When is MADness okay?

[Tinkermancer]

I’ve seen people come on the forums with a homebrew idea for a class, an archetype, or even just a feature that is powered off an ability score that is different from the class’s main attribute. Often, posters disparage said suggestion based on the introduction of a little MADness to the class.

My question is, when is MADness appropriate? Would it not make sense for certain concepts to require a character be built in a certain direction? Or should all builds of a specific class be possible and powerful using the exact same array and arrangement of attribute points?

[Mr.Spastic]

I think that doing something that requires a bit of MADness is fine. I don't see any problem with making a build that relies on several different stats. The problems that I see is that sometimes people thing that the abilities that they get from the MAD build are worth it when they aren't. I saw somebody who wanted to play a paladin/cleric multiclass once. I felt obligated to tell him that he would be too spread out between Str, Con, Wis, abd Cha.

Another thing is that being MAD is not good when using point buy.

[Greywander]

MAD is fine if you're not worried about being optimal. I'd say you can probably get by with a 14 or 16 in any important ability scores. Where MADness is a problem is when the player thinks they're creating an optimized build and they didn't roll the stats to back it up. They're going to end up disappointed because they can't perform as well as a SAD build.

In addition, I'd say MADness is a bit okay if you're expressly making a versatile character. For example, if you want to be a skill monkey with proficiency in 15 or so skills, it actually works in your favor to spread your ability scores out (especially in point buy where higher stats cost more). Sure, you're not the best at anything, but you're second best at everything, which can be a useful thing to have in a party. But again, you have to acknowledge that you won't perform on any one task as well as a SAD character.

[JakOfAllTirades]

MADness also works if you rolled for ability scores and got several high numbers to work with.

[LudicSavant]

I’ve seen people come on the forums with a homebrew idea for a class, an archetype, or even just a feature that is powered off an ability score that is different from the class’s main attribute. Often, posters disparage said suggestion based on the introduction of a little MADness to the class.

My question is, when is MADness appropriate? Would it not make sense for certain concepts to require a character be built in a certain direction? Or should all builds of a specific class be possible and powerful using the exact same array and arrangement of attribute points?

Paladin is probably a good example of a case where there's some pretty legitimate tradeoffs between choosing one stat or the other, and the overall package is still strong despite being MAD.

[Knaight]

MAD is a bit vague, really - exactly how many is multiple? 2ADness is totally fine, and if anything could use some dilution (Con tends to at least do this if nothing else does). 6ADness is sufficiently far off design to generally pose a problem. 3ADness is also generally fine (think something like str/cha/con for a Paladin, which is a pretty solid class), while 4ADness is starting to get a bit excessive.

[jdolch]

Strange Question.

MADness is okay if .... drumroll .... you have the Stats for it.

[Hears You]

Mad's fine when it's pretty limited, and is used to provide specialization on top of the primary score and not diverting resources from what you're building to.

So, your primary stat should be used to do most of your things, but your secondary stats should be cool additional you do, not something you focus on. For example I'm a big fan of 4e barbarians, and they had a build build around charisma that when you slew someone you'd howl and give your enemies -cha to attacks for a round.

That sort of thing works fine, you've clearly got a primary stat (str, for gettin that kill) and the charisma powers a cool bonus.

MAD becomes a problem when you need both stats as high as possible to function, because your stats are a pretty limited resource (especially in 5e).

[Greywander]

MAD becomes a problem when you need both stats as high as possible to function, because your stats are a pretty limited resource (especially in 5e).
I wouldn't consider depending on two different ability scores to be truly MAD, but I do see what you're getting at. Some classes are SAD, but still like other abilities when they're available, e.g. wizards only need INT to function, but always appreciate extra DEX and CON. Some classes have a primary and secondary ability score, and only really need to push the primary as high as possible, while the secondary can remain lower, e.g. paladin and STR (primary) and CHA (secondary), or monks and DEX (primary) and WIS (secondary). Even then, though, many of these classes can still function if they dump their secondary ability score.

The thing is, even if you absolutely needed two different ability scores, it's far from impossible to get decent scores in both (getting both to 20 is more difficult). Where I would say a character becomes truly MAD, though, is when they need at least three different ability scores to power all of their features. For example, a paladin/cleric multiclass needs STR for melee, WIS for cleric spells, and CHA for paladin features/spells.

Again, though, I want to reassert that MADness isn't a problem if you aren't expecting to play an optimized character. With point buy you can get an array of 13, 13, 13, 12, 12, 12. Grab standard human and you've got at least a 13 in all ability scores. Or, if you can afford to dump one ability score, you can get an array of 14, 13, 13, 13, 13, 8. Slap a half-elf onto it and by 4th level you can have a 16 and four 14s. In the extreme case where you only need three ability scores and can dump everything else, you can get an array of 15, 15, 15, 8, 8, 8.

