I’m surprised by how much I like the design space of feats in 5e, and how different they are in function and concept from 3.x and 4e. Somewhere or other I saw them described as small chunks of multiclassing, and that is about right; in this they’ve preserved some of what I liked from Specialties in the early playtests. Each time you get an Ability Score Improvement from your class progression (4th level, and 4-6 times after that, depending on class), you get one of the following:
- +1 to two ability scores of your choosing
- +2 to one ability score of your choosing
- +1 to a specific ability score, and a modest new ability
- A bundle of new connected abilities
I think it helps a lot that they introduced the “middle ground” of Option 3, making the whole choice seem less like a “binary” of either ability scores or something interesting. Anyway, some feats cover abilities that you might otherwise spend a level or three of multiclassing to acquire, though it will often be in a watered-down form, as with Magic Initiate or Martial Adept. Others support niche concepts (Crossbow Expert) or cover noncombat functions that would be an excessive power creep if they were standard options of existing skills – often things that were separate but little-used skills in 3.x. For example, Actor covers everything that a Disguise skill might, without making Deception or Performance still more potent. Observant, likewise, covers for a Read Lips skill.
I do think Wisdom (Medicine) is short-changed, though – if the functions of the Healer feat were the standard application of a Medicine check, that would not be too far out of bounds. After all, a skill proficiency is nearly as precious of a character resource as a feat.
It’s also interesting that, unlike 4e’s and 13th Age‘s approach to feats, none are explicitly class-linked. I have no problem with how 13th Age uses feats as chances to juice up some of your class talents (along with a tiny number of class-agnostic feats), and my only problem with the way 4e does the same thing is that they create way too many if/then statements to resolve for attack, damage, defenses, and damage taken. In 5e, though, there are very few prerequisites – no ability score requirement higher than 13, and otherwise only armor proficiencies and the ability to cast even one spell (which you could get Magic Initiate if nothing else).
The worst thing I can say about the overall design of feats at this point is that enough of them are critical to supporting character concepts that playing from first level to fourth without a chance to take some of those feats is rough. (The answer of “just play a Variant Human” is weak – you should never have to play a concept as one particular race just to make the first three levels of gameplay work out.) For example, the game does want to give some modest support to an unarmed brawler that is not a monk, as demonstrated in the Grappler and Tavern Brawler feats. That character – presumably a fighter – is going to have a bad time of it if he sticks to his weapon-less concept for enough encounters to make it to 4th level.
There’s a secondary but valid argument that, well, the fighter and barbarian classes aren’t really designed to do that as written, and at minimum would need a new subclass for support. That is also a solution to this problem, except that fighters and barbarians don’t get their subclasses until 3rd level. This won’t be the post to really address the greater issue of first and second level as tutorial levels that you don’t need after you’ve played a few characters.
Now, I get that the designers wanted to minimize the number of choices a player has to make at 1st level. One of the enduring criticisms of 3.x and 4e is the number of choice points that the game dumps on a first-time player, who doesn’t yet have the context to make a good decision. Anyone who bought 3.0-style Toughness for their first fighter, barbarian, or whatever knows what I’m talking about here. (“I want to be really tough! This feat will help, right?” “…you’d almost be better served to take Skill Focus: Appraisal.”)
A lot more of the 5e feats do what they say on the tin, though. That helps. At least within the initial release, I don’t think 5e has as many “trap” feats. The armor-proficiency feats may be an exception – these kind of need to be cordoned off, as the player needs an understanding of the different armor types and how we calculate AC to make good choices here. It’s not exactly expert-level stuff, but I have some very good players who just don’t engage with the rules on that level. Those feats aside, there aren’t nearly so many wrong choices, and not just because there aren’t nearly so many options. Your choices might be unusual, but as long as you’re applying common sense and not just throwing darts at your book (not recommended), you’re probably going to be okay. Choosing between two points of ability scores and a feat (with or without an ability score point) is still tough.
Anyway, the game kind of needs (and I expect we might see in the DMG) a separate “mode” for intermediate to experienced players, with a feat at first level to support corner-case character concepts. The obvious, if ham-fisted, implementation is to give characters the full statistics of third- or fourth-level characters when they have 0 xp, and just require them to earn their way up to 2700 or 6500 xp to advance.
The point of this post, now that you’ve come all this way, is that I have some new feats to present, which I’ll follow up with commentary on why I made them the way I did.
Bloodletter (New Feat)
You gain the ability to heal wounds and maladies through blood sacrifice in battle, or through leechcraft or controlled bleeding outside of battle. You can gain and spend up to five blood sacrifice points per long rest. When you take a long rest, unspent points are not lost, but you still cannot spend more than five points until your next long rest. Gain one blood sacrifice point whenever you do any of the following:
- Receive or deliver a critical hit
- Deal damage that reduces another creature below half of their hit point maximum, using any damage type other than psychic. You may reduce a creature to less than half of their maximum hit points outside of combat through the application of leeches or careful bloodletting.
