I love 5e -- it has a lot going for it as a player and as a DM. But the other thread anticipating 6e got me thinking seriously about what design aspects of 5e are actually problems.
So I felt like posting this thought exercise : if I were the lead designer, and not concerned with consensus or backwards compatibility, what would I consider needs changing to improve the game?
1. Not enough distinctiveness in the player experiences offered by the classes. Too much reliance on spells instead of giving each class unique ways to interact with the game and world. Too much overlap in spell lists. Hunter's mark never needed to be a spell, it could have been a skill-based ability triggered by stalking and studying an enemy. Weapon mastery and maneuvers could have been given to fighters only. And so on! I would redesign every core class to make its abilities serve a specific and unique play experience and trajectory, as much as possible.
Spells are just codified abilities not connected to classes or feats. Once you see spells as abilities that are magic, you understand that they offer a lot of effective use for many usecases. Making 10 different versions of some ability that works like a spell just uses a lot more page space and bloats the game. It is easier for players to all use the same spell, and easier for DM's to adjudicate the effect if they all work the same way.
Sure, some things could be made non-magical, but for a ranger the limiting resource was chosen to be spell slots. It's elegant design. That way players don't have to track multiple resources. (Yes I am aware there are classes that have more resources alongside spell slots.)
Understand that similar to spells, weapon masteries, fighting styles, feats and the like all give players consistency. That way players can change class without having to re-learn the game.
By all means, create your own version of the game, but it is important to understand why they didn't give every class a different way to make attacks and cast spells.
2. Add back choice and consequences in PC design. It is ok for a species to give an ability score penalty. It is ok if you pick a class ability from a menu that is locked in at least for a whole level, not changeable every long rest. Itbis more than ok if dumping strength cripples you in melee and spending your two high scores is an interesting choice but not an obvious one. The game is more interesting because the PCs have strengths and weaknesses, and need to rely on one another.
A very unpopular, old style kind of design. I would agree that "dumping" certain stats should have more consequences, but I don't see why we should go back to giving penalties. I would love to see better design in the core stats, so that Strength, Intelligence and Charisma have more use.
3. Surprise needs to be dangerous. If it is practically consequence-free, then you've removed one of the major functions of the exploration pillar.
We've been there, it was too swingy to be fun so they did it the new way. I think you can still use the old way if you prefer that, but I think as a game, the 5.5e way is better.
4. Beef up exploration. A solid chapter in the DMG with many examples of exploration/survival challenges. Cover dungeon, wilderness, and urban exploration. Explainnhow to run them with skills, new subsystems or both, and how some class abilities can change the nature of these challenges without avoiding them altogether. For example, maybe when a ranger fails a tracking roll, they get a "no AND" result instead of a simple no. It is OK if some classes can access tasks that others can't, or obtain unique results.
Exploration should still be optional, but I would also like more guidance on this. No new subsystems please, that would bog things down. There's
Uncharted Journeys if you are into that, I heard it's good. I use their random encounters a lot, and they really make the game better. I already have what I need, but I think the core game could do with a large amount of tables full of varied random encounters.
5. Cut down on the number of abilities acquired at higher levels. It's better to upgrade an ability, especially if it is already one if the class' core and mechanically unique ones.
Hmm, maybe. I am not opposed to this, but in many cases improving an ability means you add abilities and uses anyway. Scaling isn't an upgrade IMO.
Cool ideas! I think it would help if you thought about what the new problems could be. I think you could have a lot of fun with your 6e.
As for my changes:
- Ability scores are gone, only modifiers remain
- Modifiers are unaffected by background or species
- The usage of hands for spells, actions and the like is vastly simplified
- Merge some conditions
- Cut 30% of all spells, especially complicated ones
- Nerf the shield spell
- Add a very simple inventory system
- Weapons are simplified into categories instead of per weapon type (simple light, martial light, simple 2-handed, etc.)
… do you mean that 5e is more elegant simply by virtue of containing fewer skills than 3.5? … even though 3.5 had the characteristics you’re actually looking for 
?
