Reimagining the Rogue

 I love Rogues on a conceptual level, and in other RPGs, I've often gravitated toward them as a kind of secondary archetype (or tertiary - I tend to like heavily armored melee/magic hybirds first and then some kind of spellcaster, often with a "dark" vibe second). In WoW, for example, my Undead Subtlety Rogue was the first I ever got to the then-level cap of 70 in Burning Crusade (19 years ago!) In D&D, at least in 5E, Rogues play a sort of weird role in a group. They're designed primarily as a utility class, but without the magical spells that a Wizard or Artificer brings to the table (though Arcane Tricksters, of course, lean back this way). Thanks to having more skill proficiencies, expertise, and eventually (and coming online earlier now than it did in 5.0) Reliable Talent, they're the "good at skills" class.

Personally, I don't think this really does the premise of the class justice. While it's certainly a cool thing to have, I A: don't think that this is necessarily the first thing you think of when you think of the Rogue as a fantasy archetype and B: seems to justify disempowering them in combat.

To be clear, the math on Rogue damage isn't all that straightforward. For one thing, Sneak Attack damage goes up every two levels, so there's less of a clear "breakpoint" in their damage output. This damage isn't necessarily consistent either: certain combat scenarios will take Sneak Attack off the table, such as if there's any source of disadvantage. Because it's such a giant chunk of a Rogue's damage, though, it also means that if we can find ways to make multiple attacks, this can boost our overall damage output significantly because as long as we hit with one of them, most of our damage is going through (I say most, but this doesn't really become the case until we get to around level 10 or 12).

I think that if we want to make the Rogue work in combat, we've got to make big changes to Sneak Attack - honestly, tearing the whole thing apart and building something with the same name might be the best option.

But, let's not get ahead of ourselves:

One thing I do think is worth preserving is that Rogues should be using lower-damage-die weapons. The classic archetype is a character who wields daggers and shortswords, and the Shortbow is a better fit for a sneaky Rogue (while Longbows are Heavy and thus can benefit from Great Weapon Master, I'm inclined to believe that without that, the fact that Shortbows have Vex make them usually the better weapon in 5.5).

Frankly, the lack of proficiency in higher-damage weapons is the only "price" I think Rogues should pay for getting some kind of source of additional damage. (Rapiers are actually toward the higher damage numbers, though I think probably most Rogues who are going for melee at all are probably dual-wielding a Shortsword and Scimitar).

Part of me wants to create an incentive to wield a dagger at least in the off-hand, but I think we'll just let the Thrown property do that for us (and maybe the Piercer feat).

Crucially, though, this means that the thing we're no longer taking away from the Rogue is that massive boost that all other martial characters (even some subclasses of not-typically-martial classes): Extra Attack.

Honestly, I think this will go a long way toward helping Rogues catch up. It's not enough to fully replace Sneak Attack, obviously, but it's a big start.

But what beyond that?

Well, now we need to address Sneak Attack.

In a vacuum, Sneak Attack adds 3.5 to 35 damage per turn over the course of your leveling journey. The actual number is a little more complicated because of the chance to hit and the chance to miss, as well as the chance that you're not actually going to get it. Actually calculating the chance you get Sneak Attack is kind of impossible, as all manner of scenarios might present themselves.

So, let's consider our first change:

You just get to choose to get Sneak Attack. No requirements to have advantage or an ally nearby, and you don't lose it if you had a source of disadvantage on the attack (as I interpret it, even if you cancel out disadvantage, if there's any source of disadvantage, you're SOL).

What impact would this have, then? I'm going to do a very quick-and-dirty version of my "vs Wyvern" calculations here: if we had a +9 to hit (not unreasonable for a level 10 Rogue) and were making two Shortbow attacks (we're ignoring Vex for brevity's sake) we'd be hitting on 5s or higher, getting an 80% chance to hit. Two attacks would mean 1d6+5, or 8.5, twice, which gives us 17 80% of the time and a crit bonus of 7 5% of the time (usually I do the attacks separately, but it's just a factor of two either way) so that's 13.6+.35, or 13.95. Then, our Sneak Attack at this level is 5d6, or 17.5 average. The chance we get it at all is 96% (the chance not to would mean the 20% miss chance happening twice) so that's 16.8, and then the chance we get a crit with it is 5% (chance on the first attack) or 5%x20% (chance we crit on the second attack after missing on the first) which is conveniently a 1% chance, so that's a total of 6%x17.5, or 1.05. Add it up and Sneak Attack is giving us 17.85, so when added to our normal damage, that's 31.8.

That's actually pretty decent damage, and it's even lowballing it because of the effect that Vex will have on it.

    So, if that satisfies you, we can just kind of stop here: just say you can do Sneak Attack once per turn when you hit with an attack. And, of course, give Rogues Extra Attack.

But I actually think this could be something of an overcorrection.

One of the issues we run into, I think, is that Great Weapon Master casts such a long shadow over martial damage. I'd be inclined to think that this might, for example, leave a class like the Barbarian in the dust (though unlike Rogues, Barbarians often get additional sources of damage from their subclasses). If we were looking at only a vanilla Barbarian who just relies on their Rage bonus, Sneak Attack very clearly outpaces it.

Part of what I think balances out the several d6s of damage Rogues get with a Sneak Attack is the fact that they only get one attack (unless they're dual-wielding). But I also think this element of design has really held Rogues back - there's a certain "feel bad" element to the class even if the numbers did line up because of what I talked about in this post. In short, that more attacks "smooths out" your damage output by effectively creating a larger sample size.

My initial inclination was simply to lower the amount of damage a given Sneak Attack does, but then slap it on every attack. If we were adding, say, our PB (or the equivalent scaling with Rogue levels) number of d6s to each attack, this would ultimately wind up giving us slightly more damage by tier 4, but it would work a bit more like Rage damage, Radiant Strikes, or even Hex/Hunter's Mark.

If that were too much damage, though, I thought about making it a limited resource. Perhaps it's actually just as much damage, but the Rogue can only do it a certain number of times per Short Rest (feels right that it'd be a Short Rest resource, like a Monk's Focus Points).

But thinking about this in thematic terms, I actually think limiting Sneak Attack to once per turn feels right. Mechanically, this could represent a careful and patient sniper lining up a kill shot or a scrappy street-brawler feinting and ducking under their foe's arm while bringing the blade up into their ribs.

Now, making it a limited resource I do think actually kind of works on theme, but this also might be an unnecessary nerf to a class that already kind of struggles to keep up.

You know, the more I think about it, the more I'm inclined to the simpler solution. One real advantage here is that it also disrupts as little as possible. Cunning Strikes would still work just fine.

Let's take a step back, though:

Ultimately, what this proposal would do is primarily just add a second attack. Sneak Attack is not guaranteed, but I think for the most part, in most combats, you'll have some means to get it (most often just by having a friend near the target, though the Vex mastery might really change this. I just haven't played with a Rogue since 5.5 came out, except my own in a campaign where we weren't allowed to upgrade). This just simplifies things, and I actually think helps the Rogue with brief solo fights, which the class feels kind of inclined toward.

I think we'd need to consider some other breakpoints, though. Given that I haven't been doing a whole bunch of "reasonable fights at level 5" calculations, I don't know how useful it would be to actually do that for this version of the Rogue.

But my main inclination is that by giving them Extra Attack, you'd put Rogues on the same basic chassis as all the other martial classes, and then you could get a broad sense of whether Sneak Attack measures up to the effects of Rangers' and Paladins' spellcasting, Barbarians' Rage and subclass damage bonuses, and such, and as long as those seemed to be in the same ballpark, you'd probably be good.

Phantom Rogue (Horrors Within)

 Rogues have always been one of the more inherently "dark" classes, despite not (at least at base) having magical capabilities. The Phantom's primary subclass feature allows you to deal a little extra damage to secondary targets, which not only helps with the Rogue's actually slightly low damage output (which you might not realize is even an issue for them unless you do a lot of complicated math) but also gives you some "cleave" that martial classes rarely get.

The Phantom is also dripping with flavor, especially when you start collecting trinkets from dead foes.

Now, I've heard that people felt the subclass wasn't buffed as much as it should have been. Also, after a cursory glance at it, I'm not even sure what they changed, so let's go in and compare and contrast and see how we think of the Phantom now.

Level 3:

Wails from the Grave:

When you deal Sneak Attack damage to a creature on your turn, you can target a second creature you can see within 30 feet of the first creature. Roll half your Sneak Attack dice (rounded up) and the secondary creature takes Necrotic damage equal to the roll. You can use this Dex times per long rest.