You can absolutely have a fun time playing a MAD character as long as you don't expect to be quite as effective as a SAD character.

[DeTess]

MADness like the paladin (STR, CON and CHA) isn't a problem. What I've seen with some regularity in homebrew is classes that need 4 stats (a recent class I saw for example needed WIS, CON and DEX, and CHA for one subclass) which is one too many when assuming a 'standard' game. Basically, the moment you need to have rolled above average on dice and can barely use point-buy to play a single-classed character you have a design problem.

TL;DR 3 stat dependency (including CON): not Bad, 4 stat dependency: Bad

[jdolch]

Well, it's bad if you decide that you really want to play that MAD class and then you roll. But if you just keep it in the back pocket and wait for a game were you actually DO roll above average then it's fine, isn't it ?

[Greywander]

What I've seen with some regularity in homebrew is classes that need 4 stats (a recent class I saw for example needed WIS, CON and DEX, and CHA for one subclass) which is one too many when assuming a 'standard' game.
Yeah, that is... not okay. One primary, one secondary, full stop. You might be able to make them able to use multiple ability scores, depending on what they wanted to be good at, but it shouldn't be required to have a functional character. Although do be careful here, too; a set of really good rolls could make that character extra OP since they could actually use those ability scores for something. Probably the best way to do this would be a primary ability score for the main class, and different secondary ability scores for each subclass. In fact, you see this with the Arcane Trickster and Swashbuckler.

I don't count CON unless it's used for more than the standard. Sure, everyone, especially martials, love having HP, but most of your HP comes from your hit dice. I remember coming to the conclusion once that even on a front-line fighter, I would still benefit more by taking a feat over pumping up CON. Even on an "optimized" character, 14 CON is usually plenty. Now, for barbarians, CON is absolutely a secondary ability score since they also add it to their AC.

[Beechgnome]

MADness can be fine at an average table if you know how to optimize the crap out of everything else. In other words, go ahead and be that GOO warlock/moon druid so you can communicate telepathically when In wild shape, but make sure you choose the best spells, feats, skills and level distribution to make it sing. Stats are super important, but they aren't the only way to ensure competitiveness.

[Sception]

IMO secondary stats are fine, and a choice of multiple secondary stats, with enough support to each that a character can reasonably pick just one, is ideal. I would frankly prefer if more subclasses were written to support distinct secondary stats.

Where it gets into trouble is when you have multiple primary stats - ie, different stats that the same character is trying to use for hit rolls or save DCs of primary offensive abilities. This can work fine at the very early levels, before ASIs start piling up, or at the very latest levels, after you've had time to max out multiple stats, but for most of the middle of the game you're going to be suffering more turns where your character does nothing at all because they missed their attack or their target passed their save, and rounds where you do nothing at all just aren't fun. Furthermore, in games that use feats, a character with multiple attack stats in need of maxing out ends up making more boring, generic "+2 to a stat" selections and fewer "interesting choice that changes how your character plays" selections.

The other problem is when you start requiring multiple secondary stats on the same character (as opposed to a selection of secondary stats that a character can comfortably choose just one of), on top of every character being pressured to put some points in con, particularly on non-heavy-armored melee characters that also can't afford to scrimp on dex either. This kind of madness, while not as severe as multiple primary stat madness, can be problematic right from level one, particularly if the secondary stats are used for AC as in the case of Barbarians and Monks. Monks in particular are about as far as you can reasonably go in this case, with primary stat dex, secondary wis, on top of still needing a decent con score. So a hypothetical monk subclass with abilities based on their str, int, or cha mod would most likely be stretching a character too thin, unless that subclass also reduces their dependence on dex, wis, or con somehow. And even then, the fact that monks don't get their subclass features until like 3rd level would make the first two levels very painful for such a character to play through.

[jas61292]

Personally, I think having features work off multiple ability scores is fine. I just think it has limits. Personally, I feel that ignoring Constitution, which everyone wants, any given class should only have to rely on two stats. And, as AC is important, I'd consider Dex important for anyone who has light armor proficiency or less (with no special AC calculation).

That said, beyond that, there is a lot of MAD territory you can cover and still be fine. But the key is to make being MAD worthwhile. An interesting comparison is the Paladin and the Berserker. A number of people have already posted about why the Paladin is perfectly fine being MAD with Str and Cha, but if you look, I'm sure you can easily find people complaining about the Berserker. Ok, ok, most of those complaints have nothing to do with this, but if you actually read full analysis that get to abilities past frenzy, there is often complaining that their 10th level ability runs off of Charisma.

