Exactly. It's my whole point too, only from the reverse direction: playing an against-type class-species combo is going to be unusual, and will present a whole new set of challenges to that character's player.
It's not "unusual." It will be "almost never." That's what you keep
not getting.
How? Either you make those physical differences mechanically invisible (so why bother with physical differences) or you make the classes not dependent on abilities (so why bother with classes).
As noted above: By making it so class AND race contribute.
Because let's be real: if you're hitting the books to study for your
bar wizard exams, you're going to pick up some smarts. That's...just the nature of the beast. The training process for producing a Wizard provides opportunity to refine your mental abilities. And the same can be said for other classes: Rogues must practice their legerdemain their "but he had such an
honest face" front, Fighters must physically train, Clerics swing weapons and study theology, etc.
Being a Dwarf still affects you. Take Dwarf Cleric: maybe being a Dwarf gave you a leg up on the theology work, so you could focus on the weapon training. Maybe it just made you tougher, so you had to choose which side to focus on. But you still had to do the things that naturally result in improved ability
beyond just the things you trained for.
Obviously, this doesn't work for the game you're talking about that you run, because your system is all about punishing folks who consider playing against type and massively rewarding those who repeat whatever stereotypes (IMO,
cliches) the rules were designed to enforce. But for a game in the process of being designed--like, say, "One D&D," stupid name aside--we can do things differently.
Of course the bell curves will overlap. That said, that a Strength-10 Human is average among its people while a Strength-10 Dwarf is a weakling among Dwarves.
Irrelevant.
That they overlap at all is what matters. Because the variance is HUGE. The variance in human strength is huge! If Dwarves are comparable in terms of variance--and, as I've said, sapience, self-determination, and personal identity
ensure that this will be true--then all bets are off. You
can't meaningfully exclude much, if anything, because there are IRL humans who can't lift 15 pounds, and IRL humans who can lift ("clean and jerk") over 500 pounds. The variance is simply too wide, two or even three orders of magnitude.
So are you saying I should power up everything else to the level of what now stands out as overpowered?
No. Well, not really. I think you shot yourself in the foot by making a design with such an egregious flaw, but going back and reworking it is obviously off the table. My preference, as a designer, would be to find ways to compensate that do not require outright banning, because...well, that's a pretty draconian (no pun intended) solution to the problem. Possible alternatives (recognizing that I find some of these
really not good, but better than banning):
- XP penalty for playing a dwarf wizard. This is comparable to the idea that heavy armor is an XP penalty for a survival boost in OD&D: when GP=XP, anything that eats into how much treasure you can pull out of the dungeon is an XP penalty.
- Limited spell selection. Perhaps dwarven physiology alters the casting of arcane spells, perhaps it's a cultural thing, e.g. maybe dwarf education clings tightly to a traditional system of units rather than the modern dozenal system or something.
- Reduced durability. Maybe dwarven physiology and magic don't gel well together--so either you must undertake certain painful rites that weaken you physically but allow magic (kind of like lyrium in Dragon Age), or you accept that you'll never really be a Wizard. (Clerics, naturally, get out free because their magic is divinely gifted.)
- Subrace/variant race creation. There's already good precedent for duergar. Perhaps there's a way to look like a dwarf and act like a dwarf, but actually come from a different lineage that adapted differently.
Those are the only ones that come to mind currently, but I might be able to come up with a few more if given time.
The reason for the ALL in those is that I simply can't justify saying, for example, that only Dwarves who become Wizards don't get anything from high Con; and if I nerf those Dwarves then in the name of setting consistency I have to nerf all Dwarves equally. (thinking on it further I think I'd have the same problem with the idea of 13th-Age Necromancers not getting the same benefits from high Con that everyone else gets)
I mean, that's more or less what I would expect for any kind of "not getting what others usually get." Perhaps the dwarven tradition of wizardry is more runic in nature--and requires those runes be literally carved into one's flesh. Wizardly power, acquired at the price of sacrificing your body. Sounds like it oozes with both creative player potential and dramatic choices down the line (what
other sacrifices are dwarven wizards willing to make for power? Is a propensity toward extremes part of why few dwarves choose to make the leap? Etc.)
Of the above, IMO 4 isn't even worth considering; neither is 3 as I'm specifically trying to fight against power creep rather than add to it. 2 does nobody any favours. But you seem to be arguing for 3, and yet how can 3 happen without up-powering the whole game?
Certainly, we agree on 4--hence why I have offered other options (some of which, I recognize, are more for "design a new game" rather than "adapt an existing game.") Personally, I do actually think 3 is the best choice in the long run--because if you're okay with
one thing being powerful, it seems reasonable that other things should also be that powerful, just differently. (And, honestly, it comes across as a touch overblown, that potentially having a bit higher than usual Con suddenly makes the Dwarf Wizard unstoppable. Strong, to be sure, but utterly outclassing everything else? I'm skeptical--unless the Wizard itself is simply poorly designed!)
My proposal is that we instead do either
5: design the classes from the beginning so "has slightly higher Con than usual" isn't game-breaking in the first place, or
6: adapt around the problem with narrow, tailored solutions, like the ones listed above.