    Two subtle changes: First, you can only use this on your turn, while the other allowed you to get it on a reaction attack. Given its limited uses, I don't know that that's a huge issue, but it might be bigger later on. The other, even subtler one, is that it used to use your PB rather than Dex modifier. This means more uses early on, but fewer uses in tier 4. That trade is probably worth it. It is only about 3 uses per day, and the damage isn't massive - it caps out at 5d6, which is decent but not huge at level 19. In a tier 1-2 campaign, we're talking never more than about 10 damage.

Whispers of the Dead:

When you finish a short or long rest, you can choose one skill or tool proficiency you lack and gain it, losing it when you use this to gain a different one.

    Flexible proficiency is certainly useful, especially on a Rogue. I suspect most often this will just linger on the same option most of the time. But this is, I think, brand-new, so it's nice to have. Just gotta figure out which skill you're likely to need.

Level 9:

Tokens of the Departed:

You gain two Soul Trinkets (tiny objects that might have a randomly determined appearance/form) that will always linger within 30 feet of you. They grant the following:

Death's Knell: When you deal Sneak Attack damage on your turn, you can destroy one of your soul trinkets to use Wails of the Grave without expending a use of it.

Life Essence: While you have at least one soul trinket, you have advantage on Death and Constitution saving throws.

Spirit Query: You can take a magic action to destroy one soul trinket and cast Augury without spell components and using Con as your spellcasting ability. If you know the creature that the trinket was associated with, you can ask the creature's spirit one question instead of casting the spell. In this case, the spirit appears to you and answers concisely as possible in a language it knew in life.

When a creature you can see within 30 feet of you dies, you can use a reaction to gain another Soul Trinket. You can have a maximum of two trinkets at a time, and this maximum goes to 3 at level 13 and 4 at level 17.

When you finish a long rest and have fewer than 2 trinkets, you gain them until you have 2.

    So, this is a really important mechanic, primarily to let you use Wails of the Grave more often. For that reason, I really wish that you got this at level 3. This is honestly just a problem with Rogues overall (something I really wish they'd fixed in 5.5,) which is that subclasses have to wait an insane time to get their second subclass features. Couple changes here: first, the old version didn't let you cast Augury, which is a pretty nice spell to have (basically to ask the DM "is what I'm about to do a bad idea?") But the nerf here is that you can carry fewer trinkets. It's not always easy to do, but Rogues generally like to have their reaction ready for some kind of opportunity attack, or for Uncanny Dodge, and while I don't think you're going to need to use this every turn, I also wish that it didn't take your reaction to get a soul trinket.

    Ah, one detail I initially missed is that the old version didn't give you any "free" soul trinkets until you hit level 17 (and then only 1 per long rest). So this is nice.

Voice of Death:

You can cast Speak with Dead without a spell slot, and requiring no spell components, using Dexterity as your spellcasitng modifier (why does Augury use Con? Also, in what way does a spell without a DC or attack even need a modifier?) You can do this once per short or long rest. Also, when you cast the spell, you can target one of your soul trinkets instead of a corpse, speaking with the spirit of the creature it's from.

    This is new, and is some pretty cool and flavorful utility. I do think this will mean you're going to have to track where you got your trinkets, and DMs will have to figure out if there's anything interesting that that random Werewolf #3 knew. But it's certainly in-flavor for you to be able to interrogate the dead.

Level 13:

Ghost Walk:

As a bonus action, you can take on a spectral form for 10 minutes or until you end it. You can do this once per long rest, or can destroy a soul trinket to restore its use. While transformed, you get a fly speed of 10 feet and can hover, attack rolls have disadvantage against you, and you can pass through creatures and objects as if they were difficult terrain (taking 1d10 force damage if you end your turn within an object).

    Very subtle change, but it is a buff: you used to have to use a bonus action to end this (though there's no real detriment to using all 10 minutes). The fly speed can be useful, but probably won't see much use in combat. Still, this is a nice thing to have, and pretty flavorful.

Level 17:

Death's Friend:

You get the following:

Death's Lament: When you use Wails of the Grave, the primary target also takes the Necrotic damage.

    This is the same as before. I wish you had the option to burn this even if a secondary target wasn't in sight, which wouldn't deal as much overall damage but would make the feature usable in a solo-monster boss fight. Alas.

Draw of Death: When you roll initiative, you gain a soul trinket if you have none remaining.

    This replaces something that was incorporated into the Tokens of the Departed, but it means that you'll always be able to use at least one use of Wails of the Grave per combat, which is nice. Oddly, I actually think you're better off burning soul trinkets before using your "free" uses of the feature in order to benefit from this.

Overall Thoughts:

So, I always liked the Phantom, even if I was drawn more to the Soulknife when Tasha's came out. This does have a lot of fun utility, and I think that the revision here is definitely a buff, even if some wish that it would have been buffed farther.

The added spells here are nice: Speak with Dead is certainly situational, but can be very helpful. For DMs, I'd say make sure that you're giving the Phantom plenty of fights with multiple foes, both so that they can make use of their extra damage and also recharge their Soul Trinkets.

Rogue (Single Class) Damage Optimization

 The Rogue is a class that I really love, conceptually. While I tend to think of my go-to fantasy archetypes as some mix of heavy armor and arcane magic (basically Eldritch Knights, or Battle Smiths/Armorers if we want to add in some steampunk) I also have a soft spot for the sneaky Rogue.

If you were to ask me how to balance a Rogue in a TTRPG, I'd say that it should be basically a martial/physical-damage glass cannon, meaning that while it shouldn't be as resilient as a Fighter, Paladin, or Barbarian, it should be able to make up for that by putting out massive damage numbers.

Now, I think the design intent behind Rogues' Sneak Attack is to do just that, I think in practice a lot of theorycrafters have argued that the class doesn't really do great damage. And while they're the best at skill checks in the game, that utility doesn't quite match what a Wizard can do - being able to pick locks very well isn't that impressive when the Wizard can just cast Knock.

So if Rogues pay a damage tax for their non-combat utility, I don't think they deserve to do so. I'll note that while I still (still) haven't gotten a game of Draw Steel together, I do really like the way that their Shadow is designed. But they're ultimately very different systems.

The question, though, is whether they really are paying such a high tax.

In my last post, I calculated the damage output of a pure Armorer Artificer against a Death Knight at level 10. This would be a tough, final-boss kind of fight (or at least tier boss) where damage output would likely be the most important. My conclusion ultimately was that they could do roughly 28 damage per round. Artificers are also not known as the kings of damage output, but I'm curious how we'd compare a Rogue with them.

Thus, we'll simplify things and just do a level 10 build and figure out their damage.

Ok, let's talk about our loadout:

The Arcane Trickster is still probably the best Rogue subclass, but it's also something of an unusual one, having limited spellcasting options. At level 10, an Arcane Trickster would have 2nd level spells. That's shy of spells like Spirit Shroud or Conjure Minor Elementals, which can really add a lot of damage if we optimize for making lots of attacks. Getting a cantrip like Booming Blade can give you some more damage as well, especially if you attack a foe and then disengage as a bonus action, forcing them to take the extra thunder damage from moving.

But I suspect that our gameplan is actually going to be to try to get as many attacks in per round, and I'll tell you why:

Sneak Attack, by level 10, is going to be doing 5d6, which is more damage than I think we'd ever be doing with a weapon attack alone (our highest-damage weapon I think does a d8, so even with a Vicious Rapier and +5 to Dex, we're doing 1d8+2d6+5, or 16.5, compared with 17.5 from a level 10 Sneak Attack).

We can only get it once per turn, but if we land just one hit (with the proper conditions) we'll get the whole thing.

Thus, I think instead we want to build around a dual-wielding set-up, getting both a Nick off-hand weapon and the Dual-Wielder feat.

    Weapons:

So, in other words, we're going to go the Shortsword and Scimitar route. This will let us land two attacks as an action, and with the feat, we can get a third attack as a bonus action (I also think technically we can make that third attack with the Shortsword, as long as the Scimitar was one of the attacks of the attack action). These bonus attacks aren't going to hit very hard, because we don't (and won't be getting) the Two Weapon Fighting Style, but the main thing is to give us some help if we don't hit at first to get that Sneak Attack in.

While it would be good to get at least +1 weapons by this level, I'm going to just act as if we're deprived of magic items from our mean DM.

We're not going Crossbow Expert here to dual-wield hand-crossbows because we wouldn't be able to benefit from Dual Wielder, as they don't have Nick.

    Subclass:

If we're not using the Booming Blade trick, I don't know that the Arcane Trickster actually adds a ton of damage at this level. The Assassin has some situational boosts to damage, but to be frank, I don't want to have to calculate whether we have advantage on our first attack because we beat the Death Knight in initiative and take that into consideration for the overall damage output.

