D&D 5E Ability score generation: "buy your dice roll"

[toucanbuzz]

I was going through my old notes and came across this homebrew ability score generation option I used in 3E to give some variance with ability scores so everyone didn't look cookie cutter. This way, maybe there could be a wizard with an 18 STR because does every wizard have an 8-10 Strength? There isn't one wizard in the world who works out? I'm pretty sure I didn't create it, don't know who did, and was positing whether it would translate to 5E.

Buy your dice roll method:

You have 5 points. Purchase a roll for each ability score from the options below. You don't have to spend all 5 points on rolls. You can instead "reserve" any number of your 5 points to improve a score after the dice are rolled. The worse the score, the bigger the improvement.

Point Cost Method	Die Roll	Range	Average
A: 3 points.	15 + 1/2 of 1d6 (round down)	15-18	16.5
B: 2 points.	10 + 2d4	12-18	15
C: 1 point.	6 + 2d6	8-18	13
D: 0 points.	3d6	3-18	10.5

Reserve boost per point spent:

Ability Score	Boost
3-7	+3
8-13	+2
14-16	+1
17-18	+0

Optional: Mulligan / Superhero. Although unlikely that one may roll to the extremes, the DM and player should discuss a redo (“mulligan”) if no score has a positive ability modifier, or all scores have a positive ability score modifier.

Example:

Spoiler

I'm thinking about playing a wizard, so I know I'll want high intelligence, and I also decide I want a decent Con to survive. I use 3 points to buy method A for Intelligence because it gives me the highest range. I buy a Method Cs for Con at 1 point and make the rest Method D (costs me nothing) so I can save 1 point for a boost, just in case.

STR - 3d6 Rolled a 13
Dex – 3d6 Rolled a 12
Con - 6+2d6 Rolled an 8
Int - 15 + 1/2 of 1d6 (round down) Rolled a 17
Wis - 3d6 Rolled a 7
Cha - 3d6 Rolled an 11

Now I have a dilemma. My Con at 8 is poor, and the "boost" would get it up by +2 to a 10 so I'd have no penalty. Or, I could do the same to my 7 Wisdom to get it to a 10 (+3). Or, I might just boost something else! All depends on what you're playing. And, this wizard works out. He's got a 13 Strength. No more embarrassment asking others to help carry your waterskin when it's filled up.

 

[Krachek]

I though at first it was about paying the DM for better score!
I like the method, multiple kind of roll, some strategy, may help some.

 

[DND_Reborn]

It reminds me of the AD&D UA system of having 9d6, 8d6, etc. and choosing where you wanted to roll the most dice.

I've also seen system were you roll 5d6 for two scores, 4d6 for two scores, and 3d6 for two scores, only keeping 3d6 each time.

I think those systems and something like yours is a fine way to go, really. My personal preference would be to keep the die types the same, but that is a minor point.

 

[Blue]

This does scratch an itch I have without voiding two other concerns that point buy avoids. I really enjoy the Faustian bargain of ASI or feat when both of them are quite tempting. But in earlier editions, I like the organic "not quite what you expected" of rolling in order. This seems to give both of those. My last worry with rolling is large power discrepencies between the best and the worst in the party in terms of their prime ability score. But with this method, variation goes down on the ones you want to be good.

I'm genuinely surprised, in a good way, to see a rolling mechanic that meets my criteria in terms of party power disparities, feats and ASI as both tempting, and satisfies the organic "I'm surprised by how this ended up" itch.

I'm evaluating this as it would be at my table, which uses Tasha's ability score adjustments. Methods that can give the highest as even numbers (16 or 18 in this case) can otherwise be fodder for only pairing with the races that give +2 to it for some types of gamers, which reduced variety. Tasha's eliminates that, you can play any race/class combo without being limited.

Also this particular breakdown can give some really great scores for SAD classes (like rogue archers) with one great and most others decent, and can give a great and a good that most other classes want at the cost of all of the others being 3d6, but is a bit tight when it comes to classes that want three good scores, like a paladin or a wizard (bladesinger). Not a showstopper, just an observation.

All in all, like this a lot.

