Actually, as Tony Vargas has brought up in another thread, WotC had design goals laid out for when they were building 5E. One bit of summary:
More importantly, we must look beyond the mechanics of the game to focus on the archetypes, literary tropes, and cultural elements that built D&D. We must build a fighter that resonates as a warrior, not one simply cobbled together with mechanics pilfered from D&D's past. The key game experience of D&D lies at the game table. Our work must start by focusing on the key elements of D&D and the unique traits of a tabletop RPG. The mechanics must support those two factors, not the other way around.
I would say that they failed when it came to Sorcerer. More precisely, metamagic was in the game, and they apparently felt that they needed to carry that mechanic over, and the only place they had that could fit it was the Sorcerer class. Thus the Sorcerer became the repository of a mechanic, rather than be built from its design's intent.
Even the way that metamagic interacts with the spell lists make it feel very patched-on. About half of the metamagics are generally useful, and half are
extremely situationally useful. Even those that are generally useful heavily bias your spell selection. Unless you completely ignore the metamagics, the mechanics end up being what defines your character, rather than be a tool that supports your character.
Sorcerer is often described as weak, even though it's technically not (if built to metamagic's strengths). It also has one of the highest "unsatisfaction" ratings, alongside Ranger. And I think this is largely because they ignored their own design rule, and created a class for a mechanic, instead of a mechanic for a class. (Even Ranger can be seen in a similar light, where the class seems like a pastiche of mechanics, rather than mechanics properly designed for the class.)
Winterthorn said:
They should have based the sorcerer's spell casting on CON rather than CHA. I see well being and stamina fit the sorcerer's story better than strength of personality. With CON as the sorcerer's "fuel" it would be quite logical for them to be able to burn through a few HD to boost their spell points if they want to risk it.
Additionally, I'm thinking to allow sorcerer's to attune to up to four magic items (all other classes limited to three).
There's no reason to think that a Sorcerer should be better attuned to magic items than anyone else.
As for the stats, hmm... At first I thought it wouldn't work. Every class has its stat proficiencies split up such that one comes from Str/Int/Cha, and the other from Dex/Con/Wis. In general, the latter three are the big defensive stats (what almost all spell saves target), and thus are considered bad choices for offensive potential. However Rogue has Dex as its primary offense, and Cleric has Wis as its primary offense, and neither are considered overpowered simply because of that, despite them gaining both offense and defense from their primary stat.
So having a class with Con as its primary may actually not be such a bad deal. Plus, by having Con as the primary, you're putting your main stat value in a stat that has no associated skills, so you weaken yourself in other ways. And thematically it makes more sense, for a class that's often based around bloodlines, to have their power literally come from the strength of their body.
Int - Wizard
Wis - Cleric, Druid, Ranger
Cha - Bard, Paladin, Warlock
Con - Sorcerer
Yeah, it seems reasonable. Probably makes Paladin/Sorcerer a little easier, and Warlock/Sorcerer a little harder.