Thus, I kind of think we might go subclass-agnostic. Soulknives get Homing Blades at level 9, which could for sure increase damage output. Thieves' Stealth Attack is more useful for a ranged build. So to preserve my sanity, I'm going to not worry about subclass features.

    Feats:

We're going to take the Dual-Wielder feat, which allows us to make an attack with a bonus action if we hit with another light weapon as part of our action. If you're not familiar with the 5.5 update to this feat, the effect is that if we have a weapon with the Nick weapon mastery, we're going to be able to make our initial main hand and then off-hand attack within our Action, which then allows us to get effectively a third attack with this feat. Like the additional attack from the Light property, we don't get to add our Dex modifier to the damage, but this gives us a little bit more damage and a third chance to get off our Sneak Attack if the first two attacks missed. Also, because we did use our Scimitar as part of our action, we get to use our Shortsword for this one and potentially get the benefit of the Vex property.

As a Rogue, we will have a third feat by level 10 (we get it at level 10) and thus we can take half-feats/general feats instead of ASIs and get capped at 20 Dex by this point (assuming we took a background with a +2 Dex bonus in it, which I assume we did. Honestly, Criminal is pretty great, though ironically there's some redundancy with the Alert feat if we go Assassin, a subclass that is better but not where I hoped it would be with the 5.5 revamp). Thus, we can consider two other general feats.

Personally, I think Mage Slayer might be the martial equivalent of War Caster, as the feat that basically anyone who likes either Strength or Dexterity will want, because it gives you a limited Legendary Resistance (though only against mental saves, which is mostly Wisdom). This is a solid option, but for damage, we might consider others.

Piercer is worth considering. It won't work with the scimitar, but we could use a dagger instead if we desperately want the bonuses, and honestly, we'd only be sacrificing an average of 1 damage per turn if we went with a dagger, which also has both style points and can be a ranged option. But we get to reroll a die each turn and get a little more damage on crits. And also, if we find ourselves needing to use a ranged weapon, almost all of them also deal piercing damage, so this covers that as well.

Sentinel, actually, can be really good for a melee Rogue, if you're willing to stay in melee with foes, because the Guardian opportunity attack can potentially give you a second Sneak Attack each round, which is a massive boost to your damage output. (This works especially great if you have an Ancestral Guardian or Armorer tanking for you.) The Opp attack isn't guaranteed - the foes might just target you instead - but if you can find a way to disincentivize them from going after you, it can be great (honestly, Uncanny Dodge is a pretty decent survival ability, somewhat akin to a Monk's Deflect Attacks, though it can't ever fully negate an attack like that can).

I think that, while Sentinel sounds amazing, I suspect that monsters (like our Death Knight) might choose to just focus you down without some outside help and incentive. So I think I'll stick to Dual-Wielder, Piercer, and Mage Slayer. The latter won't really have an effect on our damage, but the others will.

Given these choices, I do think I'd actually go Shortsword and Dagger (again, the damage of our weapon itself is somewhat less important due to the fact that we get a bunch of damage from Sneak Attack).

    Attack Sequence:

I could be wrong, but I think if we want to make two attacks with the Shortsword, we'll need to attack in this order: Shortsword (thus allowing an off-hand Nick weapon as part of the action), Dagger (thus being a light weapon attack as part of the action, allowing the Dual Wielder feat to let us attack with a different weapon as a bonus action) and then Shortsword. While we could break up our action with a bonus action attack, I think that we need one attack to trigger the next. Another possible attack sequence would actually be to start with the Dagger, then the Shortsword (I think this is Kosher, but I could be wrong. Basically, if the Nick property applies to any use of the Light property, and not just the attack with the weapon that has Nick, we could do the Dagger first) but there's not much advantage to this given that we'll still basically always be sandwiching our dagger attacks between two shortsword attacks, even if they aren't on the same turn. I think Shortsword-Dagger-Shortsword is probably the most reasonable sequence.

    Advantage and Weapon Damage:

Oh boy, here's the big question: Sneak Attack is not guaranteed each turn, as we need to either have advantage or a friendly creature within 5 feet of our target.

Now, advantage should be pretty common for us - if we make our first attack on a turn with our Shortsword, we'll get it for our Dagger. We won't get it for our bonus action second shortsword attack, but that should confer it (if it hits) with the first attack on our next turn.

If we get disadvantage, we're in big trouble, as this shuts down our sneak attack entirely. Mage Slayer can help prevent some forms of disadvantage (such as saving against Fear effects) but any Rogue is going to have a terrible time if they're at disadvantage.

For the sake of sanity, again, we're going to want to come up with an abstraction of how often we have advantage. If we assume it's only coming from the Vex property, that simplifies things. So:

At this level, we should have a +9 to hit, and we're attacking a target with an AC of 20, which mercifully gives us a beautifully round 50% hit chance (yes, Death Knights can parry, but let's just ignore that, again, for sanity). If we hit with our Shortsword the previous attack, that becomes a 75% hit chance.

2/3 of our attacks will confer advantage. If these are made straight, then we're essentially getting advantage 50% of the time on one of our attacks and 75% of the time on our other two attacks, so we'll say we have an average chance to have advantage of 67%. It's actually better than that, because hitting on our bonus action shortsword attack at the end of our turn will then make it more likely we hit with the shortsword at the beginning of our next turn.

Our Dagger never confers advantage, so we'll assume that the bonus action attack from Dual Wielder is never made with advantage - oh, this might change what I said in the previous paragraph. In this case, the BA Shortsword is a 50% hit chance (and 5% crit chance), and thus the simplest to calculate.

However, because Vex can carry over from our previous turn, we might have advantage on our first Shortsword attack on a turn because we hit on our previous turn. Thus, 50% of the time, we have advantage, and 50% of the time it's a straight roll.

The Dagger, then, is the most complex, because we're more likely to have gotten the benefit of Vex, because thanks to the 50% chance of advantage on our Action Shortsword attack, we're more likely to have hit.

Let's back up:

For the Action Shortsword attack on turns 2 and onward, our average chance to hit is either 50% half the time or 75% the other half of the time. Thankfully, this just lets us average them, so we get 62.5% to hit. Then, we have a 5% chance to crit half the time and 9.75% chance to crit the other half, so that's an average crit chance of 7.375% chance to crit overall.

For our Dagger, we have a 62.5% chance to have advantage and a 37.5% chance to make a straight roll. So, basically, we're talking 62.5%x75% and 37.5%x50% for our hit chance, which comes to 46.875% and 18.75%, so our total hit chance is 65.625%. Our crit chance on the Dagger, then, is 62.5%x9.75%, or about 6.1% plus 37.5%x5%, or 1.875%, so our total crit chance with the dagger is 7.975% (roughly)

So, let's figure out the damage our weapons are doing. Remember, thanks to Piercer, we're doing a little more on average with each die, and we're also getting beefier crits. The way it's worded, I think we could do the damage reroll on our Sneak Attack if we get it with the Dagger, which is probably slightly better than using it on the d4 from the dagger, but for sanity's sake, we'll assume we can only do it on the weapon die itself.

Action Shortsword:

Hit Damage: 1d6+5, giving us 9.25

Crit Damage: 2d6 extra (not just one extra die, because Piercer) for 7 (yes, we can choose to reroll the die that rolls lowest, so it's a little better, but again, sanity).

Thus, our damage per attack with the Shortsword is 9.25x62.5%, or 5.78 (roughly) plus 7%7.375%, or about .52 damage, and thus a total of 6.3.

Dagger:

Hit Damage: 1d4, which gives us 3 with piercer

Crit Damage: 2d4, or 6 (same caveat)

Thus, per attack, we're talking 8x65.625%, or basically 1.97 ish and then crit damage of 6x7.975%, or .48 ish, so, screw it, let's say it's 2.45.

BA Shortsword:

Hit Damage, this time, is 4.25, as we're not adding Dex

Crit Damage, again, is 7

9.25x50% is 2.125 and then 7x5% is .35, so that's 2.475.

Thus, our total weapon damage, before Sneak Attack, is about 11.2. That's pretty pitiful on its own, but what of Sneak Attack?

    Sneak Attack:

Basically, we're going to assume we're using Sneak Attack the first chance we can on a turn rather than holding out for a crit.

While we will get Sneak Attack if we have advantage on our attack, we'll assume ideal circumstances where we have a friend in melee with the target as well, and that any hit will result in Sneak Attack. This won't always be the case, and we might instead use our bonus action to get Steady Aim or some other means of advantage (you can hide and get a melee attack in, I think, though I realize I might be confusing some of the rules here with the ones from Draw Steel. I'd certainly rule that you could hide and then roll that stealth check into the approach to attack someone in melee).

