry, but WAY too complicated (that and there doesn't seem to be any particular benefit of it to make the extra work worth it).
Psionx 10-03-06, 02:12 PM	Ah but there's an easy way 1- follow the link: http://d20srd.org/diceBag.htm 2- set # of d6 to 30 3- click on the d6 4- copy the numbers into notepad 5- systematically delete all the numbers and tally each: delete all the 6's and count while you do it 6- record the numbers 7- distribute them as you wish Is this technique for character stat generation balanced?
Paxar 10-03-06, 02:19 PM	Just to try it out, I used your method and these are my stat results: 18 17 17 17 15 12 That's using 30d6
Psionx 10-03-06, 02:21 PM	Remember thought that that is for a high level campaign and is the equivalent of the 5d6 (minus the reroll of course) so do those stats work for a high powered campaign type character?
Toloran 10-03-06, 02:22 PM	Ah but there's an easy way 1- follow the link: http://d20srd.org/diceBag.htm 2- set # of d6 to 30 3- click on the d6 4- copy the numbers into notepad 5- systematically delete all the numbers and tally each: delete all the 6's and count while you do it 6- record the numbers 7- distribute them as you wish Is this technique for character stat generation balanced? Thats still way to complicated for what its worth. Most of the time, if players are going to be rolling their stats, they aren't going be using a computer to do it: They'll be using dice. Granted, your system may (I'm too lazy to do that math) be more balanced but it will take far too long even using the computer. You are better off just using the "normal" rolling systems or point buy.
Psionx 10-03-06, 02:25 PM	If you have 30 d6's at home you could put em' all in a sock and dump em out then pair up the results lolz! Of course you need to have taken the lord of the dice feat at 1st level:D .
KShaver 10-03-06, 03:30 PM	If you have 30 d6's at home you could put em' all in a sock and dump em out then pair up the results lolz! Of course you need to have taken the lord of the dice feat at 1st level:D . Or just proficiency in Warhammer or Shadowrun.
pinkbunny 10-03-06, 05:55 PM	I would give 20-22d6, as you have no chance of low rolls coinciding on one set. if you roll the same number of dice, you are effectively removing all low rolls. If you roll less it give you maximum customization, with similar results
Amnarii 10-04-06, 12:07 AM	I think for this method, a standard 18d6 would be best. Trying out that method 10 times, to get a bit of an 'average' result, yielded the following 'rolls' (always taking the the highest 3 dice left for each 'roll') 4,2,5,6,4,4,4,1,1,5,6,5,2,6,4,5,3,3 = 18, 15, 14, 12, 8, 4 5,6,6,6,5,5,3,2,5,3,5,4,6,4,2,6,1,6 = 18, 18, 15, 14, 10, 5 5,2,1,1,1,2,6,3,1,6,4,2,5,3,4,2,4,2 = 17, 13, 10, 6, 5, 3 2,4,6,6,4,5,1,1,2,3,2,1,1,2,6,6,6,4 = 18, 17, 12, 7, 5, 3 1,6,3,6,6,4,3,6,5,3,3,3,6,2,1,4,2,2 = 18, 17, 11, 9, 7, 4 1,1,3,2,1,6,3,6,3,6,6,3,1,5,4,4,4,4 = 18, 15, 12, 9, 6, 3 2,4,6,6,2,6,5,4,4,6,1,6,5,4,3,6,4,5 = 18, 18, 15, 12, 11, 5 5,2,3,2,3,3,6,3,1,6,3,4,3,6,4,3,5,2 = 18, 14, 10, 9, 8, 5 3,3,3,3,6,4,3,6,4,3,5,3,5,5,6,3,6,3 = 18, 16, 14, 9, 9, 9 2,5,2,3,2,5,4,6,1,6,1,5,5,6,6,4,4,2 = 18, 16, 14, 11, 6, 4 So out of 10 sets, there were 11 18's; an average of 1.1 per set. There were 3 3's; an average of 0.3 per set. Try it in 'ranges'; from 3-6 there were 14 or 1.4 per set. From 7 to 9 there were 10, or 1.0 per set. From 10 to 12, there were 10, or 1.0 per set. Counting 13 to 15 we have 10, again 1.0 per set. Moving to the final 16 to 18, there are 16, or 1.6 per set. This indicates a fairly average distribution overall; I am sure it would even out a little more over a higher sample set - something like an even 1.5 low, 1.5 high, and 1.0 for all the middle. That meas you have to suck up some really low scores for the really high scores. Or, take one really high score and just balance the rest; you could end up with very tats under 8 if you juggled those rolls right. I think it works fine, and if you and your group are willing to spend an extra minute (or ten, for some people), it is fine to use.
