"In my game the Mental Attributes are..."

[Harshax]

I'm obviously using a simple list of basic things that have been skills in D&D game, despite this system probably more appropriately being a dicepool system, instead of inventing an attribute and skill system for the purpose of this forum post.

It's pointless to propose reason against arguments that are solely based on biased outliers.

Honest example that are kind of embarrassing:

I do or have done many Fortitude things in real life. Some still regularly. Specifically, I've been in three fist-fights this year. I've had my ass handed to me in one of them, to the degree that I thought I received permanent damage to my eardrum. Categorically, I've failed 33% of the time, yes? But, I tried 100% of the time in those encounters to resolve the situation with Fortitude things.

In my kwoon, I've met a lot of individuals that have no other observable Fortitude skills, except their martial art. They're good in set piece strikes and blocks, but when advanced students slip in something unexpected they fail Fortitude rolls very quickly, especially when they're just admins or couch potatoes the other six days a week. They lack a body of experience that makes them an adept at Fortitude things despite having ranks in Fighting with Hands.

When I'm not getting my ass kicked, I do system engineering and architecture. I have mad skills, but I'm terrible at math. Like, hardly competent at some traditional maths*. This makes a significant number of development tasks cumbersome for me. I know a fuck-lot of kernel-level Linux and can dizzy most people with my grasp of network protocols and regular expression. But, I can't idly wrap my mind around using those skills to design something new without a quiet place and a lot of personal time (taking 10).

Some people will read those statements above and isolate one sentence to dismiss the idea that skills equal overall competency in a particular Attribute. Others might see that I might be better in one category than another. Ultimately, fuck you for implying your view point is better than mine because of one particular edge case. It's just a game.

* - Eugenia Cheng is a heroine for being able to explain categorical mathematics. Check her out.

[JonSetanta]

If your game has characters with "lots of ranks in Jump, Climb, Athletics, Tumble, and Break Doors" then you have a bad game.

Speaking of which, a bit off topic of "mental stats", but I have a plan coming up to allow non-Rogues (aka Warriors with low INT and few skills in D&D) to stay relevant in those very same tasks without investing skill points.

[OgreBattle]

We use Intelligence and Will.

Intelligence impacts ranged attack rolls, DCs to resist spells and some special abilities, and resist checks against illusions. Many skills also rely on Intelligence.

Will covers far more saves, and includes how many spells/ki abilities you can use. Interpersonal skills rely on Will.

A wizard wants a high Will to cast a lot of spells and a high Intelligence to ensure they're impactful.

That sounds straightforward, animals failing an illusion save vs scarecrow can be built into the game with that.

Having the core attributes also go into Magic Points or Magic Power can work, though it's something I'm personally avoiding for my Heartbreaker.
---
What I've got in mind is... 4 or so core attributes that are core ways to interact and resist things (pick stuff up, detect ambush, resist fear, get kicked by a mule and what happens) then 3 cateogories of skills (Physical, Social, Technical) with 5 or less skills that aren't attached to attributes.

[Stubbazubba]

I use basically the 4 stats from Shadow of the Demon Lord: Strength, Agility, Intellect, Will, though I call mine Might, Nimbleness, Wits, and Spirit.

Wits is intelligence, memory, and perception. It's quickness of thought and internal processing. Perception is its associated save/defense, and it's what Wizards and Artificers use for their magic. It's tied to knowledge skills (arcana, nature, religion, society), investigation, medicine, and tinkering.

This means the knowledgeable, the widely-read, the experts are the first to notice things out of place and detect bullshit. Conversely, those who notice the little details are the ones who develop the best, most complete understanding of topics.

Spirit is mental fortitude and charisma. It's strength of ego and power projection. Will is its save/defense, and it's what Mages, Oracles, and Warlocks use for their magic. It's tied to creature handling, deception, impress (inspire/intimidate/charm), performance, and persuasion.

This means those with the strongest convictions are more impressive, more persuasive, and more able to bring others to their cause. Conversely, those who can successfully impress or manipulate others' minds don't fall for their own tricks as easily as their marks do.

[Neo Phonelobster Prime]

trying to make ability scores work better is a fool's errand.

This sentence is correct.

Ability scores in their traditional usage are for suckers and do more harm than good to TTRPGs through several means.

Design without traditional attribute mechanics period, mental or otherwise, it makes your game better.

