Whacked out attribute point buy

[Starman]

Originally posted by Plane Sailing

Have you considered the general problem which race adjustments provide in a point-buy scenario with differing values for different attribute scores?

I've thought about this before and I came up with a crazy idea, but I've never had a chance to test it out.

Instead of attribute bonuses, races would have favored/non-favored attributes. An elf, for example, would have Dex as a favored attribute and Con as a non-favored. Favored attributes would actually cost less to raise while non-favored would cost more.

How well do you think this would work? Like I said, the idea just kind of came to me and I haven't tested it out or anything.

Starman

 

[Plane Sailing]

That's an interesting idea!

In order to make it work well I'd guess that a point-buy system would have to be created with that in mind though. Hmmm

 

[seasong]

I don't think the advantage-based chart I wrote would be able to scale to that very well, but WotC's original one (based on the bonus) could - simply lop +1 off of the bonus when calculating cost.

So a favored attribute for a +2 would look like this:

8: 0
9: 1
10: 2
11: 3
12: 4
13: 5
14: 6
15: 7
16: 8
17: 10
18: 12
19: 15
20: 18

And a favored attribute for a +4 would look like this:

8: 0
9: 1
10: 2
11: 3
12: 4
13: 5
14: 6
15: 7
16: 8
17: 9
18: 10
19: 12
20: 14
21: 17
22: 20

A disfavored attribute (-2) could look like the chart below:

8: 0
9: 1
10: 2
11: 3
12: 4
13: 6
14: 8
15: 11
16: 14

 

[Starman]

Hmmm...that looks interesting. I might have to make up a few characters to see how it plays out.

Starman

 

[haiiro]

Hey Plane Sailing! That's at least one good reason to use the 56 points 1:1 system - no extra points, at least.

Plane Sailing's point is an excellent one, and one of the main objections I have to WotC's point buy system.

I used to use a straight (1:1) point buy system when I ran 2e, generally with 74-78 points to work with (and a minimum score of 3 in any attribute, of course). I treated racial attribute modifiers as altering the maximum points you could allocate to an attribute, not as affecting its total. For example: elves could allot 20 points to DEX, but only 16 to CON. From a mathematical/statistical standpoint, I'm not sure quite what effect this has -- but it seemed to work out fairly well in practice.

In 3e terms, this would gave races like the races like the half-orc a mechanical boost -- instead of losing 2 points after applying modifiers, they simply have one more attribute that gets capped at 16 (in the half-orc's case, at least).

Expressing a 1:1 system as "3 in each stat and xx points to spend" is much more elegant; I'll have to remember that.

Am I correct in thinking that the 56 point approach works out to a fair amount more than any of WotC's point buys? By my math, I can squeeze 18/18/10/10/10/8 out of 56 points, which is more than the 18/18/8/8/8/8 I can get from WotC's 32 point buy. Is this correct? Is 56 points equivalent to a particular level of WotC-style point buy?

seasong (or others), do you have any more general comments on the nature of the 56 point 1:1 system? In any case, thanks for a very interesting read so far -- you guys are good at this math stuff.

 

[seasong]

Am I correct in thinking that the 56 point approach works out to a fair amount more than any of WotC's point buys? By my math, I can squeeze 18/18/10/10/10/8 out of 56 points, which is more than the 18/18/8/8/8/8 I can get from WotC's 32 point buy. Is this correct? Is 56 points equivalent to a particular level of WotC-style point buy?

The 56 points is not equivalent to WotC's point buy - it's equivalent to all "average" rolls with 4d6/low (as opposed to the earlier system I proposed, which attempts to simulate the advantage gained by any particular 4d6/low roll). For example, 18/18/10/10/10/8 yields an average of 12.33 in each attribute, which is just a hair off from the 12.24 average you would get from 4d6/low. WotC's 18/18/8/8/8/8 gives an average 11.33, well below the 4d6/low average.

seasong (or others), do you have any more general comments on the nature of the 56 point 1:1 system? In any case, thanks for a very interesting read so far -- you guys are good at this math stuff.

Not really. I would use it, but if I'm going to use D&D (as opposed to GURPS), I actually prefer making my players roll .

 

[haiiro]

The 56 points is not equivalent to WotC's point buy - it's equivalent to all "average" rolls with 4d6/low (as opposed to the earlier system I proposed, which attempts to simulate the advantage gained by any particular 4d6/low roll

I understand, and I like that equivalency. But in deciding how to approach things for the next game I run, I'm comparing WotC's point buy method to the 1:1 method -- rolling for stats is already out of the pitcure, except as a useful benchmark. Based on what I've seen in this thread so far, and on my own experience with 1:1 point buys, I think that's what I'll be going with. Your 56 point system does exactly what I want it to, and it addresses the one flaw inherent in the WotC approach.

Any comments on the "racial modifiers only apply as stat caps, not as modifiers to the final totals" approach?

 

[seasong]

Racial modifiers as a stat cap only (and possibly a stat minimum) mostly kills the advantage of having a stat boost in a scaled point system (like WotC's). In the 1:1 system, it doesn't really matter either way - stat boosts or stat caps, if the plusses and minusses balance out for the race, they balance out either way.

If the race has more plusses than minusses, a "stat cap only" hurts them.

If the race has more minusses than plusses, a "stat cap only" helps them.

The only problem I can see with the 1:1 system is that it encourages specialization (taking low stats to pay for high stats), because having a +3 or +4 in your core competency is worth being a drooling idiot or clumsy weakling for, and overall you get more than you lose (since you have friends to cover for your weaknesses).

 

[haiiro]

The only problem I can see with the 1:1 system is that it encourages specialization (taking low stats to pay for high stats), because having a +3 or +4 in your core competency is worth being a drooling idiot or clumsy weakling for, and overall you get more than you lose (since you have friends to cover for your weaknesses).

I actually like encouraging this approach, as it gives players an easy jumping-off point for roleplaying their character (while preserving the option to go down the 14/12/12/12/12/12 route) and makes stats below 8 seem like a potentially interesting option. I and many of the folks I play with often tend towards that type of specialized character, so that figures into it as well.

Thanks for the confirmation on the stat caps/stat mods issue.

 

[haiiro]

One heck of a bump.

cptg1481's thread in GD about rolling, point buys and fairness (here: http://enworld.cyberstreet.com/showthread.php?s=&threadid=50541) got me thinking about this issue again.

When I looked at the 56 point 1:1 point buy system, I looked at it from the perspective of how many high stats one can purchase. Like so: 18/18/14/8/8/8, with a net bonus of +10. Compared to WotC's 32 point buy: 18/18/8/8/8/8, with a net bonus of +8. At first glance, it looks unequivocally better than 32 point buy.

When trying to model more uniform stats, however, 32 point buy can turn out the following: 14/14/14/14/12/12, with a net bonus of +10. 56 point buy (1:1) can only generate 14/12/12/12/12/12, with a net bonus of +7 -- clearly worse, and not even as good as WotC's 28 point buy (which can produce 14/14/12/12/12/12, with a net +8 bonus).

I understand that the original intent of the 56 point 1:1 system was to model the "4d6, drop the lowest" method -- which it does, and does well -- and that it isn't really intended to be compared to WotC's weighted/scaling point buy system. Nonetheless, I think that's the benchmark most detail-minded players (and DMs like me) will use for comparison -- and the inconsistency of the results of this comparison make me scratch my head.

Can anyone sort this out for me and explain it in layman's terms? Also, is there a number of points that makes a 1:1 point buy system consistently slightly better than WotC's 32 point buy -- or is that mathematically impossible?

Thanks in advance.