Psionx 10-04-06, 02:35 AM	thx, I feel that this system allows for alot of flexibility as well as luck so thats why I suggested it, just wasn't sure exactly how many to roll.
Amnarii 10-10-06, 10:59 PM	Just for fun and out of sheer boredom, I went ahead and ran the test (above) several more times. I made 1,000 sets of 6 numbers; each set of 6 numbers derived from 18d6 rolls and using the top 3 dice for each number... And, I came up with the following (roughly balanced) averages: 3/4: 811 (.8 per set) 5/6: 738 (.7 per set) 7/8: 747 (.7 per set) 9/10: 803 (.8 per set) 11/12: 731 (.7 per set) 13/14: 697 (.7 per set) 15/16: 666 (.7 per set) 17/18: 807 (.8 per set) (note that I made a slight error in balancing the above groupings, as the first group was 4 numbers [3-6] and each following group was 3 numbers [4-9, 10-12, 13-15, 16-18]...that is why I am using groups of 2 now) I started to try the same using a few extra dice; first 20d6 and then 22d6; I was planning on moving up to 24d6. Unfortunately I lost the spreadsheet I was using in the midst of this and only managed to save 250 of the 20d6 and 250 of the 22d6, none of the 24d6. Anyway, the averages seem to jump disproportionately as you add 2d6; dropping out the lower (3-6) and focusing more on the upper (15-18). I would say 20d6 would work for a high powered game, 22d6 is a bit much and 24d6 is probably right out for any game (unless you want your players to have 3 18's and nothing under a 10).
Mitchell 10-10-06, 11:39 PM	I used something like this (a full 24d6 were too much, imo, given the extra flexibility), but I won't do it again. My thought was noble, I believed. let my players have a good, solid, high number or two. They immediately decided that a 3 cha was okay, a 6 wisdom wouldn't hurt, etc. If they wanted a fighter type, it was all about the highest possible physical numbers, and everything else didn't matter. if they were casters, then it was caster stat+. The charismas were so low, I had to wonder why they would group up with each other... All I could think of is that no one else would hang out with them. They weren't characters anymore, they were caricatures. The strong, tough, somewhat nimble fighter with an IQ comparable to their unintelligent weapon. No thanks. The only way I would do this again would be to install floors on stats, forcing them to use some mid-range numbers in what they consider to be non-essential stats.
Jo_Pistachio 10-10-06, 11:51 PM	So make minimum scores of 8 or so, like in points buy. Course, the problem comes with people like me, since I can roll 25 1s with 30 dice. :rolleyes:
Sorti 10-11-06, 01:19 PM	In a high-powered campaign I DM, we used the following method: Roll 9d6 and write every result. Every stat starts at 9. Assign 1 dice to each of 3 stats and 2 dices to each of the other 3 stats. Results above 18 are 18. It works quite fine.
zioth 10-11-06, 05:22 PM	I wrote up a quick program to calculate the results of your method. Here they are: Average Avg # Avg of Dice Attribute of 18s 3 lowest stats 18 10.52 2.02 6.31 20 11.29 2.33 5.88 22 11.98 2.68 5.07 24 12.61 3.01 5.21 26 13.11 3.34 4.97 28 13.55 3.68 4.74 30 13.90 3.99 4.53 32 14.22 4.31 4.37 34 14.53 4.65 4.25 36 14.79 5.01 4.09 38 15.01 5.30 3.98 40 15.24 5.65 3.87 So, while stats aren't ridiculously high on average, the number of potential 18s is. Using the 30d6 method, min-maxers will have an average of four 18s. Using your low-end value of 24d6, you can have three 18s. Those same min-maxers could have three stats in the 5-6 range, making their characters even more silly.
Max Octavious Xavier 10-11-06, 05:52 PM	roll 24d6 times and record each number that has been rolled and how many times each die was rolled. The creator may then distribute those numbers as they choose for an ability which almost guarantees an 18 in something, but at a cost to another score and the dice pool. For higher up games, 30d6 may be rolled and the dice results are pooled and distributed. How does this idea sound? 24d6 is too many, do you really think 4d6 drop zero is a good idea? That is an average of 14 per stat! Try: roll 18d6 (for high power, try 20d6 drop 2 low) divide them into 3d6 per stat Also, this sort of idea gets posted every few weeks. That may be why it didn't get a quick response. To bad there isn't a good search function.