I tried it on impulse and it's rewarded me non stop since.

I've gone over this before.

[MGuy]

I am using the traditional DnD 6 stats. There's no special reason for it beyond I started with those and haven't had a reason to change the number. I changed the names of some of them and I use my own descriptions to better fit how things work in my setting. Like strength is still being strong but also is a measure of the character's tactile kinesis. I use some degree of sympathetic magic in my setting so the stronger you are better able to do things related to strength and magic abilities related to strong guy stuff are also affected by strength and its related skills. Likewise with mental stats. I use wits, intelligence, and presence. They still correspond to most of the skills you'd expect. Wits deals with detecting stuff, intelligence with knowing/remembering/building stuff, and presence is mainly use for affecting others. Just as strength is used to do strong person things magic interacts similarly with the other attributes. So any kind of detection magic relies on wits (and related skills), presence (and related skills) are used to influence others, and int (and related skills) is used for knowing/remembering/building stuff. Int also determines the number of 'memory slots' you have whenever an ability can be expressed in a way that I'd like to limit: like illusions and alternate forms. Int is also used for any kind of magic based damage resistance. Presence is used to augment magic damage. Wits, combined with Agility, effects initiative. The stats also effect saves/defenses. Wits, along with Agility effect Reflex, both Int and Presence effect Will. In both cases the highest bonus is used along with the worst penalty (if applicable).

There is no special reason for some of the choices I made. I could've left the names alone but I felt like changing them at some point. Sometimes it was because I liked the word better (like Agility>Dexterity) or I felt the new word was more fitting for what it would do (Presence>Charisma). These work well enough and I haven't run into a mechanical issue that made me want to reconsider any of that. I think the only thing I've had to think twice about was whether or not I wanted to nix the traditional way DnD counts its stats. Right now I only use bonuses and penalties since that's good enough for most things related to these stats and it's just easier to work with just the bonuses. There are probably a bunch of details I am not thinking of right now or feel like explaining at length but my set up is pretty close to regular DnD anyway and isn't anything really groundbreaking.

[Neo Phonelobster Prime]

I am using the traditional DnD 6 stats.

So... the traditional 6 D&D stats only called different names, applied to different things and applied in different ways. And also then you replaced the actual traditional attributes with their derived attribute bonus/penalty. Those traditional 6 D&D stats.

I don't want to sound too much like I'm making fun of you here, what you are actually doing sounds OK, at least a mild improvement in a few ways even if it wouldn't satisfy many of my fundamental issues with base attribute mechanics.

But you know I couldn't walk past "I'm sticking to tradition, by changing everything!".

[MGuy]

I am using the traditional DnD 6 stats.

So... the traditional 6 D&D stats only called different names, applied to different things and applied in different ways. And also then you replaced the actual traditional attributes with their derived attribute bonus/penalty. Those traditional 6 D&D stats.

I don't want to sound too much like I'm making fun of you here, what you are actually doing sounds OK, at least a mild improvement in a few ways even if it wouldn't satisfy many of my fundamental issues with base attribute mechanics.

But you know I couldn't walk past "I'm sticking to tradition, by changing everything!".

I'd suspect most people could, given that what I said wasn't strange. I think this is just a you thing.

[Neo Phonelobster Prime]

Hey look, something that is not a defense of presenting relatively boring information in an amusingly dumb ass way.

But it IS full of personal spite, that's gotta count for something right?

[JonSetanta]

Anyway, in my experience, whichever stat system is optimal, I've found the ones that split "dodge" from "ranged attack" are ideal, but as far as The Mage Stat vs. Social Bonus goes, simplicity is best.

[MGuy]

Hey look, something that is not a defense of presenting relatively boring information in an amusingly dumb ass way.

But it IS full of personal spite, that's gotta count for something right?

I'm gonna guess that the new account and name is an attempt to see how fast you can get people, probably not just me, to put you on ignore. You've made a lot of progress in two posts. I'm eager to see what gets you over the finish line.

[erik]

I mean. The crustacean isn’t wrong. The only good reason to keep the 6 stats is because people recognize them. The further you walk away from that by changing their names and changing how they function the less value you get from the association.

I don’t think he was being insensitive or offensive by making the observation. Hell. He even complimented it partway through. That’s damn near civil.