Sneak Attack at this level is 5d6, and while Piercer could apply to its damage to bump it up a little, I'm going to apply its bonuses instead to the weapon damage, as I have already. This is suboptimal, as we should add an extra d6 on a crit even with the dagger, but just for ease of math, I'm doing it this way.

So, we'll need to consider the likelihood of getting a hit or crit with the Action Shortsword, applying that to Sneak Attack damage. Then, if we missed and thus don't get advantage on the dagger attack, we find the average hit and crit damage, with a coefficient for missing on the first Shortsword attack (because if we did hit on it, we'd have already used up our Sneak Attack). Then, finally, we have to find the chance that we missed on both the first Shortsword and the Dagger attack, and use that as our coefficient on the adjusted-for-hit-and-crit chance on the bonus action attack.

To break it down:

Our Sneak Attack damage is 5d6, or an average of 17.5. Because this element is pure dice, it's actually easier to deal with crits, because with a 50% chance to hit and 5% chance to crit, we can just multiply that 17.5 by 55%. But we don't get to do that just yet.

As we determined before, we have a 62.5% chance to hit on our Action Shortsword, and a 7.375% chance to crit, so our Sneak Attack damage on that first attack is 17.5x about 69.9%, or about 12.23.

Now, there's a 37.5% chance we missed on the first attack. On its own, we have a 50% chance to hit with our Dagger (because we didn't get advantage) and 5% crit chance, so 55%x17.5 is 9.625. But we only do this if we missed on the first one, so 9.625x37.5% is about 3.61.

Lastly, that 50% chance to miss with the Dagger times the 37.5% chance to miss on the first Shortsword attack means we've got an 18.75% chance that we need to hold out hope that our bonus action attack is the one that finally gives us our Sneak Attack damage. After missing, of course, our chance to hit and crit is the same for the same amount of Sneak Attack damage, so that's 9.625 again, but this time multiplied by 18.75%, or roughly 1.8.

    Bringing it All Together:

So, now we just add it all up: our weapon damage from our three attacks came to 11.2ish, and our Sneak Attack damage adds up to 17.64, roughly. Which means that our total damage output is 28.84.

Some Caveats:

Again, we're not truly using Piercing at its best - we aren't strictly limited to rerolling the weapon dice, and I think in theory we could add a d6 to a Sneak Attack crit and not just a d4 if we crit with the dagger. Still, I think these are pretty marginal boosts. The damage should probably be slightly higher, but only by a very small amount.

Unfortunately, there are some downsides: if we don't have a friend within melee range of our target, we will need to rely either on Vex or Steady Aim or some other source of advantage to get our Sneak Attack. While Vex makes it likely we will get it, bad luck on the dice could screw us, and chop into the greater part of our damage.

Now, we are out-damaging my Armorer build by a little bit, so that's good. I'm tempted to try out an Eldritch Knight build, which I think probably outdoes this, but I'm not sure. I'm certain that a Wizard with Conjure Minor Elementals and Scorching Ray blows maybe anything else out of the water (is there a Sorcerer subclass that gets that? Because Innate Sorcery could make that even better).

UA: Magic Stealer Rogue

 The Rogue, of course, already has a subclass that focuses on magic. The Arcane Trickster is a popular option as a 1/3 caster. But did you want to play a Rogue who could interact a lot with magic while not actually casting any spells themselves?

Like the Spellguard Paladin, the Magic Stealer (can't call it Magic Thief I guess because of the Thief subclass) has a kind of love/hate relationship with spellcasters, with some features that disrupt enemy spellcasters and some that can aid friendly ones.

Let's take a look:

Level 3:

Empower Sneak Attack

When a creature within 30 feet of you casts a level 1+ spell (I see we're using more efficient language here) you can use a reaction to absorb some of the spell's energy. Until the end of your next turn, the next time you hit with a Sneak Attack, you deal extra Force damage equal to a number of d6s equal to the spell's level.

You can take this reaction Int times per long rest (minimum 1).

    So, on one hand, we have a similar issue with the Scion of the Three Rogue - a feature that is very cool, but also very limited in use. Not that it makes this better - arguably it makes it worse - but you might not encounter many enemies casting leveled spells, especially at early levels, but also depending on the foes you face. Against a Lich? This will be quite nice. Against an Elemental Cataclysm? Not so much. The damage boost is nice - Rogues honestly need a bit more damage-boosting in their subclasses, as the base class' damage output leaves something to be desired. But it's very limited use - likely 3 times per day until level 10 or so. Luckily, you get a second 3rd level feature, and this one's... maybe really good.

Drain Magic

As a magic action, you can touch a willing creature under the effects of a 1st or 2nd level spell. You end the spell's effects on them, and then that creature regains an expended spell slot of 2nd level or lower (their choice). You can do this once per short or long rest.

    Make no mistake: this lets your friend level up a spell slot.

    While it would obviously be best to use on some hostile spell - someone casts Blindness/Deafness on them, for example - the more likely way for you to use this is to have a, say, Wizard cast Detect Magic on themselves, and then right when the 10-minute duration is up, you end the spell six seconds early and give them a 2nd level spell slot.

    The limit per short rest of course prevents you from going too crazy with it, likewise the spell level limitation (though check back in with me later,) but the ultimate fact here is that you're essentially turning a 1st level spell slot into a 2nd level spell slot, which is cool. And unlike the other feature, you're almost certain to be able to use this regularly in any campaign.

Level 9:

Magical Sabotage

You gain the following new Cunning Strike options:

Spell Susceptibility (cost 2d6): The target has Disadvantage on the next saving throw it makes against a spell until the start of your next turn.

    While somewhat costly for our damage, this can set an ally up very nicely.

Disrupt Spell (cost 3d6): Whenever the target casts a spell before the start of your next turn, they must make an Intelligence saving throw or the spell dissipates, wasting the action/bonus action/reaction used to cast it, though if it was cast with a spell slot, the slot is not expended.

    Couple things: First, this is essentially a preemptive counterspell that targets Int saves. Granted, while a lot of monsters don't have good Intelligence saving throws, spellcasters often can. However, this also works on all spells they try to cast before your next turn, so if they fail to cast a spell with their action, they still also have to save to cast a spell as a bonus action. Another nuance here that I think we need to figure out the ruling on for DMs is this: NPCs don't use spell slots anymore. When a Lich casts Chain Lightning as a 1/day spell, does it get to retain the spell slot if it's counterspelled or prevented with this feature? Or, because it's not a spell slot, is that use still expended? Is this meant to favor players? Or are we meant to treat those X/day spells as using spell slots?

Steal Resistance (cost 2d6): Choose one kind of damage. If the target has resistance to that kind of damage, until the start of your next turn, the target loses that resistance and you gain it.

    This becomes way more powerful if you're fighting foes from pre-5.5 sources that have resistance to bludgeoning, piercing, or slashing damage made with non-magical weapons. That was everywhere in 5.0, and most monsters also didn't have magical weapon attacks. More likely, though, is doing something like grabbing a Demon's fire resistance.

    One hitch here is that a lot of very elemental-themed monsters are fully immune to the damage they're associated with. This doesn't work if you use it on a Fire Elemental, both failing to make them any more vulnerable and failing to grant you resistance. I think my fix might be to say that if they're immune, they remain immune, but you still get the resistance.

Level 13:

Improved Drain Magic

You can now use Drain Magic as a bonus action. Additionally, you can now end 3rd level spells as well as 1st or 2nd as before. The target recovers an expended slot of 3rd or lower now.

    3rd level spells are a pretty huge leap from 2nd, so this is a big bonus. Being able to turn a Mage Armor into a, like, Fireball, is a pretty huge deal. Also, by making this a bonus action instead of an action, it's much easier to use on detrimental spells (though I don't know how often you'll have 3rd or lower spells cast on your friends when you're level 13). Still, I think this is a good improvement to what is already the best feature for this subclass.

Level 17:

Eldritch Implosion

When you use Empower Sneak Attack, you can force the target to make a Con save (DC based on your Dex). On a failure, the spell dissipates with no effect and the target has the Stunned condition until the start of its next turn.

    You know, I actually think this is a pretty cool capstone. A counterspell (though notably, unlike Counterspell 5.5, they don't get to keep the spell slot) that also stuns the target is pretty great. Again, you're only going to be able to do this a few times per day, and your foes are going to need to be casting actual spells for this to work.

    But I like that this effectively disincentivizes them from casting spells - not only might it not work, but they might get stunned. That dragon might elect to stop casting spells as part of their multiattack.

