While I agree with everything else you say, I simply cannot find any this to be true, there's simply nothing in 5e to suggest any kind of party building to me, I'm in 2 5e games and we basically have a stack of damage dealers and a cleric cuz someone had to heal in each party, and one of those clerics is really unhappy with his role. I tried to build my paladin to be a "defender" but I'm simply not sticky
5e is great, I'm really loving it, but in my opinion losing Defender, Striker, Leader, [Controller] is hands down the biggest mistake WOTC has ever made.
Well, there's two separate things here. The first is that niche protection does not necessarily mean party building. It's more about how nobody is sneakier than the rogue, nobody hits harder than the fighter, and nobody is better at tracking enemies down through the wilderness than the ranger. The classes are actually good at what they're supposed to be good at, it isn't all just taken over by people with cool spells. It seems to me that 5e is based on the assumption that either the DM will deliberately provide good challenges that cater to different character strengths, or the players will play to their strengths and come up with strategies that take advantage of what they're good at (not at a combat level, but more on a larger strategic approach level). I agree that there really isn't much party-building strategy in 5e other than you probably want a healer.
The other thing you say here is that we lost a lot from 4e with the loss of roles. I personally felt that the roles in 4e had several major flaws in their implementation. I would like to see a 4e-style game that handles roles well, but I also don't think we really lost that much going away from roles, since they weren't done very well anyway.
My problems with 4e's roles is that they forced classes into roles they didn't thematically fit, and they never really embraced the roles enough to make them interesting.
A major problem in 4e is that there is no way to build a big, tough guy who focuses on hitting things. You can build a melee striker, but you're shockingly squishy, or you can build a defender who has to constantly think about controlling the battlefield. Nobody fills the role of, "I wanna hit some dudes." The barbarian, for instance, just doesn't make much sense as a striker. The barb is supposed to be tough, not squishy. There should have really been a bruiser role.
The controller role also had significant issues IMO. The controller classes got all their control in their powers instead of features, making it much less clear what their "thing" was supposed to be. At first it was area damage, but then the sorcerer came along and was good at that, and the controller became all about lockdown and forced movement. These abilities, however, are extremely annoying for DMs to deal with, they slow down fights in an already slow game, and they make fights against solos almost impossible to run without ridiculous cheesing from the DM to even things out.
The leader role was clearly defined in its ability to heal, but IMO healing should not have been made such a crucial part of the game. Especially with 4e's mechanics, I don't think there was a need for so many classes with healing as their thing. Plus, healing slowed down combat in an already slow game. I wish they would have beefed everyone up a little bit and made healing a cleric thing instead of a leader thing. Maybe let other classes have healing as an option in their powers, but not as a core feature of a full ~1/4th of all classes.
Strikers were defined by their ability to deal damage, but there was the issue that the role was divided into two fairly distinct groups. The first was melee strikers, the second was ranged strikers. The rogue was designed to easily flip between the two styles, and I think was the best designed striker by far. The ranger could be built to do both, but the natural build for most was to pick one or the other. The problem here is that the ranged striker and the melee striker had very different approaches and needs, but were built on the same class. This led to some of the problems I already talked about, like squishy barbarians.
IMO, the role system should have been heavily revised. If I was in charge of updating 4e, I would make the PC roles closer to the monster roles 4e had, with a skirmisher role, artillery role, bruiser role and defender role. Skirmishers would be characters like rogues and rangers, focused on mobility, utility, and damage at close and medium range. Artillery would be characters like sorcerers, wizards, and maybe a new archery oriented class that focus on doing big damage at range while being squishy. This would cover both area effect and single target damage. Bruisers would be tough melee characters built to both dish it and take it. The fighter would fit into this role, as well as the barbarian. The defender would be basically the same, although even tankier than in 4e. The leader role would be made something that is optional for a number of classes. I would keep the warlord, but make them a defender with a build that has some healing. The paladin would be a defender, and the cleric would be a bruiser, and both of them would have some healing options. I would allow many classes to build in a controller direction by choosing control oriented powers, although artillery and defender would really lend themselves to that style.
This is just how I would do it, not the only way or anything. But I think that the particular roles they went with in 4e caused them a lot of unnecessary trouble in their class design, and it led to a lot of confusion as classes had trouble aligning thematically with their supposed role.