There is absolutely nothing wrong with a Barbarian having an ability run off of Charisma, but because it is not a core ability for the base class, this feature needs to be worth it for someone to invest in it. And, arguably, it is not. Like, imagine if Paladin had all spells and features all run off of Strength, except uses of Divine Sense and Cleansing Touch. Do you think Paladins would still invest in Charisma? No. Do you think people would complain that the game is locking features behind inferior abilities? Of course they would.

Now, imagine a Berserker's Intimidating Presence still works off of Charisma, but was a bonus action, locked down movement, could be used on multiple enemies at once, and lasted until they make a save, rather than needing to be refreshed manually every turn. I'd bet you the complaints about a barbarian having to use Charisma would vanish. Because it would be worth it.

That's really the question you need to ask when looking at something being MAD. Is having the ability to use both ability scores well worth it, in comparison to just focusing on one along side other generically good things.

[Gastronomie]

I believe MADness is fine when the three scores required for creating an optimized character of that class are:
(1) One out of STR and DEX
(2) CON
(3) One out of INT, WIS, and CHA

Because that is how the official classes are designed.

[Psikerlord]

MADness is fine if you get your players to roll their stats. It's only a potential problem if you enforce point buy.

[Arkhios]

MADness is fine if you get your players to roll their stats. It's only a potential problem if you enforce point buy.

Even with point buy, it's a problem only in people's minds, because 5th edition in particular doesn't require you to have maxed out ability scores to be effective. A 14 in an ability score, for example, is quite enough, even for someone whose spellcasting relies on it. Essentially, if you have a positive modifier, you're doing great.

[KorvinStarmast]

When is MADness okay?
When you are a ranger.

[Psikerlord]

Even with point buy, it's a problem only in people's minds, because 5th edition in particular doesn't require you to have maxed out ability scores to be effective. A 14 in an ability score, for example, is quite enough, even for someone whose spellcasting relies on it. Essentially, if you have a positive modifier, you're doing great.

100% true you're right. My fighter had 16 Str and did perfectly well.

[gkathellar]

As a general rule, MAD is a term used to describe the reason why some classes don't function properly, and the reason why some classes scale up in effectiveness far more dramatically when their array of ability scores is improved.

It is neither intrinsically good nor intrinsically bad.

So really, there are multiple answers to this, but the most straightforward is probably: MAD is okay when, using an average ability score array, the class works properly.


Even with point buy, it's a problem only in people's minds, because 5th edition in particular doesn't require you to have maxed out ability scores to be effective. A 14 in an ability score, for example, is quite enough, even for someone whose spellcasting relies on it. Essentially, if you have a positive modifier, you're doing great.

Mm ... no. 5E has a very tight math curve, and while it's certainly possible to keep up with low stats, you're suffering from a noticeable disadvantage if your friends hit on an 8+ and you only hit on an 11+. For some classes it's worse: monk with 14s in both of their main stats gets hit somewhere between 20-30% more than is expected, depending on level. That's a pretty big difference in competence in practice.

If anything, spellcasters typically have it better than martials, because slightly lower DCs isn't actually that bad compared to persistently reduced damage and tanking ability.

[Rixitichil]

The existing classes with MAD elements tend to work best where they allow multiple different choices for the stats. The question of whether your Barbarian maximises their Str, Con or Dex first is an interesting one. There is a degree to which you can make the abilities that rely on a Tertiary stat that little bit more generous as you know that in order to capitalise on them, a character will have lesser performance in another core area.

[jas61292]

Mm ... no. 5E has a very tight math curve, and while it's certainly possible to keep up with low stats, you're suffering from a noticeable disadvantage if your friends hit on an 8+ and you only hit on an 11+. For some classes it's worse: monk with 14s in both of their main stats gets hit somewhere between 20-30% more than is expected, depending on level. That's a pretty big difference in competence in practice.

I strongly disagree with this, a least in an array or point buy game where every point missing somewhere is a point gained somewhere else (for the most part). The measure of what is competent is all about how successful you are, not about how much more or less competent a theoretical other character could be. If a ability score of 14 is enough for you to be competent in most scenarios (it is), then it doesn't matter how competent anyone else is.

Yes, 5e has a tight math curve, but that just means sp much of your success is going to be based on random chance. A 5 or 10% difference in outcome may sound significant in a white room scenario, but in actual play, it's far less relevant, as it only comes up infrequently, and will rarely be in a situation that would totally change the outcome of a scenario.

And I mentioned array and point buy at the start because, in those, having as lower main ability means your have a higher score elsewhere. And if a 5% lower chance to hit is significant, so may be a 5% greater chance to circumvent the situation in some other manner.

[Psikerlord]

I strongly disagree with this, a least in an array or point buy game where every point missing somewhere is a point gained somewhere else (for the most part). The measure of what is competent is all about how successful you are, not about how much more or less competent a theoretical other character could be. If a ability score of 14 is enough for you to be competent in most scenarios (it is), then it doesn't matter how competent anyone else is.