Overall Thoughts:

    Once again, like the Spellguard Paladin, I think this subclass could do very well in campaigns with a lot of spells flying around. But truly, I think Drain Magic has a lot of potential. Even if the subclass doesn't wind up being that powerful on closer analysis, I applaud that it truly feels very different from any other I've seen, even the Arcane Trickster.

Rogue Damage in D&D

 I realize that I might be biting off more than I can chew here, but let's consider the following:

The general consensus among the optimizers of YouTube is that Rogue damage falls behind compared to other Martial classes like Fighters, Barbarians, and Paladins. On a design intent level, this has always felt, if it's actually true, kind of wrong.

Balance in games is always a tricky thing - in a solo game, you could reasonably say that a build or class that leans more toward defense might sacrifice some damage to do so, but if you have a game like, say, World of Warcraft, where survivability is not really a primary concern for anyone other than a tank, it doesn't seem fair that the heavily-armored Fury Warrior would do less damage than the lightly-armored Frost Mage because the Warrior's armor isn't really doing anything to protect them if they're not the one getting hit. (For those who don't play or know about WoW, there's plenty of magical damage that gets tossed around at all players, but the characters playing the tank role should essentially always be the ones that are taking the direct physical hits from enemies and not any of the damage-dealers).

D&D's a bit different - no one is a tank, at least not in such explicit terms. Some subclasses, like Ancestral Guardians or Armorers (or Cavaliers, I think?) can at least incentivize a foe to attack them, but it's not the kind of deterministic thing it is in a game like WoW.

Rogues are a little squishier than most other martial characters - they only have a d8 hit die, and they only get light armor. While they can eventually hit equivalent ACs to medium armor, it takes longer (a level 2 Artificer who makes Smoldering Half Plate could easily get their maximum AC out of it while a Rogue needs to cap their Dexterity to get all they can out of Studded Leather).

Thus, it has always seemed to me, thematically, that a Rogue ought to be among the highest damage-dealers in the game. While a Fighter might suit up in heavy armor to endure the hordes, I think the fantasy of the Rogue is to be the one who takes the risk traveling light in order to go fast and hard to take out their foes.

Martials tend to do better in single-target situations due to the way that the game treats attacking with a weapon, of course (though often it can be demonstrated that spellcasters can overcome this intended balance) and I think the Rogue makes sense to really focus on that single-target damage anyway, striking with surgical precision (though I'll admit that this is really the "assassin" archetype, and you could argue a Swashbuckler maybe ought to be better at fending off multiple attackers, even if its subclass mechanics imply the opposite).

Rogues work differently than other martial characters. They're the only ones that don't get Extra Attack. Extra Attack does a couple things for classes that do get it, initially just doubling their damage potential, but also smoothing out bad luck. You can still miss twice on your turn as a Paladin, but if you have, say, a 60% hit chance, extra attack means there's only a 16% chance you don't hit at all on your turn.

Instead, Rogues get their damage scaling via Sneak Attack, adding a number of d6s to their damage equal to half their Rogue level rounded up.

There are conditions here that need to be met - you can never get Sneak Attack if you have any source of disadvantage (even if it's cancelled out by a source of advantage) and to get it, you either need to have advantage (pure advantage, thanks to the aforementioned clause) or have a non-incapacitated ally within 5 feet of the target.

Now, in fairness, Rogues also have numerous ways to get advantage - by level 2 they can hide as a bonus action to try to attack without being seen, or they can sacrifice their movement and bonus action to get advantage with careful aim. Lots of Rogue-friendly weapons like Shortswords and Shortbows have the Vex property, so one hit will set them up for more hits in the future.

However, in terms of game feel, I think you could argue that this creates a feel-bad situation. Rogues basically need Sneak Attack in order to feel like they're working properly, which places an onus on them that other martial characters don't have. A Barbarian will be more or less just as effective fighting some goblins out in a field as they would in a barn with a lot of posts holding the roof up. But a Rogue is going to have far fewer options for how they want to achieve their Sneak Attack if there's nothing but grass.

Rather than feeling like a cool bonus to their damage, Sneak Attack is the assumption. Rogue damage is balanced around that assumption, and that means losing it leaves the Rogue feeling broken (not in a good way).

But how good is Sneak Attack, then?

One thing I really have to hand to Sneak Attack is that it's one of the smoothest-scaling features in the game. Getting another die every two levels ramps your damage up much more granularly than, say, a Fighter getting an additional attack at 5, 11, and 20.

I'll also say that I think the theming makes sense: if we lean more into the Assassin archetype, the Rogue's goal is to hit hard with a precision attack for maximum devastating effect, rather than some mad frenzy of stabs. Fighters are the machine-gun class while Rogues are the sniper rifle class.

In a vacuum, let's see what kind of damage this come out to.

Fighters seem a good comparison. Generally considered the best martial damage-dealers, I think the Rogue actually ought to match them in power.

While the Rogue has less obvious breakpoints, we can look at those breakpoints for the Fighter:

At level 5, a Fighter gets Extra Attack, while a Rogue's Sneak Attack goes up to 3d6.

Let's assume a similar weapon loadout - the Fighter is going with a dual-wielding build (which I don't think is quite as strong for pure damage, but it also leaves them fairly capable with a ranged weapon when they need to be if they're going Dex).

    Fighter 5:

We can assume a couple things: the Fighter likely has the Two Weapon Fighting Style, and I think it makes sense that they'd take the dual-wielder feat at level 4. Their Dex is +4 at this point.

Thus, the Fighter is making three attacks that deal 1d6+4 apiece (two with the main hand, one with the off-hand scimitar) and then a fourth, bonus action attack for just 1d6. Total damage output is 4d6+12, or 26 (yes, we're ignoring the higher crit chance thanks to probably using a Vex weapon in the main hand).

    Rogue 5:

The Rogue won't get a Fighting Style. We could potentially imagine that they've taken the Dual Wielder feat as well - Rogues have more consistent need of their bonus action than Fighters, of course, but in situations where they have an ally granting Sneak Attack, they might like to have a way to weaponize it.

If we assume that such an ally is there (maybe that fighter) we can look at their damage in the following way:

Their main hand attack with a shortsword deals 4d6+4, and then they can make a scimitar attack for 1d6. Then, with Dual-Wielder, they make another Scimitar attack for 1d6. So that's basically 6d6+4, or around 25. Actually quite close.

As a note, dual-wielding on a Rogue is, in fact, quite good, because you get that many more chances to land the Sneak Attack. If you had a 60% hit chance, making these three attacks would give you a 93.6% chance to land your Sneak Attack, and that's before factoring in advantage from something like Vex.

    Fighter 11:

By this point, the Fighter has likely capped Dexterity (even if only taking General half-feats, if they start off at 17, they could hit 20 thanks to their first bonus feat at level 6). I also think it's highly likely that they'll have at least a +1 weapon, and by 11, probably two (it will also make my life easier to assume the same bonus on both weapons). Dual Wielding doesn't quite benefit from feat-stacking like a GWM/PAM build does, but they might have gotten something like Mage Slayer and then, like, Medium Armor master or something.

Anyway, the Fighter is now making three attacks with its +1 Shortsword and then a nick attack with its +1 Scimitar, followed by a bonus action attack with said scimitar. That means basically four attacks for 1d6+6 and then a single attack for 1d6+1. So, 5d6+25, which comes out to 42.5 damage.

    Rogue 11:

Likewise, we're looking at +1 weapons in both hands. Actually, Rogues and Fighters will have the same number of feats at this level, as Fighters get their first bonus feat at 6 and Rogues get their only bonus feat at 10. So we can, I think, likewise assume that Rogues are capped on Dexterity at this point.

The Rogue's Sneak Attack is now up to 5d6. So, their main attack is going to be 6d6+6 damage, and then they can make two off-hand attacks for 1d6+1 each, so we're looking at essentially 8d6+8 damage. That winds up looking like 36 damage.

That's more significantly behind the Fighter

    Fighter 20:

11 to 20 is an enormous leap. At this stage, I'd want to have, like Vicious weapons, if not crazy legendary weapons. Vicious Weapons (or Flame Tongues) scale really well with more attacks, and the Fighter would undoubtedly shoot into the stratosphere.

So, to keep this somewhat more fair for the Rogues, we're just going to use +3 weapons.

The Fighter is now making four main-hand attacks on top of their two bonus dual-wielding ones. So, 5 attacks that are doing 1d6+8 each and then a single 1d6+3 attack. That's basically 6d6+43 damage, which comes out to 64 damage.

    Rogue 20:

Sneak Attack by this point is 10d6. So, we have our main attack that deals 11d6+8, and then two off-hand attacks that deal 1d6+3, meaning a total of 13d6+14. That's 59.5 on average.

