If 4d6dl is the baseline, what's the PB/Array?

[delericho]

4d6 drop the lowest is equivalent to 30-31 point buy.

Huh. I was of the impression it was 28 points. Guess I was wrong.

25 or 28 is probably the best non-random equivalency in power.

Agreed (I would go with 28 points). For the array, then, I would suggest 16/15/13/12/10/8, which works out to a nice 29 points - the more control you get, the fewer points, so this fixed array is very slightly better than straight 28-point buy.

Assuming that 5e has the same "hopeless characters" clause for rolling, the same modifiers for the stats, and the same point-buy costs as 3e, of course.

Probability in Char. Gen. - Page 6 - EN World: Your Daily RPG Magazine

One of the bestest thread I ever started. Also the mathiest.

That is a fascinating thread.

 

[Gort]

Ugh, I'd hoped rolling ability scores was dead and buried, but here it comes again to blight a new edition!

 

[delericho]

Ugh, I'd hoped rolling ability scores was dead and buried, but here it comes again to blight a new edition!

Both the rolling and point-buy camps have strong opinions on this, such that I don't think there can ever be a consensus. That being the case, I'm hoping that both methods are fully supported (and reasonably equivalent - something 4e fell down on).

It's also perhaps worth noting that the rolling method is much better for new players, who then only need to understand what the six scores mean, rather than having to grasp all the intricacies of the point buy method in use.

 

[Kaodi]

Personally I would pretty much always choose point buy as my method for generating ability scores. Or, if momentarily lean towards old school, translating rolling 4d6 for targets and using point buy to get something close to that (maybe we can call this method Reliable Rolling, or Roll Modelling).

However, I could compromise my opinion on what I think should be the standard generation method if it could be show that rolling (edit: or standard array) as default is better for newer, and in particular, younger players. Because really, there are just sometimes when veterans and grognards have to suck it up for the good of the hobby. Because their preferred stat generating method will still be presented in the Core game. And because they should be best able to handle choosing their preferred option over the mystic "default" because they are grown ass adults.

 

[delericho]

However, I could compromise my opinion on what I think should be the standard generation method if it could be show that rolling (edit: or standard array) as default is better for newer, and in particular, younger players.

I would have thought this was quite easy to demonstrate.

In order to understand the rolling method, a player needs to understand the meaning of the six stats. And that's it - he then just rolls and assigns the numbers to match up with his image of the character.

In order to understand point-buy, a player needs to understand the six stats as before, but also needs to understand the non-linear relationship between the points and the stats spent. And, indeed, to be any good at doing point-buy, he then also needs to understand the relationship between the raw stat and the modifier that that gives you.

My experience with newer players is that random rolls are vastly preferable. That said, it is possible that I just haven't explained the system very well.

(It's also worth noting something that I hadn't previously considered - which is that a standard array is better still for new players, as it removes one step from the process, and neatly eliminates the 'bug' of some characters being rolled better than others.)

All that said, though, I think 4e got the general approach right - present the three methods together without comment, and let people choose what works for them. (Where 4e fell down was in making the rolling method significantly worse than point-buy, but that's something 5e can easily fix.)

 

[Kaodi]

Probability in Char. Gen. - Page 6 - EN World: Your Daily RPG Magazine

One of the bestest thread I ever started. Also the mathiest.

I do not have background in computer science, so I cannot really read the functions that you guys thought up to calculate this stuff. But I cannot help but get the feeling that there may have been a key element missing from the entire exercise: the question of what the best point buy costs for modelling 4d6 characters would be. Am I merely mistaken and such considerations were somehow included in the calculations?

 

[Kaodi]

In order to understand point-buy, a player needs to understand the six stats as before, but also needs to understand the non-linear relationship between the points and the stats spent. And, indeed, to be any good at doing point-buy, he then also needs to understand the relationship between the raw stat and the modifier that that gives you.

Good points, but I am not sure that by itself is it sufficient to prove that rolling is better, because you only have to wrestle with point buy in the first session, but you would have to deal with the consequences of sucky roles for the entire game (though it is for standard array being best). To answer reliably I think you need to identify what the "breaking point" of D&D is, and use the method that protects against that.

 

[avin]

I've used point buy for 3E and 4E and will be using point buy in 5E, no matter what the book suggests.

 

[Kaodi]

I was sort of messing around with a spreadsheet for a while, doing things the hard way because I have too little idea of how to use one properly, and I derived a point buy system that looks kind of like a skewed version of Pathfinder...

12 Pt Buy

07 -4
08 -2
09 -1
10 00
11 01
12 02
13 03
14 04
15 05
16 07
17 10
18 18

With this you could do the ol' Standard Array of 15 14 13 12 10 8; or its Eighteen Array works out to 18, 10, 10, 10, 8, 7.

 

[howandwhy99]

3d6 and keep 'em:

3 1/216 0.00463
4 3/216 0.01851
5 6/216 0.02778
6 10/216 0.04629
7 15/216 0.06944
8 21/216 0.09722
9 25/216 0.11574
10 27/216 0.125
11 27/216 0.125
12 25/216 0.11574
13 21/216 0.09722
14 15/216 0.06944
15 10/216 0.04629
16 6/216 0.02778
17 3/216 0.01851
18 1/216 0.00463

4d6 drop the lowest:

3 0.10%
4 0.30%
5 0.80%
6 1.60%
7 2.90%
8 4.80%
9 7.00%
10 9.40%
11 11.40%
12 12.90%
13 13.30%
14 12.30%
15 10.10%
16 7.30%
17 4.20%
18 1.60%

EDIT: The point buy cost matters entirely on the value of the score to a target number or, if they use ability modifiers, then the modifier distribution.