Which is a really strange bias. And is very specific, restrictive, and exclusionary. And doesn't really follow from any principle or precedent that I'm aware of.
I fear we are simply talking past one another. I find having one clear, best choice for each character design decision to be far more "specific, restrictive, and exclusionary" than to have choices that make characters meaningfully different, providing an array of varying strengths and weaknesses which, over time, play out to relative equality of contribution. And, as I said above, pretty much every edition change, and many errata/rules alterations, are aimed at enhancing that balance, in my experience. For someone who claims to be widely read about the hobby, you can't be paying much attention if you have overlooked this.
A pretty harsh characterization.
So is dismissing a class as "a silly concept".
I don't know a lot of D&D characters that fit it. Again, a bard can be somewhere in between a sidekick and the main combatant.
You keep coming back to "combatant" where I use "contributor". This again seems to suggest you prioritize melee combat as by far the most important aspect of the game. This also explains a party made up entirely of martial characters - since melee combat is the common path to success, the players gravitate to characters skilled in melee combat.
If the game features a wide array of challenges, not all of which can be resolved by dealing damage, many of which are best resolved in other ways but could be resolved by killing whatever stands in your path, others which could be resolved equally effectively by combat or by some other approach, and still others for which combat is the only viable approach, then it is a much richer game, and it provides room for the melee brute (who shines when combat is the best, or only, option), the trickster/negotiator, the spellcaster, the knowledgeable sage and an array of other possible character concepts to each shine in their own areas of expertise.
I'm also sensing an anti-sidekick bias. There's nothing wrong with being one.
Poll your players and see if they feel the same way, or if they want characters capable of holding the spotlight. By the way, I would not classify a character whose main contributions are bolstering and enhancing their teammates as "a sidekick", nor would I consider a character who shines out of combat, solving meaningful challenges through knowledge, expertise, skill or negotiation to be a sidekick. A sidekick is a character who is of clearly lesser importance, utility and value than the main protagonist(s). Their contributions are routinely overshadowed by those of the other characters, not just useful in different types of challenges.
From Wikipedia, "A
sidekick is a close companion who is generally regarded as subordinate to the one he accompanies. Some well-known fictional sidekicks are
Don Quixote's
Sancho Panza,
Sherlock Holmes'
Doctor Watson,
The Lone Ranger's
Tonto,
The Green Hornet's
Kato,
Munna Bhai's Circuit,
Batman's
Robin and
Hercule Poirot's
Hastings.
Annabeth Chase is nobody's sidekick."
Sure. If you run a campaign that favors the bard, the bard becomes more useful. No issues there.
I wouldn't say that campaign favours the Bard so much as it disfavours the melee combatant classes.
It is certainly possible to build an unbalanced campaign. This might be intentional (this campaign will focus around gladiatorial pit combat, so build characters accordingly), or an unconscious bias (antagonists are always designed so they must be fought - negotiation, stealth, etc. never resolve the challenge in the long term; opponents are generally resistant to magic, immune to sneak attacks, and encountered in close range, so melee thrives; NPC's are strong willed and never influenced by interaction skills or abilities; or the converse, everything is far more powerful than the PC's, so combat is a death sentence). The key difference between intentional and unconscious, to me, is that the GM makes his design clear and suggests appropriate characters in the former case. In the latter, the players fairly quickly figure out that some character choices are ineffective, and gravitate away from them - often with the conclusion that such choices are underpowered in general, not just when the game is designed to be unfriendly to them.
I wouldn't. Perhaps the most overpowered PC I ever saw was a barbarian. Or so I thought. Then other characters naturally took over the spotlight, and then others. The beauty of D&D is that it's very difficult to break.
If that is the case, then none of the character choices would be "suboptimal". They would each have their turn in the spotlight. That's not what you're telling me is the Bard's situation, nor that of the Halfling fighter. You have repeatedly stated these will both be comparative weaklings when compared to the half orc barbarian.
I'm inclined to err on the side of "let the player play their cool and possibly unbalanced class" as opposed to "let me tell the player that their character is too powerful and compromise the character in some way in the name of balance".
And yet you are clearly opposed to my SpellSage...
Case in point:
One of the really meaningful changes from 2e to 3e is the cleric. It goes from 7 spell levels to 9, picks up spontaneous healing and various useful combat capabilities, while losing some restrictions (all changes made in the name of creating an "equally viable character" out of a rather misguided belief that divine casters were not good enough). But I let people play them, and they play fine. No big deal. Does the 3e cleric need a downgrade? Probably. But only because it got an upgrade in the name of "balance" that it didn't need.
Again, a lot depends on campaign style. Our group has never set out to break the game, and so the game is not broken. I prefer the 3e/3.5/PF model of the cleric to the "SOMEONE has to play the healer while the rest of us grab the glory" sidekick model we had in the past. I find Pathfinder brought the classes a lot closer together by beefing up everybody, but some classes more than others. It also provided a lot more choice for most classes, such that characters can be better customized (among other things) to compete in the environment set by the GM, whether intentionally or unconsciously.
No it shouldn't. Barbarians are defined not solely by physical might, but also by prowess and survival skill and probably other things. A character built solely around physical might probably shouldn't be all that great.
Those 2 or 3 skill points per level after you dump INT and select a half orc leave pretty limited choices for skills. An extra hp per level is nice, but it largely offsets the reduced AC of Rage and armor restrictions. I have a 3.5 half-orc barbarian who started with a 20 STR. He's very powerful in combat, and he leaves the talking and the thinking to others much more capable than himself in those areas (likely including his horse...). That doesn't relegate any teammates to sidekick status, and it doesn't even make him universally "the best combatant". He gets his turn in - and out - of the spotlight.