It's still behind the Fighter, though only by 5.5 points of damage.

    Nuances:

There are a few elements here that I think we need to consider.

First, in the Rogue's favor:

The fact that the Rogue can get their full sneak attack damage as long as any of their attacks hit means that we've actually been devaluing their damage slightly. If we did the full "damage per attack" rather than "damage per hit" calculations, this fact would really help the Rogue's total.

Also, Rogues benefit more from critical hits because so much of their damage comes from rolled dice. If our Fighter lands a crit, they're only adding 3.5 damage on average, while a Rogue who gets lucky with a crit (on their first hit of the turn, to be fair) could, by level 20, be adding 38.5.

But in the Fighter's favor:

Fighters have tools to deal more damage. Action Surge can give them a giant boost in damage, and while this is a limited resource, in a game where combat tends to only last a couple rounds, a higher fraction of a Fighter's turns than you might think can have action surges in them.

Fighters will benefit enormously from anything that increases the damage of an attack. If they can get Conjure Minor Elementals up (easily done with a tier 4 Eldritch Knight) they'll be able to multiply that bonus thanks to all their attacks. The same goes for vicious weapons or even just picking up the Hex or Hunter's Mark spell somehow. Rogues don't get to do this as much because they focus everything into a single attack (though our build here does, at least, give them more of a benefit from it).

And again, there are scenarios in which a Rogue's sneak attack can be shut down - if the Rogue is poisoned, say good bye to sneak attack. Underwater, there are some weapons they can still use, so it might not be so bad. If a Fighter is blinded, they're less likely to hit, but they will still deal full damage. Not so with the Rogue. And even in the regular ebb and flow of combat, a Rogue might just not be able to get the advantage they need. Technically, the 2024 rules have Hide give you the Invisible condition, which means that in theory, a foe with Truesight or Blindsight cannot be hidden from, and even someone merely under the effects of a See Invisibility spell. (Granted, I tried that briefly as a DM and now think I'll rule differently. I actually kind of like the idea of mundane beating magical detection, like how using a Disguise Kit can sometimes work better than the Disguise Self spell because there's no magical illusion to see through, just physical make-up, wigs, and prosthesis).

The point I'm trying to make here is that there's a world in which I think that the Rogue ought to deal more damage than a Fighter. I get that a Fighter, you know, fights, and that's their whole thing. A Rogue does have its powerful skill-monkey features. But I think it would be nice if a Rogue got a little boost to damage.

What Subclasses Should Get a Revision: Rogue

The Rogue is one of the truly central "fantasy character class" archetypes. It's also one that covers a really broad spectrum of classic characters. They can be the morally-ambiguous assassins, spies, or thieves, but they can also be swashbuckling heroes (Robin Hood being the classic "heroic thief").

Mechanically, there are some complaints to be made about the Rogue class as a whole - a class one would expect to really put out among the highest damage often lags behind in at least theoretical builds. WotC instead orients them to being the best class at skill checks, with lots of expertise, tons of skill proficiencies, and the incredibly powerful Reliable Talent.

But setting aside those concerns, subclasses tend to revolve around new ways to either get or enhance the Rogue's primary damage mechanic, the Sneak Attack.

One of the really interesting new mechanics added in the 2024 PHB is Cunning Strikes, which gives us the option to deal less damage with Sneak Attack (spending d6s we'd be rolling as part of it) to get other effects, and some subclasses grant the Rogue new options for this mechanic. While it's a little painful to spend some of our potential damage on these, given that, despite the hefty damage boost Sneak Attack provides, it's still generally not enough to overcome the fact that we can only make one attack per turn (though dual-wielding is a very viable option for Rogues, especially thanks to the Nick weapon mastery).

The point is, a lot of these subclasses could potentially use their own bespoke Cunning Strike options, which would necessitate some kind of revision. I don't think that's enough to justify just reprinting all of them, and in fairness, not all of the PHB subclasses even have a special Cunning Strike.

Another factor in Rogue subclasses is that, of all classes, they have to wait by far the longest to get their second set of subclass features. Artificers get their third set of subclass features at level 9, while Rogues get their second. That means that, especially in a world in which many campaigns only get to around level 10, Rogues really care far more about the subclass features they get right at level 3 than anyone else.

While the Swashbuckler was initially intended as the PHB 2024 fourth subclass, they swapped it instead with the Soulknife, tying a subtle psionic theme into the PHB options. Personally I think this was probably a mistake (even though the Rogue I've played is a Soulknife) because the Swashbuckler is one of the most classic, iconic Rogue archetypes. But we'll talk about it when we get into our list:

We'll be looking at the Mastermind, the Swashbuckler, the Inquisitive, the Scout, and the Phantom.

Mastermind: No

    This is going to be rough, because I honestly think nearly all of these could potentially benefit from a revision. And the concept here - of being a Rogue whose primary weapon is their allies - is pretty cool. But I also think that this is a subclass that feels more like a general personality. I think there's space for a Rogue who truly works more like a Draw Steel Tactician. If so, I think you could call it a Mastermind, but you'd want almost totally different mechanics.

Swashbuckler: Yes

    They've printed this twice already (SCAG and Xanathar's, if memory serves, same as the Mastermind) but I truly think that this subclass is such a profoundly core part of the Rogue fantasy that it really feels like it needs to be given a mechanically-sound update. Could you argue that a Rogue/Fighter multiclass kind of gives you the Swashbuckler anyway? Yeah, probably. But if we're going to have Pirates in the Monster Manual who more or less represent the NPC equivalent of this, surely we can make a good subclass that does it.

Inquisitive: No

    Another heartbreaker. This was clearly meant to replicate the portrayal of Sherlock Holmes from the 2009 Guy Richie/Robert Downey Junior version of Sherlock Holmes (the way that time slows down for him as he fights and he evaluates the precise way in which he can maximize his impact on his opponents). I also love a good detective character, and Rogues, in their truly expansive class fantasy, can very much fit the Noir-ish, hardboiled pulp detective character, which this subclass embodies. Boy, I might be talking myself into a yes. But I guess it might be time to move on from this.

Scout: No

    The Scout is actually not a terrible subclass, but it's basically "hey, do you really not want to play a Ranger?" It's always been the dullest Rogue subclass, and one I never see anyone take (though to be fair, I don't see a ton of Rogues in the first place).

Phantom: No

    Genuinely one of the cooler Rogue subclass ideas, and one that feels supernatural without having explicit spellcasting (something I think we can have more of in D&D). I think this one's just far enough out there that we don't necessarily need to revisit it. It's also probably ok as is.

Overall, I feel kind of bad about being so negative here (I also wonder if I've gotten more negative as this series goes on). A lot of these are genuinely cool ideas, and part of me wants to see them all match their potential. In particular, I have a lot of affection for the Inquisitive, and I do feel like there's some potential in the Mastermind that was never really achieved. The Phantom is the closest among these to being "fine as is," but I'll admit that I wouldn't be too sad to see it get another coat of polish.

Scion of the Three Rogue

 Thanks to the computer games, Baldur's Gate is a very well-known part of the Forgotten Realms, and the Dead Three, the gods Bane, Bhaal, and Myrkul, are also quite notable.

This is the rare subclass where you'll need to work fairly hard if you want to play a good-aligned character and justify it. It's a brutal, violent subclass, literally themed around the blessings of three evil gods.

Level 3:

Bloodthirst:

When an enemy you can see within 30 feet of yourself takes damage and is bloodied after taking that damage but not killed outright, you can use a reaction to teleport to an unoccupied space within 5 feet of them and make one melee attack. You can use this feature Int times per long rest.

    In the right circumstances, this can be a big boost to your damage, but with a few considerations: the target already has to be bloodied for this to work (though it can trigger on the damage that gets them to being bloodied). That means the fight is already going poorly for them. That being said, I've seen some dismiss the power of this feature because a bloodied creature is close to death, but running high-level D&D where most monsters have like 200 HP, there's still plenty of damage needed to be done to take down a bloodied creature. However, if you want Sneak Attack damage on this, it'll probably have to be on someone else's turn. Then again, its being on someone else's turn means that you might be more likely to get Sneak Attack because you might wait for a melee ally to hit them in order to trigger this. The teleport is also nice.

    The limitation on uses, though, is pretty rough: at best we're talking 3 times per day until much higher levels.

Dread Allegiance:

When you get this subclass feature and when you finish a long rest, you can pick between Bane, Bhaal, or Myrkul to give you their blessing. Each gives you a cantrip and a damage resistance. Intelligence is your spellcasting ability.