 

[Deset Gled]

It's a complex system. Whether that's good or bad is a matter of personal preference, but this is a bit too complex for my taste. It does seem like the numbers work out well, though.

Beyond that the only specific critique I have would be on this section:

Optional: Mulligan / Superhero. Although unlikely that one may roll to the extremes, the DM and player should discuss a redo (“mulligan”) if no score has a positive ability modifier, or all scores have a positive ability score modifier.

You would have to work very hard to get a set where there isn't at least one positive ability modifier. Conversely, I think there are a number of cases where a player could game this system to get all ability scores at a low positive modifier if they decided they want to force a re-do. It would probably be easier to just say "mulligans at the DM's discretion".

 

[Charlaquin]

I like this idea a lot, though I feel like the option to improve a score once rolled complicates it too much for my taste. Simply spending points to choose a dice code for each score seems good enough for me.

 

[toucanbuzz]

I like this idea a lot, though I feel like the option to improve a score once rolled complicates it too much for my taste. Simply spending points to choose a dice code for each score seems good enough for me.

I think the "reserve" is supposed to be like insurance. If a disastrous roll happens, you might be glad you had some.

But it might be interesting to turn the insurance into a gamble by further removing any "plusses" to anything 15 and above.

 

[toucanbuzz]

...It would probably be easier to just say "mulligans at the DM's discretion".

Simpler is better. While it sounds silly, I suppose I could see players opting 3d6 and ditching characters until they get the rolls they want. Of course, the DM should step in and nip that type of practice in the bud.

 

[EzekielRaiden]

If folks would like a simpler method, do something like this:

You have five points. By default, roll 3d6 in order. Spend one point on a stat to instead roll 2d6+6 for one stat. Spend two points to instead roll (best 1 of 2d6)+12. If you have points left over, you can spend them to add 2 to any score less than 8, or 1 to any score 8-12. Alternatively, you may swap two stat values by spending one point.

Nice and simple, no need to faff about with big numbers or major modifications. Takes all of five sentences, two of which are very short. Spending 2 points is almost certainly worth the cost, but still a gamble. (If you prefer, just make it 1d6+12 to make it that much more of a gamble, and the process that much simpler.) Players who decide to just roll have a major cushion against really low stats, while players who specialize risk hyper-specialization. Everyone always has at least one stat that rolls 3d6, so odds are decent at least one score will be on the low end. There isn't assign-to-taste, but if you hold onto a point, you can at least swap one stat if necessary.

As baseline examples, I'll do a Dragonborn Paladin and a Half-Elf Bard, since those are characters I've played in 5e before.

Spoiler: Dragonborn Paladin

Gonna take (2d6k1)+12 for Charisma and Strength, 3d6 for everything else, keeping that reserve point just in case.
"High" rolls: 14, 13 (in order Str/Cha)
Remaining rolls: 10, 6, 12, 7, which go in order Con/Dex/Int/Wis.
This gives
Str: 14+2 (racial) = 16
Con: 10
Dex: 6
Int: 12
Wis: 7
Cha: 13+1 (racial) = 14
Got some real crap rolls on those first two! Guess I shouldn't have gambled so hard. I could put my remaining point into bumping up Dex, but I think I like the idea of a knight who fights really well but is as coordinated as a fledgling goose outside of that context, so I'll bump Wis to 9. May invest an ASI into Dex at some point just 'cause that IS a painful hole though.

Final scores
Str 16
Con 10
Dex 6
Int 12
Wis 9
Cha 14
Not ideal, but not totally awful either.

Spoiler: Half-elf Bard

Half-elf: Charisma is really the only "key" stat for a Bard. I'll do the high roll for Charisma, and medium rolls for Dex, Con, and Int. The other stats fall where they may.
High roll: 17
Mid rolls: 17, 10, 13 (order Con, Dex, Int)
Low rolls: 14, 15 (order Str, Wis)

Wow, holy crap, this is an insanely good set given what I went for--dramatically better than the previous! We end up with:
Str: 14
Con: 17
Dex: 10
Int: 13
Wis: 15
Cha: 17+2 = 19

I have the two floating half-elf ability bonuses, but they can't be applied to Cha. Con 18 is a no-brainer for this character, but it's a bit tough to choose between the other two. Bards should be well-rounded though, so I think a bigger Int modifier is better.

