Bard

1. I feel weird about grouping Bards in with other expert classes because they’re full spellcasters, on par with Wizards. Well, part of my problem is that there are full spellcasters for each class group except Warriors, but I’m also bothered because I think being a full spellcaster is likely to be much more mechanically significant than having Expertise in a few skills unless the game is substantially changed to make skill use a peer to at least low-level spellcasting.
   1. Also, Bards have spellcasting starting at 1st level, but only gain Expertise at 2nd. They feel more focused on spellcasting than on Expertise.
2. The discussion of being able to use legacy subclasses with a new class design (even if it’s only for playtest purposes) seem to impose a pretty low upper bound on the quality of feedback to be received from playtesting.
3. Providing default selections whenever a player would have to make a choice (of skill proficiencies or spells known, for example) is good for a number of reasons:
   1. Helps players generate their characters (though this is in tension with the idea of everyone custom-building their backgrounds from the Character Origins packet…)
   2. It causes the Bard to have Perform by default, meaning a player would have to make an affirmative choice to not have it. That makes a lot of sense to me.
4. Multiclassing rules directly in the class entry: well, it has the advantage of being specific instead of requiring players to try to apply general rules to specific classes. I’m not sure how well that would work in a whole Player’s Handbook, where I think I’d rather look at the general rules for multiclassing to see how they work.
   1. I note that multiclassing is still treated as an optional rule even though feats are being promoted to a standard rule. I’d rather it was the other way around, or that both of them were kept optional.
5. Bardic Inspiration gives a creature a Bardic Inspiration die to add as a bonus to a roll, in contrast to Inspiration which lets a creature re-roll a d20 Test. I think the overlap in terms is unnecessarily confusing, especially when the Musician feat allows the user to give characters Inspiration, and being a Bard lets a character hand out Bardic Inspiration, and a fair number of Bards are likely to take the Musician feat.
   1. I also note that Bardic Inspiration dice are handed out as a reaction, which clearly isn’t enough time for any sort of a performance. I like the idea of bards inspiring people with their performances, but this seems like a mechanic without any real thought to what it looks like in-character – I mean, the bard might not even be in a position to give someone finger guns and say “You got this, buddy”. I’d much rather just give all bards Musician and then leave them to have their 1st-level feat choice to be more differentiated.
6. Bard spellcasting – I wish we could back down from Bards being full spellcasters – I think putting them on the same schedule of spell level gains as Wizards and the like makes those other casters feel less special. I don’t think the spell school limitations are enough to make up for that, though it’s good to explicitly have them because it makes it easier to see what future spells and spell effects are open to bards, though I have no idea why Transmutation is one of the spell schools they have available – it doesn’t seem to match the theme of Bards as I understand them.
   1. Prepared spells – This is probably easier for new players, but it’s once again a buff not just to spellcasters (who haven’t needed any since 3.0), but to bards in particular (who certainly didn’t need any buffs over other spellcasters in 5E)
7. Expertise – this would be more meaningful if the main player-facing book gave concrete examples of what specific skills and tools accomplish at specific DCs. It would also help if DC tiers weren’t 5 points apart, since Expertise won’t make that big a difference until 13th level – at 2nd level when this is gained it’s only worth a 2-point difference, and it’s hard to feel the value of this feature on anything other than opposed rolls like Perception and Stealth, or Deception and Insight. What can Performance do at various DCs, or Persuasion, or a musical instrument proficiency?
   1. Part of the reason expertise doesn’t feel like much of a class feature: compare how much space is dedicated to Expertise in the class write-ups. Now compare how much space the Ranger’s relatively limited spellcasting takes up, both in the spellcasting description and in the table of class features. That presentation sure makes it seem like the game cares more about spellcasting than about expertise, or about being good at skills in general. 

