I enjoyed the heck out of 4e, and would love to see a 4e2. Lots of great stuff in this thread! Some musings on what I'd tweak:
- Split power source/encounter abilities from professions and roles and themes and archetypes. So you can have a party where there’s a divine warrior (aka Paladin), skirmishing archer (aka Ranger), primal warrior (aka Warden) and primal spellcaster (aka Druid) who all take the “Ranger” profession to gain exploration, tracking, survival, herbalism, and etc abilities.
(My
Trades and Professions supplement was born out of an initial idea of this split between “Power Source” mainly for encounters and “Profession” mainly for the other pillars.)
(BTW, it’s also worth noting that 4e was very similar to 1e in this regard, where out of combat stuff was much more nebulous and open to freeform interpretation. Personally, I prefer a bit more support for it, hence the ideas above.)
It might be overkill (would have to test), but I think I'd like Power, Profession, and Themes/Archetypes, along with species traits and upbringing (which may include cultural aspects).
- I think I’d like reworked math with less overall spread than +½ level… maybe +1/3 level?
- More granularity to skill training than a one-time +5.
- Continue to embrace what Essentials started with a variety of ways for classes to use various means of power distribution within the AEDU context without having every class being strictly AEDU. Plenty of the ideas upstream in this thread could be cool to see.
- Also continue to embrace the amazing flavor of the classes, as well as their versatility of purpose.
- And ensure that MAD doesn’t bite a class in the butt.
- Reduce the number of feats, and make them be meaningful by adding new options/abilities more than just flat bonuses to something you can already do. 5e isn't bad in this regard, with most feats providing a few cool things thus making them feel significant.
- Remove the need for magic items as +s to hit/defenses (which can be part and parcel of the reworking of the math above) and allow items to be more flavorful.
- In addition, many of the abilities on many of the magic items could also easily be learned or trained abilities. Lots of these could be included in the character’s profession or theme/archtype as noted above. (Many of the GMs allowed us to refluff magic item powers in that way, so your ability to kip up wasn't because of your boots, it was because you learned how to do it, or that resistance to fire damage was because the artificer carved protection runes all over themselves, rather than because they found some cloak somewhere.)
- Slightly less interrupt powers to keep things flowing better.
- For skill challenges, re-work how they’re described. I was fortunate in that I heard one of the designers describe how they used a skill challenge way in the early days before the game was released, and using that as a template all my skill challenges in games were great. Essentially they were always mechanically hidden from the players, simply that there was a situation described and the first person who went started things and then could say who went next (people could pass) around the table so they each got turns. If the players used a resource, it would count as an extra success. Keep going round until the number of successes or failures were reached or until time ran out (measured by total number of player actions). Made for very inventive and exciting scenarios.
- Fully embrace and expand Rituals and what they do. (I gave a Paladin in our group access to a handful of Rituals that mirrored some of the more traditional Paladin spells or abilities, and they could cast using Healing Surges (or residuum if they wanted). They loved it.)
- Spell Lists. Even if we keep spells as powers, no need a power list to every caster class. Wizards and warlocks, for example, could both take "spells from the arcane power list". Perhaps same idea to martials, like A5e maneuver schools.
I'd not be keen on this, as I see these differences as a feature, not a bug. I'd even say to lean even more into how each Arcane (or whichever power source) wielder does it differently and/or has their shtick. So Wizards would be flexible, with a wide variety of spells they can memorize from every day. (Also add options for different Wizard builds that could be built either as a Controller or Striker.) Sorcerers are more innate casters built around a specific theme, with a set of linked powers -- the Elementalist was an example of this, and we could have plenty of others as well. Ditto with the Warlock and it's specific flavor.
Though... that said... when it comes to Divine or Primal spells, I feel there could be more overlap. But I did appreciate the difference between a 'traditional' Cleric (melee + spells) and the Invoker (full priest caster/wizard equivalent) and the Paladin / Avenger divine warrior type. That last one is a place where I could see some power sharing and or perhaps a rework (a bit like the Wizard suggestion above) of a core class with options for either a Striker or Defender.
So I never played 4e, and the reason I never did was because, when I first read through the books, it was really, really clear it was a combat game, to the point that most spells with non-combat purposes weren't in the PHB, they didn't have good or neutral monsters in the MM, to the point that dryads, I believe, became plant monsters, and they didn't even include the bard because bards are far more face characters than combat characters. Maybe they improved on this later on in the game, I dunno--my initial readings turned me off from the edition completely
That is so interesting; maybe it's because I cut my teeth on 1e and had been playing for so long that it never screamed "COMBAT ONLY!" at me. At all. It was more freeform in that area (like 1e), but I never took it as an implication that they thought there shouldn't be the other stuff. (Though, as noted above, I did like the idea of some mechanical support for it, and I did so with my broad Trades and Profession "skills".) Certainly in all the games I played it never felt any different in that regard to 1e, 2e, or 3e.
6- Rework the Controller role so it feels more like a proper role and have the controllers more consistent.
Also interesting to me! At our table we found the following adage quite true: "A striker's role is to do damage. A defender's role is to eat damage. A leader's role is to make people shine. And a controller's role is to make the DM pull their hair out."

The controllers in my groups were most effective in messing up my battlefield plans or denying my grand moments of power in a variety of ways.