[erik]

[doublepost sorry]
I do confess my bias that I also think mental attributes are a dinosaur that deserves to be extinct. Mechanically not every system needs a set of core additive bonuses to apply to rolls. And of those that do go that route they are almost certainly still better off avoiding the traps of trying to quantify intelligence and wisdom and instead just pinning them to their spellcasting if that’s why they are being kept. Call em Lore, Divinity, and Will. Or some such.

There are certainly avoidable squickfeels that spawn from mental attribute modifiers such as racial mods to those scores.

I mean other than very broadly for people with mental retardation there isn’t a reason to quantify finer granulation of intelligence as it is not just one thing. IQ tests are bullshit except for clinically significant outliers to show developmental delay. And the rest of the differences aren’t relevant to a Ttrpg unless you are playing some learning/school simulator.

Mental stats make very little sense if examined even gently, which is why most people just shrug and don’t examine it at all- accepting it just as an arbitrary conceit of the genre.

[OgreBattle]

Anyway, in my experience, whichever stat system is optimal, I've found the ones that split "dodge" from "ranged attack" are ideal, but as far as The Mage Stat vs. Social Bonus goes, simplicity is best.

Yes I see benefit in not having a single core attribute govern both offense and defense. Either more core stats like Shadowrun, or move offense to another part of the character sheet.

"Mage Stats" is a whole topic to itself, if the setting as magic as nature or an outside spooky force only for book worms and so on.

Mental stats make very little sense if examined even gently, which is why most people just shrug and don’t examine it at all- accepting it just as an arbitrary conceit of the genre.

I can see INT being replaced by a skill system, lore, knowledge tags. But there's some aspects like "detecting someone following me" "resist running away from the frightening skeleton" that almost all player characters should have baked into the system.

[erik]

Awareness is pretty universal. I wound up making it one of the core attributes I cared about in a fantasy/scifi game. Resisting magically induced terror or resisting a flight or flight response could be a universal that all characters need, depending upon the game.

A good system should look at the kind of rolls every character is expected to make and the ones that are gated.

A fantasy game will probably want to cover feats of strength, speed, combat, lore, awareness, stealth, along with however you resist damage and magic. Some of those may be able to be conflated or expanded depending upon how much granularity your want. Social stuff if even mechanically resolved, can be a separate gated minigame or it could be central to the game as an attribute or several attributes if it is a diplomacy game.

Ideally attributes should be something you can describe so players can understand what characters are generally capable of. Like lifts so many kilos. Can run so fast. Sometimes in relation to each other in case of opposed attributes like stealth and awareness. So stealthy can slip past a guard dog. So alert can notice a sneaking cat. Most games do a good job of this.

D20 is actually a bit bad at it since stats are a pile of wharbargl. The system for attribute checks is pretty garbage as a +1-4 modifier on a d20 is not significant. And a 30 stat where the bonus actually becomes meaningful the stat itself is incomprehensible as it is far outside the range of humanity. So attributes only act as a bonus pool for other things that already get tons of different bonus types. They’re the essence of fungible.

[MGuy]

I mean. The crustacean isn’t wrong. The only good reason to keep the 6 stats is because people recognize them. The further you walk away from that by changing their names and changing how they function the less value you get from the association.

I don’t think he was being insensitive or offensive by making the observation. Hell. He even complimented it partway through. That’s damn near civil.

Problem is I never made a claim that I keep them cause if their legacy value. I said that's what I started with and I haven't had reason to change them. Though I think it's pretty obvious every edition of DnD has tweaked what they do and the renaming serves no purpose outside of the ones I outlined. They are still strong, tough, fast, perceptive, smart, and influence stats. They still apply to mostly the same skills. The only major difference I mentioned was that they work with magic differently and that's because magic works differently in my heartbreaker.

So PL saying lol you used the word traditional but you changed things (as if they haven't been changed before in the game referenced) is... Let's say not helpful if I am being charitable. You are assigning his post more charity then I would given the words he elected to use in both.

[merxa]

I'm using:
Strength (STR)
Dexterity (DEX)
Mind (MND)
Spirit (SPR)

and i may refer to STR + DEX as body, MND + SPR as soul.

skills are made up of general skill groups: Athletics, Guile, Thievery, Nature, Magic, Academics, Perception.

Taking Athletics as an example,

Athletics

Your bonus to athletics is calculated by adding together your STR + DEX and dividing the sum by two ( (STR+DEX) / 2 ). Encumbrance penalties apply.