Ranger

1. Rangers in general – rangers are originally supposed to be about surviving, exploring, and traveling through the wilderness. As long as 5E doesn’t care about those activities – and it hasn’t so far – any Ranger class will be pointless at best.
2. Favored Enemy – finally an implementation that doesn’t amount to game mechanical bonuses for racism. But this ignores so many of the usual rules for a spell, I wonder why it’s even treated as a spell – just make it a non-magical class feature.
3. Spellcasting – so Rangers get this at 1st level now. And they get cantrips, which means they have an unlimited amount of magic. That’s kind of odd to me – like, a Ranger can just decide one day that they’re going to prepare Produce Flame and then they don’t have to worry about firebuilding skills (or ammunition for close-range attacks). Also, they can have Goodberry as a 1st level spell, so there’s no need to think about foraging for food for the party – once again, magic makes mundane skill pointless.
4. Roving – giving rangers Climbing and Swimming speeds is the first change made to the ranger that makes them feel more like skilled wilderness travelers. On the other hand, all it does is make them move a bit faster – it doesn’t give them an improved chance of success when attempting to swim or climb is difficult. That’s a larger issue with Swim and Climb speeds, though.
5. Tireless – the temporary HP seems pretty minor – it’s essentially 1 free HD of healing per short rest. Good if you’re taking those all the time. Reducing exhaustion by 1 level every short rest – and in particular without food or water – defeats a large part of the value of exhaustion as a mechanic. It would be better if this didn’t work if the character had suffered exhaustion due to deprivation.
   1. It’s pretty strange to see a short rest-based mechanic given the way the rest of the the edition is trying to move away from abilities that refresh on a short rest. It almost feels like this got copied out of another source without being updated to the new design philosophy.
6. Nature’s Veil – another magical ability, this time relying on 5E’s second-favorite innovation: the completely unexamined concept of “spirits”.
7. Hunter
   1. Hunter’s Prey was pretty bad before and it still feels bad now – an additional 1d8 damage once per turn against a creature which is already wounded is both not very mechanically significant and not very useful for hunting small prey. I’d think this would be better handled by something like sneak attack.
   2. Hunter’s Lore – This is actually decent and thematic and arguably a reason to keep Hunter’s Mark magical.
   3. Multi-attack doesn’t have much to offer for melee-focused characters. It’s also kind of a weird concept for a Hunter, since it involves an area attack on their surroundings – like they’ve ended up in the middle of an area potentially full of targets. But really it feels more like the hunter is the one being hunted.
   4. Superior Hunter’s Defense doesn’t seem like an ability that’s useful in actual hunting (which means I don’t see how the character would come by the ability), since it requires the character to be fighting a group of enemies

Rogue

1. I’m nonplussed by the suggestion rogues take Acrobatics but not Athletics, because I keep forgetting that people generally don’t have to roll to climb things (bearing in mind that Climb Walls was classically a Thief/Rogue skill). I’d personally prefer this was handled by creating very low Climb DCs, ones that very poor climbers (aka weak characters without Athletics proficiency) might not be able to pass, and allow someone beating a climb DC by some fixed amount to treat the climb as not being difficult terrain – but none of this is an issue with the playtest changes.
2. I see it’s no longer possible to take Expertise in Thieves’ Tools, or in other tool proficiencies
3. Evasion has moved back to 9th level, 2nd archetype feature to 6th. Other subclass features moved up from 13th to 10th, and 17th to 14th.
4. Blindsense replaced by Subtle Strikes. This turns the rogue into more of a team player, since it gives them free advantage when supporting party members in melee with an enemy (note that the Rogue would qualify for Sneak Attack either way)
5. Thief
   1. Fast Hands – didn’t this used to have Use Object as an option? If it was removed, you need to say why (at least in playtest documents) so people understand both that it’s an intentional change and the rationale; otherwise you’ll get a bunch of noise in the feedback about a design choice that’s likely to be treated as an error or a whim.
   2. Second-Story Work
      1. This seems like one of the issues of class-defining abilities not coming online until an archetype is selected, around levels 2-3 for many classes. That’s OK, but I think it requires some explicit discussion of prior levels as being apprentice levels, and perhaps a wider discussion of that concept and why it’s included (to allow player to get used to the game and how the class works, for example)
      2. I notice that a Climb speed never obviates the need for an Athletics check, so it’s possible to end up with a surface Conan could theoretically climb faster than other folks – except they make the Athletics check to climb it and he doesn’t.
      3. The ability to use Dexterity rather than Strength for jumps is thoroughly unimpressive. At 3rd level this is at best a +4 difference, but at least this makes some (minor) difference in how much distance you can actually jump
   3. Supreme Sneak now allows you to move at full speed, but doesn’t work if you’re wearing medium or heavy armor – in effect, discouraging players from multiclassing in a way that tends to dilute the archetype
   4. Use Magic Device.
      1. I like the new version, especially compared to the old one. The ability to bypass class restrictions tended to hurt niche protection, and the ability to bypass alignment restrictions felt like…I don’t know, an ability to tinker on a metaphysical level? That could be a valid character ability, but doesn’t feel right for a Thief.
      2. The ability to attune to more magic items is straight out a power boost assuming characters have access to enough magic items
      3. The ability to avoid using charges 1 time in 6 feels pretty minor – it won’t come up very often, and items tend to regain some charges each day. It’s also worth noting that “charges” have wildly different values in some items – for a Wand of Magic Missiles, a charge is worth 1 spell level; for a Luck Blade or a Ring of Wishes, a charge is worth 1 Wish spell. I’d probably want this to trigger more often, at least for some items – maybe 50% of the time if the item regains charges on its own, but still 1 time in 6 if the item doesn’t regain charges, like the various wish-granting items
      4. The ability to use scrolls is good, though I don’t see why there would ever be scrolls of cantrips – at-will casting makes those pretty redundant. The ability to use higher-level scrolls via an Arcana roll using Intelligence as your spellcasting attribute is interesting, especially when you end up using spells usually not available to Intelligence-based casters. This does create an incentive to take Expertise with Arcana, but your last Expertise pick is at 6th level and I don’t see anything about being able to retrain Expertise decisions or skill proficiency choices.