Bane: Psychic, Minor Illusion

Bhaal: Poison, Blade Ward

Myrkul: Necrotic, Chill Touch

    Each resistance is decent to have. I don't think you're likely to use Chill Touch as a Rogue, but if you're facing a lot of undead (especially incorporeal undead,) it'll be good to have that resistance. Blade Ward is good if you can get it off before combat starts.

Level 9:

Strike Fear:

You gain a new Cunning Strike option: Terrify (cost: 1d6): The target must succeed on a Wisdom saving throw or be frightened for 1 minute. While frightened in this way, you get advantage on attack rolls against them. They can repeat the save at the end of their turns, ending it on a success.

    Ideally you'll have other ways to get advantage, like a Vex weapon, but fear can be a really good debuff to put on foes.

Level 13:

Aura of Malevolence:

When you use Bloodthirst to teleport, each creature of your choice within 10 feet of either the space you left or the one you enter (your choice) takes damage equal to your Int modifier, the damage being the same as the resistance associated with your current Dread Allegiance choice. This damage ignores resistance.

    There's very few creatures with resistance to any of those damage types, if memory serves. It's not a ton of damage, but Rogues don't tend to get AoE damage options.

Level 17:

Dread Incarnate:

You get two benefits:

Cutthroat: You regain a use of Bloodthirst when you finish a short rest.

    Bloodthirst was pretty limited in use, so getting a way to get more of them is certainly nice.

Murderous Intent: When you roll Sneak Attack damage, you treat 1s and 2s as 3s on the d6s.

    This changes the average roll of a d6 from 3.5 to 4. At 17th level, we do 9d6 with Sneak Attack, so that's a boost of 4.5 average damage, and 5 at level 19. It does make the floor on our Sneak Attack 27 damage.

Overall Thoughts:

I like the flavor of this one. I think we could stand to get more uses of Bloodthirst. As always with Rogues, it's tough that they don't get their second set of subclass features until level 9, and for a class that feels like it struggles a bit to keep up with other martial classes in terms of damage, I feel like they played it a little conservatively here.

The storytelling potential of this subclass, though, is enormous.

UA: Phantom Rogue

 Continuing our horror-themed subclass UA review, we're on to the Rogue, a class that likes to stick to the shadows, but fit in many fantasy sub-genres. The Phantom, though, is one directly associated with the haunting specter of death. Published initially in Tasha's Cauldron of Everything, the Phantom gets a little bit of "cleave" damage. Let's see what the proposed changes are:

Level 3:

Wails from the Grave remains the signature features of this subclass, and allows you to essentially hit a secondary target with half of your Sneak Attack (rolling separately, which I think means it won't scale up with a crit) and dealing Necrotic damage, as long as you can see the target and they're within 30 feet. You can do this a number of times per day equal to your Dexterity modifier (minimum 1).

    The old version of this was limited to a number of times equal to your proficiency bonus per long rest, so, this is likely to mean better scaling until mid-tier-3, and then falling behind in tier 4. (I think they also generally have tried to move away from PB-scaling class features because they don't require as much class investment if you're multiclassed).

    I haven't actually seen a Phantom in play (which, to be fair, describes plenty of subclasses,) but I think that this is a nice bit of extra damage for a class that probably needs a slight damage boost, and also has the added benefit of letting you put some damage on a target that either has a very high AC or on which you're struggling to get Sneak Attack. Naturally, though, in single-monster fights, this gives you nothing.

Whispers of the Dead allows you, when you finish a short or long rest, to gain an additional skill or tool proficiency until you use the feature again. This is unchanged form the Tasha's version.

    Rogues of course benefit a lot from skills proficiencies, especially with Reliable Talent coming online earlier than the 2014 version.

Level 9:

Tokens of the Departed has you gain two Soul Trinkets when you finish a long rest, which last until you finish your next one, and they teleport to you if you move more than 30 feet from them. You get a third at level 13 and a fourth at level 17. You also regain an expended soul trinket if you use a reaction when a creature within 30 feet of you dies.

You can use the Soul Trinkets in the following ways:

Death's Knell lets you use Wails of the Grave without expending a use of the feature, destroying the trinket.

Life Essence means that while you have at least one Soul Trinket on you, you have advantage on Death and Constitution saving throws.

Spirit Query lets you take a Magic action to expend the trinket and cast Augury with no spell components, using Dexterity as your spellcasting ability.

    So, the big change here is that you get some free Soul Trinkets every day without having to be near anything dying, which is a real buff and makes this more reliable (especially because Rogues often do want to use their reaction). The other interesting thing is that Spirit Query in the Tasha's version simply let you use up a soul trinket to ask the spirit associated with the trinket a yes-or-no question, whereas now you full-on cast the Augury spell. This is probably better, because if the spirit was of a hostile creature, they might not be all too inclined to answer truthfully.

Voice of Death lets you cast Speak with Dead once per short or long rest with no spell components, which is brand new, and thematically cool. Speak with Dead is, of course, the kind of utility spell that really depends on the adventure, but if we consider this a ribbon, that's fine.

Level 13:

Ghost Walk lets you, once per long rest, or if you destroy a soul trinket, to use a bonus action to get a fly speed of 10 feet (with hover,) a hazy form against which attack rolls have disadvantage, and incorporeal movement, for ten minutes.

    While worded differently, this feature appears to be unchanged, with the sole difference being that it's no action to end it early, rather than a bonus action.

Level 17:

Death's Friend has two sub-features:

Death's Lament allows you to deal your Wails of the Grave damage to the target of your attack as well as the secondary creature.

    Interestingly, I don't know that this works in a single-monster encounter, as I think you still need a second target to trigger Wails of the Grave. I'd probably change that, because this could potentially be a nice damage boost for this level.

Draw of Death gives you one soul trinket when you roll initiative with none remaining.

    I like this sort of "little more gas in the tank" feature to help with long days with lots of fights.

And there we have the Phantom revamp. I believe this is all buffs, which is probably good. I've always liked the concept of the Phantom, though given that it came out with the Soulknife, I was more drawn to that subclass (and am currently playing one).

Rogues are a class that I think a lot of combat-focused theory-crafters tend to think is underpowered, but I do think their value as powerful skill-focused characters shouldn't be undervalued. And while the Wails from the Grave feature is the real headliner here, and one that is a combat enhancement, I think there's some fun utility to this class.

Really the question is how this compares with the old version, and clearly, this is a modest rejiggering rather than a thorough redesign. It's better than the old version, but I'm not extremely excited about it either.

UA: New Subclasses in the Forgotten Realms Player's Guide

 In something of an echo of the Sword Coast Adventurer's Guide, among the first new books that will be coming out for "5.5" Edition is a two-book set that will be a big campaign setting book for the Forgotten Realms. While ten years of published adventures primarily in the Realms along with the aforementioned SCAG, not to mention the popular Baldur's Gate 3 computer game, and the D&D movie, probably means that anyone with any understanding of D&D beyond the surface level is well-aware of the Forgotten Realms or at least elements therein, these books will likely serve as the new definitive setting books. Will we get stuff beyond the Sword Coast, and maybe even beyond the continent of Faerun? No idea.

But, as someone who has actually never run anything in the Forgotten Realms, and who really likes setting-agnostic sourcebooks so that he can use them with his homebrew setting, the number one top appeal here is that, like SCAG, this comes with several subclasses.

And we just received the Unearthed Arcana to test said subclasses.

The subclasses are a mix of returning and brand-new. They are:

College of the Moon for Bards

Knowledge Domain for Clerics

Purple Dragon Knight for Fighters (don't worry - it looks like this is mechanically entirely different than the old one, which made the Champion seem complex and interesting)

Oath of Noble Genies for Paladins

Winter Walker for Rangers

Scion of the Three for Rogues

Spellfire Sorcery for Sorcerers

Bladesinger for Wizards

    You'll note that some of these have very specific FR themes - the "Three" likely refers to the Dead Three, the trio of allied evil gods often worshipped in Baldur's Gate, and even the college of the Moon is specifically linked to the Moonshae Isles. This, thus, might require a bit of lore-bending for other settings, but we'll look at it mechanically.

College of the Moon:

    Moon Bards weave tales of druidic magic into their Bard powers. At the core is your 3rd level feature, which lets you swap between three different "Folktales," each of which give you a different capability. You remain in this "mode" until you swap to another one.

Tale of Life allows you to expend a Bardic Inspiration die when you restore HP to a creature with a spell, adding the result rolled to the heal (no more than one BI per turn). One area of clarification that would be good to get is how this works on AOE heals, such as Mass Healing Word.

Tale of Gloam allows you to Disengage or Hide when you use your usual bonus action Bardic Inspiration feature.