Athletics deals with all types of movement. Activities associated with athletics includes running, climbing, swimming, flying, riding, acrobatics, balancing, and intimidation. Athletics also influences your base movement speed.

and then I have write ups for adjudicating climbing, swimming etc. I might give general bonuses, something like 'for every 3 ranks in athletics increase your base speed by 1', and i also like PF2 idea of skill feats, so maybe award someone a new feat in athletics every 2 skill points they have in it, so someone could become super good at jumping(10x your jump), or holding their breath, calculate your athletics bonus as only STR or only DEX instead of average etc.

That's the basic framework.

[Neo Phonelobster Prime]

Anyway, in my experience, whichever stat system is optimal, I've found the ones that split "dodge" from "ranged attack" are ideal, but as far as The Mage Stat vs. Social Bonus goes, simplicity is best.

Both those things are symptoms of more complex dynamics.

Splitting dodge (related base attribute) from ranged attack (related base attribute) is not inherently good or bad.

Now traditionally in D&D it would maybe be a minor improvement because melee fighters have their attack and defense base attribute contributions split, while ranged fighters do not. Sort of. So you are making the MAD for the two roles a little fairer.

But Multiple Attribute Dependency is probably at best a tolerable bad thing, but is definitely a very bad thing at it's worst. Making everyone "equal" in how much MAD effects them is certainly "fair" but if you balance at a level of MAD that is itself bad (and that might be any level at all) all you have done is go from having SOME roles/classes/??? crippled by MAD to having ALL your roles/classess/??? crippled by MAD.

Further one of the biggest problems with Base Attributes period is how they generate the same attribute profile over and over again with cookie cutter character clones. Characters dependent on a single primary attribute can at least "mess around" with pretty much everything else. If everyone has even several attributes they MUST invest in to be effective because everyone is a MAD character you will end up with even more of a cloned character attribute profile problem.

Without further clarification it would be equally rational to claim that reducing the MAD for melee fighters instead of increasing the MAD for ranged fighters is the "correct" course of action to create richer more varied viable character profile options.


As for Social attributes being "best" tied with magic attributes.

Yes and no.

Because Social stuff is usually tacked on unimportant and mechanically pretty terrible, you know traditionally and because generally it is something that either all characters, or at least some character/s in the party are just expected to do as well as the actual main thing they really do it IS best if it is tied to one or more base attributes that actually do "real" and important stuff.

But that's mostly just because if you have some base attributes that do things that really matter, and some other base attributes that do things that definitely do not really matter you've just created a slum made out of dumping attributes.

Any and all actions and activities that share the same frequently informal, frequently broken and non-functional, and generally unimportant position as social stuff normally fills has this same dynamic occurring with base attributes. Knowing and accounting for it is nice, avoiding it in the first place would be nicer.

And for that matter while social being unimportant, is best lumped in as a free extra with something actually important, there is no particular reason it should, or shouldn't, be magic specifically. Might as well be lumped in with Muscles or Communism for all it matters.

[deaddmwalking]

Anyway, in my experience, whichever stat system is optimal, I've found the ones that split "dodge" from "ranged attack" are ideal, but as far as The Mage Stat vs. Social Bonus goes, simplicity is best.

Making everyone "equal" in how much MAD effects them is certainly "fair" but if you balance at a level of MAD that is itself bad (and that might be any level at all) all you have done is go from having SOME roles/classes/??? crippled by MAD to having ALL your roles/classess/??? crippled by MAD.

Further one of the biggest problems with Base Attributes period is how they generate the same attribute profile over and over again with cookie cutter character clones. Characters dependent on a single primary attribute can at least "mess around" with pretty much everything else. If everyone has even several attributes they MUST invest in to be effective because everyone is a MAD character you will end up with even more of a cloned character attribute profile problem.

I don't think that this is necessarily true. First off; much of the game is relative ability. If everyone is 'crippled', but so are the opponents, then you're just starting a little lower on the power-treadmill. We'd have to agree on what makes a character 'crippled', but I'd go with a definition of 'completely unable to fulfill the functions of their role'.

Likewise, if every stat is important and people need to make necessary trade-offs, cookie-cutter characters are unlikely. One melee-character might choose more toughness while another one chooses more defense - if both are equally valid choices than you're likely to have variation between builds.