Final scores
Str 14
Con 18
Dex 10
Int 14
Wis 15
Cha 19
A surprisingly "physical" character--almost worth considering Valor Bard over the alternatives. That (relatively) low Dex actually prevents the approach I would typically take, sadly, but perhaps this could work as a Bard that dabbles in Bladelock?

 

[Ruin Explorer]

I was going through my old notes and came across this homebrew ability score generation option I used in 3E to give some variance with ability scores so everyone didn't look cookie cutter. This way, maybe there could be a wizard with an 18 STR because does every wizard have an 8-10 Strength? There isn't one wizard in the world who works out? I'm pretty sure I didn't create it, don't know who did, and was positing whether it would translate to 5E.

Buy your dice roll method:

You have 5 points. Purchase a roll for each ability score from the options below. You don't have to spend all 5 points on rolls. You can instead "reserve" any number of your 5 points to improve a score after the dice are rolled. The worse the score, the bigger the improvement.

Point Cost Method	Die Roll	Range	Average
A: 3 points.	15 + 1/2 of 1d6 (round down)	15-18	16.5
B: 2 points.	10 + 2d4	12-18	15
C: 1 point.	6 + 2d6	8-18	13
D: 0 points.	3d6	3-18	10.5

Reserve boost per point spent:

Ability Score	Boost
3-7	+3
8-13	+2
14-16	+1
17-18	+0

Optional: Mulligan / Superhero. Although unlikely that one may roll to the extremes, the DM and player should discuss a redo (“mulligan”) if no score has a positive ability modifier, or all scores have a positive ability score modifier.

Example:

Spoiler

I'm thinking about playing a wizard, so I know I'll want high intelligence, and I also decide I want a decent Con to survive. I use 3 points to buy method A for Intelligence because it gives me the highest range. I buy a Method Cs for Con at 1 point and make the rest Method D (costs me nothing) so I can save 1 point for a boost, just in case.

STR - 3d6 Rolled a 13
Dex – 3d6 Rolled a 12
Con - 6+2d6 Rolled an 8
Int - 15 + 1/2 of 1d6 (round down) Rolled a 17
Wis - 3d6 Rolled a 7
Cha - 3d6 Rolled an 11

Now I have a dilemma. My Con at 8 is poor, and the "boost" would get it up by +2 to a 10 so I'd have no penalty. Or, I could do the same to my 7 Wisdom to get it to a 10 (+3). Or, I might just boost something else! All depends on what you're playing. And, this wizard works out. He's got a 13 Strength. No more embarrassment asking others to help carry your waterskin when it's filled up.

It's cute but it fundamentally doesn't make sense.

It looks like one of those things someone came up with by themselves and absolutely never tested - the initial concept re: paid rolls isn't terrible though I'd argue the numbers/rolls, but the concept of "reserving" points to shore up bad rolls is frankly idiotic. The return is extremely poor, mathematically, and it requires you to essentially assume you'll get a crap roll, which is let's be clear - NOT HOW HUMANS WORK - sorry for the all caps but I've been talking about RPGs for what, 30 years now, on the internet, and one leading cause of truly bad system design is people don't understand how humans work, don't understand like the basic psychology involved.

And the reserve concept is a prime example of that.

The other two main causes, btw, are "bad math/didn't bother to do any math", and "doesn't understand the rules to start with, but has decided to change them". This also has the "bad math/no math" problem. Like there's no way these numbers are the result of anything but some dude's "gut feeling". Which is why the reserve boost numbers are so low.

For once I don't have a ton of time, but in short it doesn't make mathematical sense to reserve stuff, the odds strongly favour rolling 6+2d6 over reserving any points, and the players aren't going to reserve anything even when they don't immediately realize that, because again, it requires them to assume they'll get terrible rolls, which players do not.

And in the end, what it's trying to achieve is essentially the same as 4d6DtL and standard array, but it's considerably more complicated also has more potential to produce outright unplayble characters than either. So yeah I'd this is one of thousands of stat gen methods destined for and deserving of the scrapheap of history.