D&D 5E Remember the "3d6 For Stats In Order" Thread? I'm doing it!

[Blue]

Any thoughts on race/class/build for these stats above (bolded)? I need some ideas for my backup character. All official content plus UA permitted. Feats/multiclass okay.

Well, nothing MAD. Hmm, if you go for a race with +2 DEX you can get a 15 DEX and 13 CON, which is playable, especially with DEX being useful in many different ways. I'd suggest some sort of archer, but if it's fighter, rogue, ranger, or a multiclass.

You won't have the ability scores to spend lots of feats on SS and crossbow expert so you won't be a DPR king. But you should be able to hold your own.

I'd start with a level of fighter - Archery style can boost your attack roll, with a +2 Dex mod medium armor will give you the best AC for now, and you can change it later. Plus fighter MC never hurts for martial characters. Each level is good. Go rogue from there if you want, or stay fighter.

Alternately, the UA Ranger beastmaster would be an interesting replacement. Because a good part of the heavy lifting is now on your companion's ability scores and not your own. And you can still use archery.

Instead of a +2 DEX race, vHuman gives +2 mods for DEX and CON and an archery feat.

Of course, you could also ignore your two best scores and go for a moon druid. Replace all your physical scores when you get up to wildshape. With a wisdom of 11 you'll have to focus your casting on buffs and healing when not in form, but that's viable if restrictive. IIRC the starting party has several neggative CON people. Goodberries dished around to everyone in the morning can be a quick feed to get people back on their feet. Efficiet, and can be delivered by whomever, not just on your initiative.

 

[clearstream]

Well, nothing MAD. Hmm, if you go for a race with +2 DEX you can get a 15 DEX and 13 CON, which is playable,

I wanted to focus on "playable" (emphasis mine) as I feel like it reveals some of the philosophy behind roll-in-order. The character can be optimised, the optimisation is flawed because some ability scores are not where one might ideally want them; even so, it is playable. Roll in order isn't about stopping players from optimising, rather it asks them to focus on the inherent playability of a character.

 

[Blue]

I wanted to focus on "playable" (emphasis mine) as I feel like it reveals some of the philosophy behind roll-in-order. The character can be optimised, the optimisation is flawed because some ability scores are not where one might ideally want them; even so, it is playable. Roll in order isn't about stopping players from optimising, rather it asks them to focus on the inherent playability of a character.

I always liked rollling. Back in AD&D 2ed I played with a DM for yers that had in-order rolls though much higher than 3d6 (4d6 drop the lowest, do three sets and pick one). Felt very organic.

In 5e though, I enjoy the Faustian bargains of ASI vs. feat. But from playing some games that are point buy and some that are rolling, I find it only applies to point buy. For games that are rolled, you usually have several characters that must take ASIs to keep up their support of the party as much as a generic replacement. Others who were blessed with high (and well fitting) rolls on the other hand already are good or great, and they can then branch out with feats. And many players will take feats that multiply their effectiveness.

So rolling you end up with the haves and the have-nots, and because of how 5e advances ability scores that divide will only get worse.

Look at Kol, with his 18, two 15s, and a 14 before racial modifiers, vs. the backup character who's highest pre-racial are two 13s. Which of them will gain more spotlight both in combat and in other pillars of play that reply on skill checks? It's not even.

 

[Inchoroi]

Oh man...

Back in the day, I was playing B/X, and rolled. Didn't get anything higher than a 9. I had 1 hit point. However, Celsius survived up to level 3, which made everyone at the table go, "Yep, every character you have from now on is going to die horribly. You've burned all your luck."

I can't remember the name of the adventure, but another character got liquified and teleported through a telescope lens in a wizard's tower. I keep telling my current 5e players that I'm going to force them to play B/X someday, just to torture them.

I always liked rollling. Back in AD&D 2ed I played with a DM for yers that had in-order rolls though much higher than 3d6 (4d6 drop the lowest, do three sets and pick one). Felt very organic.

In 5e though, I enjoy the Faustian bargains of ASI vs. feat. But from playing some games that are point buy and some that are rolling, I find it only applies to point buy. For games that are rolled, you usually have several characters that must take ASIs to keep up their support of the party as much as a generic replacement. Others who were blessed with high (and well fitting) rolls on the other hand already are good or great, and they can then branch out with feats. And many players will take feats that multiply their effectiveness.

So rolling you end up with the haves and the have-nots, and because of how 5e advances ability scores that divide will only get worse.

Look at Kol, with his 18, two 15s, and a 14 before racial modifiers, vs. the backup character who's highest pre-racial are two 13s. Which of them will gain more spotlight both in combat and in other pillars of play that reply on skill checks? It's not even.

I thoroughly enjoy having my players roll for stats, but do it a little differently with bonus feats (I eliminated or re-wrote the worst feats) so players don't have to choose between ASIs and Feats. It works fairly well with my groups, save for one player who has such abysmal luck that I started asking another player to roll stats for him. I've been thinking that, for my next campaign, having them do characters randomly, via the tables in Xanathar's Guide. It would amuse me, but I don't know how much they'd enjoy that.

 

[FrogReaver]

One idea for rolled stats might be to give a feats if certain low stat conditions are met.

No stat above 14 pre racial then +1feat
Con below 8 then +1 feat
Sum of all stats below 58 then +1 feat

Something like that could be fun

 

[clearstream]

In 5e though, I enjoy the Faustian bargains of ASI vs. feat. But from playing some games that are point buy and some that are rolling, I find it only applies to point buy. For games that are rolled, you usually have several characters that must take ASIs to keep up their support of the party as much as a generic replacement. Others who were blessed with high (and well fitting) rolls on the other hand already are good or great, and they can then branch out with feats. And many players will take feats that multiply their effectiveness.