To clarify - as long as there is a strict hierarchy of relevance of stats, you're going to see characters in the same roles with the same attribute profiles; when every stat is equally valuable you'd effectively expect a uniform distribution of stats. Making every stat 'equally valuable' may not be possible across all roles, but your conclusions do not naturally follow if that were to be achieved.

[Neo Phonelobster Prime]

I think anyone can agree there is at least a point in MAD where a character is genuinely crippled by it, where there just aren't enough attribute resources to go around for the final result to function. I think people would disagree as to the exact point where. But I also think anyone should be able to accept it would be at some point.

Equality does not guarantee functionality. Even if everything, including opponents, was likewise equally crippled, you just end up with a mockery of a game where all events are about failure and ineptitude on behalf of all parties.

And where do you get the "even the opponents" bit from? A traditional system made out of base attributes and class/role/??? by no means suggest "opponents" work the same way or even have classes or attributes. In fact the opposition may not even resemble characters or monsters, sometimes it's just a DC or an obstacle or a trap. Equality of MAD for PCs tells us nothing about anything about the opposition.

Meanwhile making all attributes equally viable to all characters, at some sort of hypothetical unachievable extreme could, I guess could result in a situation where everyone who wants optimal effectiveness always picks nothing but 12s... but... probably not.

More likely you will not achieve abstract perfection, much less especially edge case twists on it, and instead your classes will just have a strict hierarchy of MAD for each class/role/???. And what you get is fairly predictable attribute profiles with fairly strict hierarchies of priorities for your attribute allocations. We are very familiar with this from 3.x D&D, we all know that a Con dump wizard is a daring edge case build likely to explode in your face while Cha dump wizards are dime a dozen and extremely viable.

Another more likely scenario if you could approach the abstract perfection of "all attributes of equal value to all classes" is the creation of meta-roles. Then sure, you maybe end up with a DPS-Wizard that has a dramatically different stat profile to a Tank-Wizard and maybe both are viable. But your Tank-Wizard and another Tank wizard are going to be basically the same.

And of course, what is likely as you move closer to perfection is much more important than what might happen if you achieve perfection. Because we know MAD is a pile of poop, the idea that possibly if you could just exacerbate it until an abstract point of total perfection and then and only then in would turn into something good... is not a reasonable goal.

I think it is very, very, reasonable to suggest that MAD is a bad thing, that in general more MAD is more of a bad thing, and that any design plan that revolves around adding more of a bad thing and requiring perfectly balanced application of that bad thing is a design plan setting itself up to fail, and fail explosively.

[deaddmwalking]

I continue to disagree. MAD is not, in the abstract, bad.

In the real world, people want to be rich, funny, smart, suave, athletic, etc. Some people are all those things, some people are not. If in your game all of those things are important and desirable, even if players must accept trade-offs that isn't necessarily a bad thing.

In the abstract, if you have 3 attributes (call them A, B, and C) and resolution is simply comparing numbers (ie, I 'attack' your A, and if mine is bigger, I deal a negative condition, but if mine is lower nothing happens) you have incentives to raise all of your stats above the level of your opposition but also have a more powerful incentive to ensure that one of your attributes is as high as possible. In any case, it doesn't actually matter WHICH attribute is your highest/lowest until you encounter an opponent that you're either safe from or unable to meaningfully affect.

Third Edition is not a good example of MAD. Wizards bump Intelligence at all times, and every other stat is 'nice to have'. Monks on the other hand need Str/Dex/Con/Wis. Theoretically, if the cost of an 18 is equivalent to four 16s, MAD isn't really an issue. It's just a matter of costing the abilities appropriately. Likewise, using INT for DC, but adding other attributes (like CON for spells/day and Wis for highest level spell) doesn't imply that a Wizard can't be effective.

[Foxwarrior]

Then sure, you maybe end up with a DPS-Wizard that has a dramatically different stat profile to a Tank-Wizard and maybe both are viable. But your Tank-Wizard and another Tank wizard are going to be basically the same.

Yeah this bit is basically the best you can hope for, really... the ideal that a balanced attribute system would achieve if done reasonably well.

The thing is that ability scores have such simple outputs, that if there's any way to tell which stats you need for what you want to do, you just put your points there and you've solved it, congratulations, a charopper is you. So trying to make ability scores interesting is a bit of a fool's-errand.

