Ability Score Rolling/Point Buy Hybrid

[Guest 6801328]

I originally posted this back on the WotC forums (RIP) but thought I'd share it here.

I'm trying to address the downsides I see in the two main approaches, rolling dice vs. point buy/standard array.

4d6-1 Dice Rolling "suffers" (depending your point of view) from being too variable. You can easily end up with a big disparity between characters, and because you can arrange your scores you still get "dump stats".

And the problem with Point Buy/Standard Array is that you end up with cookie cutter characters with predictable dump stats. You almost never get a highly intelligent fighter, or a strong wizard.

Yes, I know there are some/many of you already typing "Not in my campaign! We have...etc." but in my experience, and from reading these forums, my impression is that most players have noticed both of these phenomena and to some extent don't like them.

So here's my proposal:

1. Roll 3d6 for each stat, in order.
2. Now spend 12 points to increase those ability scores using the normal costs for Point Buy. (With a cost of 1/point up to 8.)

Mathematically it ends up being roughly equivalent to Point Buy/Standard Array. The average of 3d6 is 10.5, so 3 10's and 3 11's, which starting from all 8's would cost exactly 15 points, leaving 12 points.

(4d6-1 averages 12.5, which is like 3 12's and 3 13's, which would cost 27 points via Point Buy.)

It's possible you would have a tough time building your character concept if you happened to roll badly on the key abilities. But even if you rolled, say, a 6 you could get to a solid 14 by spending 9 of your 12 points.

I played around building some example characters and it results in some quirky stat arrays without the predictability of Point Buy. You get some intelligent fighters and strong wizards.

And of course you could tweak it in any number of ways if you wanted more powerful or more customizable characters

• More/fewer points
• 4d6-1, but still in order
• Roll two arrays and pick your favorite

But I'd stick with the "in order" requirement to avoid the predictability of dump stats.

Thoughts?

 

[Cap'n Kobold]

It can depend on how the player decides on their character for this particular game. Some people start with a specific concept that they'd like to play. (Such as a high-charisma Barbarian who is effective in combat but insists on posing and trying to look good in every situation.)
Some decide based on rounding out or optimising the party. ("We don't have a spellcaster? I'll play a Wizard then.")
In those situations, being able to assign abilities is important.

Other times people might not have a definite vision or preference and are willing to see where the dice fall and work from there. That is where randomly assigned scores might be useful to give a starting basis.

This doesn't just vary from person to person, it can often be the same person having different preferences for different games.

 

[Blue]

Not 100% sure what you are saying. I think you are saying if I roll a 13 and want to bring it to a 15, that's (9 - 5 = 4) 4 points from point buy. Is that it?

If I roll a 5, what's the cost to bring it to a 10? How about just up it to a 6?

But really, my biggest problem is since the point buy can't get over 15 but you can roll over 15 that you can still get a lot of swinginess. Think like this - roll one 18, and you can start with a 20 (+5 modifier) after racial while everyone is starting at most 17 (+3 modifier). Plus you're not spending any of your point there, so the REST of your stats will be better because you have more points to spend on them. So you get double rewards for a single good roll. Chance of a 16-18 on 3d6 is 1/6. Chance that you will have one in 6 rolls is 1-(5/6)^6, about 2/3 of the time. So a good chunk of your group will have a better stat then the rest can buy and get to focus their points on just a few remaining stats (most characters need only 2-3 good stats). Yes, you'll also have really low, but since we don't know how you value raising those it seems like it will be either really cheap, or just used for a dump stat.

This seems to be a lot more swingy then just 4d6 drop the lowest when you consider this is a distribution across a bell curve and likely variations off the average will have a big effect on costs.

 

[Xeviat]

I like this, but I think it only works if you roll characters in the direction the PHB says: stats > race > class. Most players think of class > race > abilities, from my experience. Lets see what I get for 3 example characters (and their point buy equivalent)

Str 8
Dex 9
Con 8
Int 8
Wis 13
Cha 12
Point total: 10

Str 10
Dex 9
Con 12
Int 5
Wis 4
Cha 8
Point Total: 0?

Str 8
Dex 14
Con 15
Int 12
Wis 8
Cha 9
Point Total: 21

There's still a lot of swing. If you're trying to avoid swing, I wouldn't quite use this.

What I am thinking of doing, because what seems to really make characters interesting (and still playable) is a random high stat that's not used by their class, would be to have players make their stats and then roll a 1d6 coresponding to Str/Dex/Con/Int/Wis/Cha. That stat then gets randomly bumped to 16 or gets a +2 if it's already 16. This means if it lands on a stat they already have high, it won't be a huge bonus, but if it lands on a stat they had low then they'll get a big, but less useful, bump.

Alternately, roll 3d6, calculating point buy total as you go. When you reach 27 points, you stop. If you exceed 27 points, you stop and the rest of your stats are 8. If you roll 6 and you still have points left over, you get those points to adjust.

 

[Blue]

For years (back in AD&D 2nd ed) the game I was in was that you rolled 4d6 drop the lowest in order, but you rolled three sets and picked which one. You often would get scores that might not be the class the party needed, but that lead to lots of different feel between different groups. I had the high-Int, high-Dex (and 17 min Chr) paladin. I had the 18 Con Bard (when only fighter types could benefit from more than a 16 Con, the bonus was capped) who could drink people under the table. I like non-standard characters. But in 5e where you hace a choice of advancing a stat or taking a feat, having wildly variant scores between different party members throws off the balance between them. So how can we move forward with that?

4d6-1 Dice Rolling "suffers" (depending your point of view) from being too variable. You can easily end up with a big disparity between characters, and because you can arrange your scores you still get "dump stats".

And the problem with Point Buy/Standard Array is that you end up with cookie cutter characters with predictable dump stats. You almost never get a highly intelligent fighter, or a strong wizard.

Just riffing off this, what if you used a (lower total) point buy to set a floor so that you can play the character you want? This addresses the low-side volatility which is half the 4d6 drop lowest, and the predictability of point buy.

So you pick some lesser total, like 21. Do out your stats. Then roll in order, keeping whatever is higher. 21 point buy you will not be able to get what you want so there is a "threat", but the downside is much lower than the upside and players need to think about if they want to ensure high minimum in the prime abilities for their class or if they want to ensure no penalties elsewhere.

So if you were building a paladin you might have the following

Point buy (cost)/ Actual Roll = Final result (before racial)

Str: 14 (7) / 17 = 17
Dex: 8 (0) / 9 = 9
Con: 13 (5) / 8 = 13
Int: 8 (0) / 15 = 15
Wis: 10 (2) / 13 = 13
Chr: 14 (7) / 8 = 14

Heh, I use an online dice roller for that but it's really almost your example of a smart fighter. They didn't protect dex and will have a penalty (which won't hit AC in heavy armro but will affect saves and init). They protected Chr and Con which is good because an 8 wouldn't have cut it in either location. Str they luckyed out with a 17, that could be an 18 or 19 with racial modifiers.

Point buy the final would be ... well, the 17 is off the charts to calculate. But considering how the second highest being Int you definitely have a paladin that would be unlikely to be built with point buy.

Would this fit what you are looking for?