So rolling you end up with the haves and the have-nots, and because of how 5e advances ability scores that divide will only get worse.

Look at Kol, with his 18, two 15s, and a 14 before racial modifiers, vs. the backup character who's highest pre-racial are two 13s. Which of them will gain more spotlight both in combat and in other pillars of play that reply on skill checks? It's not even.

Overshadowing comes up repeatedly as one of the chief faults of roll-in-order; although it is a general fault with rolling at all. That is why I'm trialling drawing cards (without replacement): so that all characters have the same total for abilities (though not necessarily total modifiers, although one could do it that way).

Character Generation
A deck of eighteen cards is used to generate ability scores; comprising five “2”s, four “3”s, four “4”s, and five “5”s. Draw (without replacement) and sum three cards for each ability, allocating in order drawn. All eighteen cards will be used—once each—to yield six scores from 6 to 15 that collectively sum to 63.

Ignoring that I am also working within a 6-15 range - which one could readily expand to 3-18 - I think draw without replacement has a lot to commend it. I prefer it to setting bounds for keepable characters... although right now I can't quite articulate why. Maybe because if the intent is to bound, I prefer a design that produces values falling within the bounds?

 

[Blue]

Overshadowing comes up repeatedly as one of the chief faults of roll-in-order; although it is a general fault with rolling at all. That is why I'm trialling drawing cards (without replacement): so that all characters have the same total for abilities (though not necessarily total modifiers, although one could do it that way).

Let me start by saying everything you want to do is valid, and if your table is happy with it then run with it. I'm just discussing nuances.

I agree with you about rolling at all bringing those inequalities to 5e. That's why for the games I run I prefer point buy. That's a 5e-ism bases on the ASI vs. feat point I mentioned before, not a gaming preference in general or even with earlier editions of D&D.

The deck idea is definitely a step in the right direction.But just as the concept of "dump stat" is around, D&D is really a game where you need a few good ability scores in the right places. The the "off ability scores" are moderately high doesn't contribute as much.

This has two effects. One is that someone with a 15 (very rare) or a 14 will be doing better then a character with 11 high, even if the 11 high is three 11s and three 10s for not a single penalty mod. (This can be drawn with the deck mentioned.)

The other is when you get a character with ability scores that don't work well together - say the moderate high INT and STR that leaves everything else at a penalty.

The deck is a step in the right direction, but how about just randomizing the order of the standard array?

I've seen another where everyone rolls, and then everyone picks which set they want, including multiple picking the same. To qualify for in order for you you've have to do something like pre-roll a rank (highest to STR, next highest to WIS, ...) or just randomize the order afterwards. The second is more random, the first does allow a player who wants to play a MAD character a little bit of knowledge if they should go for a array with a good score vs. an array with several moderate scores.

 

[clearstream]

The deck is a step in the right direction, but how about just randomizing the order of the standard array?

I considered something similar, I tabulated the permutations of points-buy to see how that looked. So players could roll for which valid points-buy array they got. Net modifiers range from about +5 to +9 after race, which sets the game difficulty at "easy" for players with high system mastery. I want a harder setting. (IMO the default game difficulty is correctly set and I do not use "easy" pejoratively.)

With the parameters I am using, the range is +0 to +4 after race. That mitigates overshadowing.

It is not my goal to ensure that players have optimal characters (which is not the same as being opposed to optimising, if you see what I mean). At the same time it is my goal to disrupt party balance. Given fiat over race and class, even with say high Strength and Intelligence, a player can thread the needle. They might be mildly overshadowed, but no more than can arise from disparate degrees of system mastery, or levels of interest in optimisation.

 

[FrogReaver]

I considered something similar, I tabulated the permutations of points-buy to see how that looked. So players could roll for which valid points-buy array they got. Net modifiers range from about +5 to +9 after race, which sets the game difficulty at "easy" for players with high system mastery. I want a harder setting. (IMO the default game difficulty is correctly set and I do not use "easy" pejoratively.)

With the parameters I am using, the range is +0 to +4 after race. That mitigates overshadowing.

It is not my goal to ensure that players have optimal characters (which is not the same as being opposed to optimising, if you see what I mean). At the same time it is my goal to disrupt party balance. Given fiat over race and class, even with say high Strength and Intelligence, a player can thread the needle. They might be mildly overshadowed, but no more than can arise from disparate degrees of system mastery, or levels of interest in optimisation.

My issue with adjusting the stat spread is that some classes need more stats than others.

Classes like monks foe example feel really bad in a low stat system. Paladins kind of do as well. It'd be nearly impossible to play any of the melee bards this way.

 

[clearstream]

My issue with adjusting the stat spread is that some classes need more stats than others.

Classes like monks foe example feel really bad in a low stat system. Paladins kind of do as well. It'd be nearly impossible to play any of the melee bards this way.

I think this works out for MAD classes better than both points-buy and 4d6k3, relative to other party members. The spikiest array is 15, 14, 12, 9, 7, 6, while the flattest is 11, 11, 11, 10, 10, 10. The most any character will have after race is +3 on their highest ability.

MAD classes typically want three good ability scores, while non-MAD want two. With points-buy, non-MAD have a clear advantage: they sink their points into those two. With 4d6k3, characters are likely to have one or two very strong scores, but far less likely to have three (unless they have godly stats, like the paladin in my previous campaign who overshadowed all bar one of the other party members).

The flatter stat-spread from draw-three-without-replacement supports MAD, relative to fellow party members.