Spend blood sacrifice points on the following:
- 1 point: cure wounds as a first-level spell, +1 point per additional level of the spell.
- 2 points: lesser restoration
Finally, while you are at or below half your maximum hit points, any non-cantrip effect you cause that restores hit points heals 3 additional hit points.
Poisoner (New Feat)
You have the ability to modify poisons, both magical and mundane, for the unusual demands of battle.
- Whenever you would deal poison damage, you may instead deal only half damage, but change the damage type to psychic (neurotoxins) or acid (caustic compounds).
- You gain proficiency in the poisoner’s kit.
- You may apply blade poison as a bonus action.
- When you deal damage with blade poison, the poison damage is doubled.
Design Commentary
The point of Bloodletter is that I’m looking back to an idea I had some years ago that eventually grew into Aurikesh: all of the gods (even the good ones) love blood sacrifice of some kind. The good ones just prefer that it be voluntary and, unless you’re deliberately sacrificing your life for a miracle, survivable. I also like supporting leechcraft in games – in both Dust to Dust and Aurikesh, humorism is objectively true. (They’re fantasy games – why wouldn’t it be?) So between the vampiric healer concept (deal damage to heal wounds, that is) and the bloodletting physicker, I wanted to provide support that isn’t in evidence in the rulebook.
I think the limit of five blood sacrifice points is a questionable one – it’s definitely possible that it is too high, compared to the throughput of other feats.
- Healer is the obvious point of comparison, at 1d6 + 4 + target’s maximum number of Hit Dice, once per short rest or long rest.
- Healer is a lot more potent if several people in the party are injured, or if you’re taking multiple short rests in a day. Healer also scales, albeit slightly, by level.
- Magic Initiate (cleric), taking cure wounds as your one spell, is obviously weaker than Bloodletter, though you don’t have to get in a fight to “earn” your one casting of cure wounds.
- Neither Healer nor Magic Initiate so much as open the door to any second-level spell, though some racial abilities do (drow, tieflings, and the currently-previewed eladrin subrace). On the other hand, bloodletting can’t cure disease or conditions without lesser restoration, or a rather more involved rules structure imitating it.
For that matter, I think having to track both points earned and points spent lacks elegance, but I couldn’t think of a better handling for it that didn’t shove a lot more complication into triggered effects, while also sharply reducing the feat’s usefulness by requiring you to spend the effects you’d earned immediately. (If they aren’t spent immediately, that’s just tracking the full effect rather than the point value – “Problem Adjusted.”) Tweaking the point limit and point costs for effects is the main way I’d expect to change this feat upon playtesting.
With Poisoner, I’m looking for a partial fix to support characters who rely heavily on poison damage. Not too long ago, I posted my thoughts on 5e’s heavy use of immunities in creature design. Right now, it would be hard to build a character to rely on poison damage; a wizard with chromatic orb and cloudkill or a rogue with a curiously bottomless supply of blade poison are the two best means that I see. This feat, then, is more in preparation for the introduction of new spells, subclasses, and poisons (the last of which I expect to see in the DMG). I felt that allowing poisoners to change their effects to flesh-eating poisons (behaving more as acids) or neurotoxins (behaving more as psychic attacks) was tolerably justifiable. Cutting them to half the damage they would otherwise deal means that the target’s damage resistances still matter.
What I didn’t care to do was to increase the damage output of poison spells – from my perspective, that would exacerbate the issue, making them too good against non-immune targets and useless against immune targets. The one blade poison in the Player’s Handbook is 100 gp for 1d4 extra damage, so I felt like doubling that in exchange for the feat was not out-of-bounds. If there are highly potent blade poisons in the DMG for bargain-basement prices (which I do not expect), I’d retune this feat considerably.
I pondered giving Poisoner an additional benefit when attacking a target suffering the poisoned condition. I like spells or attacks that key off of conditions on the attacker or defender – it speaks to the WoW rogue in me, what can I say? I decided against it in this case, partly because this too would deepen the divide in effectiveness between immunity and non-immunity to the poisoned condition, and partly because the rest of that idea didn’t really jump out at me. I think there might be some interesting design space for effects that dealt additional instant damage to the target, but also gave the target an additional saving throw immediately – basically giving the attacker a chance to “cash in” more.
These feats had a pass a pretty high bar to justify their existence to me. I applaud the current restraint of content bloat, though the Adventurer’s Handbook (due on 17 March of next year) is going to reveal a whole lot about the future of the edition and their approach to expansion material. One of the few other things I see as a gap in the existing feats is a way to “class-dip” into the ranger for improved exploration abilities – I originally intended to write something like that for this post, but I haven’t yet resolved how to go about it without just duplicating Natural Explorer.