It's kinda funny how pretty much the entire character creation system for D&D is about trying to generate a clear stereotype for your character isn't it, you have mental ability scores not because they're interesting, but because that makes it easier to say "wizards are bookish types who know a lot of things, while clerics give sage advice without knowing as much". Almost like the character creation system was designed for some sort of role-playing game...

[Neo Phonelobster Prime]

Third Edition is not a good example of MAD. Wizards bump Intelligence at all times, and every other stat is 'nice to have'. Monks on the other hand need Str/Dex/Con/Wis.

What.

That IS a good example of MAD. The wizard has less and the monk has it pretty bad, before also sucking for all other reasons as well (interesting how most MAD character archetypes do that hey?).

But also.

What.

The wizard in 3.x is NOT "and every other stat is nice to have".

They are a fairly good example of MAD at what is probably a practical rather than a hyperbolic best case scenario. And yet...

The 3.x wizard has a fairly strict priority of attribute dependency. Its not going to suck to death for not having wait, holy shit, FOUR SIXTEENS just four sixteens hey... you really... FFS. Look whatever, lets not get hung up on that piece of ludicrous.

It's probably easiest to see if you are rolling stats and then allocating them to attributes. At which point you almost every time allocate in a strict order from highest to lowest...

Intelligence
Constitution
Dexterity
Wisdom
Charisma
Strength

Now, you could vary it slightly. Dex could come before Cons, Str could come before Cha. That's maybe two choices that would be worth maybe a +1 difference each. But even that difference... is probably technically a net loss in optimization.

You can discus various bullshit highly specifically builds that can make something stupid like a Strength wizard at least slightly viable, but they are not just edge case but have even stricter build requirements, depend on very specific spells and feats, and generally are NOT viable through all levels of play like the above default stat priority for wizards in 3.x definitely is.

In practical terms if you didn't want your wizard to sooner or later suck, you used the above attribute priority list I've mentioned. Any variation at all in the exact numbers on any given 3.x wizard that wasn't gimping itself was exclusively down to possible use of a random attribute generation method or the needlessly high granularity of the base attributes to the point that it didn't REALLY matter if you are only +1/-1 bonus off perfect.

I though everyone knew this?

I seriously didn't think anyone would say "3.x wizards, you can put their attributes ANYWHERE, mine goes Int, Cha, Str, Wis, Dex, Con!" that's a god damn nonsense gimp is what that is. Just because you recognized the primary attribute doesn't mean you didn't build a gimp.

Oh, and I'm fine with trade offs.

But. The 3.x edition wizard, probably close to the best you will manage to achieve trying to balance attributes and MAD, punishes you for having a wizard that is also charismatic regardless of, indeed BECAUSE of what you have to "trade off" to do that. And that's attributes arbitrarily punishing viable and desirable character concepts and that is in the end THE core problem of base attributes, why they are a needless burden mechanically, and more MAD absolutely will never and can never fix that.

[JonSetanta]

Concerning social stats being a thing in too many RPGs, I'm thinking lately that you could very well have an RPG that has NO stat for "bonus or pool of dice for social checks", instead using situational modifiers that add up to the same thing depending on reputation, timing of conversation relative to other campaign events, in-combat or not, or how many beings the persuader/liar is attempting to influence.

If an RPG writer really, really, really wanted to slap on some kind of skill or numeric bonus, just make it level based and call it done.

Also, a decade ago I wrote a Mage class that was mostly a Generic Spellcaster with MAD: Int determined the number of spells readied each day, Wis was a mana bonus, and Cha added to the Save DC.

[deaddmwalking]

A Wizard is the definition of SAD - Single Attribute Dependent.

Yes, you have a hierarchy of secondary abilities, but if you have 1 18 and all 10s, you don't care, because you have the single attribute you need in the place you need it.

MAD - Multiple Attribute Dependent - needs multiple high ability scores.

That said, you've still not made a case for why MAD must suck and more of it is worse. You're just arguing that your preference is a universal truth that anyone should recognize.

I disagree.

If you use Attributes, you should want players to want to have good attributes in multiple places.

It's not a problem that people who want to do the same thing in the same way should prioritize the same abilities; it is a problem if all characters look exactly the same. It's okay if all bookish wizards are smart; it's not as okay if they're all tough/smart or all dodgy/smart. It's much better if some bookish wizards are dodgy and some are tough.