Yes, 5e has a tight math curve, but that just means sp much of your success is going to be based on random chance. A 5 or 10% difference in outcome may sound significant in a white room scenario, but in actual play, it's far less relevant, as it only comes up infrequently, and will rarely be in a situation that would totally change the outcome of a scenario.

And I mentioned array and point buy at the start because, in those, having as lower main ability means your have a higher score elsewhere. And if a 5% lower chance to hit is significant, so may be a 5% greater chance to circumvent the situation in some other manner.

The ease with which adv or other bonuses are obtained on attack rolls also supports this.

[GlenSmash!]

Anytime.

Try starting with a bunch of stats at fourteen. Look for ways to get advantage on attack roles and checks. Or use superiority dice to increase hit chance. Bump main stat later if you like.

Or pick spells that contribute to the party without calling for roles. Bless for example.

[JakOfAllTirades]

Even with point buy, it's a problem only in people's minds, because 5th edition in particular doesn't require you to have maxed out ability scores to be effective. A 14 in an ability score, for example, is quite enough, even for someone whose spellcasting relies on it. Essentially, if you have a positive modifier, you're doing great.

This might depend on the table. Most games I've been in, a 16 is just okay, but I'd have trouble getting by with a 14, especially for a spellcaster. Too many monsters making their saving throws!

[Arkhios]

This might depend on the table. Most games I've been in, a 16 is just okay, but I'd have trouble getting by with a 14, especially for a spellcaster. Too many monsters making their saving throws!

It boils down to accepting and realizing that it doesn't really matter in practice how high or low chance you have to accomplish X/Y/Z. Dice roll how they roll; unless you always roll the extremes, when it raises a question whether you have loaded dice.

Bounded Accuracy is the key. Ability Scores matter to a certain point, until they don't matter.

Personally I find it very annoying when people try to prove a point with hair-splitting calculation of probabilities.

[Sception]

A 14 attack or save DC stat is awkward but workable at level one, sure. But at level 10, not so much. Becuase the ACs and saving throws of the enemies your generally fight do tend to increase with your level, "bounded accuracy" or no. Frankly, the fewer bonuses you have access to, the more the ones you do have matter, so bounded accuracy tends to result in a small difference in stats being more important, not less.

So the problem with madness in primary stats is less that you can't start all of them maxed out. It's actually often possible to get two or even three stats to a +3 bonus at level one with the right point spread and racial choice, and even if you don't want to be that obnoxiously min-maxed, the difference between a +2 and a +3 bonus at level 1 isn't at all unworkable. Heck, the loss of a point of damage on hits might matter more than the point of accuracy at that level. The problem again is when you get to level 10, where most campaigns are wrapping up. Lets imagine a split attack stat character who started with two 14 attack stats. At level 10, one of those can be up to 18. Still behind where it could be by a bit, but the difference between a +4 and a +5 mod at level 10 is no more severe than the difference between a +2 and a +3 mod was at level 1. But the other attack stat is still a 14, and the difference between +2 and +5 is just *a lot* more noticeable. Now, you could split stat advancement, and have 2 16s, but +3 to +5 is still a pretty hefty difference, and now it applies to all your attacks, not just some of them.

Eventually the overall stat cap of 20 makes itself felt. Single stat classes stop pulling ahead, and multi-stat classes have a chance to catch up. So at the lowest levels and the highest levels MADness in attack stats isn't as much of a problem, but in a huge chunk of the middle levels, where most campaigns spend most of their time, it's a real hassle. And in particular it's *most* noticeable between levels 8 and 11, which again is where most campaigns are wrapping up with climactic fights against their main antagonists.

[Arkhios]

Proficiency bonus increases as you level up. Really, whether you have a +2 or +3 bonus from your ability score is not that big deal.
I hear time and again that having 16 in your primary score for the rest of your career (as far as up to 20th level) is enough. Why is the difference between 14 and 16 (only one point) such a big deal?

You don't absolutely have to get that ability score to 20. You can get by with 14 and be just fine. I'm not saying it's the best of the best. But it's not horrible either.

[DeTess]

Proficiency bonus increases as you level up. Really, whether you have a +2 or +3 bonus from your ability score is not that big deal.
I hear time and again that having 16 in your primary score for the rest of your career (as far as up to 20th level) is enough. Why is the difference between 14 and 16 (only one point) such a big deal?

You don't absolutely have to get that ability score to 20. You can get by with 14 and be just fine. I'm not saying it's the best of the best. But it's not horrible either.

Is this what you imagine it'd be like, or have you actually played a character that's a a couple of points behind in bonus to the other characters? Even if maths say that it doesn't matter I can tell you that in actual play it feels bad when you're 2-3 points behind in attack/skill/whatever bonus.