Tale of Mirth allows you to, when an enemy succeeds on a saving throw within 60 feet of you, use a reaction and spend a BI die, forcing the enemy to subtract the roll of that die from their saving throw result, potentially turning it into a failure.

Given how reliant on saving throw spells Bards are, I think this last one is probably going to be your "default" option, and this is the one of these tales that I think could be a huge, clutch play - if some major enemy gets an 18 against your DC 16 spell, this could really turn the tables.

At 6th level, you get Blessing of the Moonwells, which automatically gives you the Moonbeam spell. You can cast the spell as a bonus action for free once per long rest, or by expending a 3rd level spell slot, and when cast this way, you get some extra benefits: while concentrating on it, you glow with dim light for 5 feet, and whenever a creature fails its save against the spell, another creature of your choice within 60 feet regains 2d4 HP.

Moonbeam is an ok spell, and this does a lot to buff it. I don't know that I'd rely on the healing, given that it requires the target to fail a saving throw, but making it a bonus action also means you can do plenty of other things with your turn.

Finally, at 14th level, you get Bolstered Folktales, which allows you to simply roll a d6 instead of expending a Bardic Inspiration die when you use Tale of Life and Tale of Mirth, and then, if you're using Tale of Gloam, you can teleport up to 30 feet as part of the bonus action.

This is a nice feature - basically allowing you to use the feature without ever running out. Again, getting to toss a d6 at a monster's saving throw, especially after you know what they rolled, can be very powerful.

I'll confess that I'm really bad at evaluating Bards. It's a class I can never totally wrap my head around. I feel like this is decent, but probably not blowing anyone's minds in terms of power - though that Tale of Mirth feature feels like it could be the gamechanger.

Knowledge Domain:

    The first PHB subclass to be reprinted outside the PHB following the 2024 culling of Cleric and Wizard subclasses, Knowledge is the Cleric of the mind.

From the start, the first thing one will notice is that they get more domain spells than other subclasses - 3 for each spell level of 1-5, rather than the typical 2. (Remember that Cleric subclasses come at level 3 now, so you'll be getting your 1st and 2nd level domain spells at the same time.)

Domain spells are:

1st: Command, Comprehend Languages, Detect Magic

2nd: Detect Thoughts, Identify, Mind Spike

3rd: Dispel Magic, Nondetection, Tongues

4th: Arcane Eye, Banishment, Confusion

5th: Legend Lore, Scrying, Synaptic Static

    That's a mix of some really solid spells to always have prepared, some very niche ones, and a few combat bangers (I love Synaptic Static).

Beyond this, it appears there's pretty extensive changes, so let's get into them:

At level 3, you get Blessings of Knowledge, giving you proficiency in a type of Artisan's Tools as well as not just proficiency, but expertise in two of the following: Arcana, History, Nature, or Religion. Damn, build yourself a Knowledge Cleric with the Thaumaturge Divine Order and you'll have the highest Arcana/Religion modifier by a large margin.

Also at 3, you get Mind Magic. You can use Channel Divinity to cast any spell on your Domain Spells list without expending a spell slot or requiring material components. That is... actually that's damn good: Scrying for just a Channel Divinity? No need to get the ivory strips and such for Legend Lore? (Also, I think this might turn the casting time into an action as well - I don't know the casting times for all those spells off the top of my head, but still pretty good). The other massive takeaway is that this means a 7th level Knowledge Cleric could potentially cast Banishment several times per day. Not bad at all.

At 6th level, you get Unfettered Mind, gaining telepathy out to 60 feet, and allowing you to simultaneously contact a number of creatures equal to your Wisdom modifier. Also, if you make an Intelligence check and the total is lower than your Wisdom score, you can take the score instead as your total. Granted, with your expertise in two knowledge skills, this might not come up super-often, but it does mean that if you have a -1 to Intelligence, and you beef an Investigation check looking for a trap or something, you could just say "nah, that's a 16." It's not quite Reliable talent, but at the least, if you're trying to figure out some major lore reveal, you have a pretty decent floor on what you can get.

At 17th level, you get Divine Foreknowledge. As a bonus action, you can give yourself advantage on all d20 tests for an hour. You can do this once per long rest, or by expending a 6th or higher level spell slot.

So, this is almost Foresight - a 9th level spell - once per day for free, or as a 6th level spell and as a bonus action. You can only cast it on yourself, and it doesn't impose disadvantage on attackers, but still. That feels... pretty damn good. A 6th level slot is certainly not cheap, but it's a powerful effect that could be clutch.

I've got to say: I don't know if this Cleric is going to be superpowered or anything, and the 6th level feature, while cool, does not really enhance your combat power or anything. But I could imagine playing one of these and really enjoying it, smartypants know-it-all that I am.

Purple Dragon Knight:

    Ok, here we come to what, in the SCAG-version, I consider to be the least-inspired subclass in all of 5th Edition. A subclass that basically just took the three core Fighter features - Second Wind, Action Surge, and Indomitable - and let you spread them to your party, but adding absolutely nothing new to you.

Thankfully, though, this version just utterly trashes the SCAG PDK/"Banneret" in favor of a subclass with an actual identity. And that identity is: Fighter Pet Class!

At 3rd level, we get two features, but let's start with the headliner: Purple Dragon Companion.

The Purple Dragon (likely meant to represent an Amethyst Gem Dragon, as seen in Fizban's) has a scaling stat block, and like a Beast Master's pet or a Battle Smith's steel defender, uses the typical battle pet rules - they act only if you use your bonus action. You can always revive your dragon with the expenditure of a Second Wind use (and 1 minute - so not in-combat) or as part of an hour ritual that can be made as part of a short or long rest.

The Dragon scales with your level and your Intelligence modifier, so you'll really need to work on both your main attack stat and that. Its AC is 13+ your Int and its attack modifier is 3+ your Int (which means I suspect it'll fall behind a bit at high levels - I'll probably suggest in the response to this that it'd be nice to have the dragon scale better by using your Strength or Dexterity instead). The dragon does a Rend attack that deals Force damage, as well as a 2/day Gravity Breath, which doesn't deal damage, but can push or pull targets caught in it.

Also at level 3, harkening to the diplomatic role that PDKs are supposed to play, you get Knightly Envoy, giving you one additional Language and the ability to cast Comprehend Languages as a ritual.

At 7th level, you get Dragon Rider. Your dragon grows to Medium size, and you can ride it as a mount if you're Medium or smaller, but the dragon will fall to the ground at the end of a turn if it's airborne and being held aloft only by its Fly speed. It also only takes 5 feet of movement to mount or dismount. Additionally, the Gravity Breath now deals 2d6 Force damage to creatures that fail the save against it. Finally, when you use Second Wind to regain HP, your dragon regains HP equal to 1d6+ your Fighter level, and regains a use of Gravity Breath.

One thing I'd want to be really specific about is what kind of mount the dragon is. We'd probably want it as an uncontrolled mount so that it can use its own actions. And given that it acts on your turn and not immediately after, this should work fine, as the dragon can move you before you take your action.

At 10th level, you get Rallying Surge. When you use Action Surge, up to three allies in a 30-foot emanation (which could include your dragon, I guess) can use a Reaction to either Advance or Retreat. Advancing allows them to make an attack with a weapon or Unarmed Strike, or if the ally is your dragon, they can make a Rend attack. Retreating allows them to move up to half their speed without provoking Opportunity attacks.

This is... ok, so this is actually a bit like one of the subclass' old features, but what it does is turn your Action Surge into a real battlefield-management ability. I might prefer that this be its own resource, but this ensures that Action Surge will be a really climactic moment.

At 15th level, there's Amethyst Pinnacle. The Dragon grows to Large size and both its Speed and Fly Speed increase to 40 feet, and the dragon can now remain in flight even if you're still mounted on it. Additionally, you get Tandem Attack, which allows you to trade out one of your attacks to command your dragon to attack with a Rend, or spend two of your attacks to have them do their Gravity Breath.

This is fine, but there's an issue: I think the chance that your Dragon's Rend attack is more powerful than one of your own weapon attacks is deeply, deeply unlikely. Unless you are fighting something that resists Bludgeoning/Piercing/Slashing damage (which is more likely in the 2025 Monster Manual, as "magical" versions of those damage types don't exist anymore), even a Shortsword is probably going to deal more damage because it'll be using your preferred stats. I think the dragon could use better damage scaling - its attacks literally don't go up at all unless you can boost your Intelligence, meaning that there's a chance that it never does.

Finally, at level 18, you get Enduring Commander, which gives both you and your dragon resistance to both Force and Psychic Damage. That's pretty cool, actually - Force is a damage type we do see in some endgame campaign bosses, like the Monsters of the Multiverse-versions of the Demon Lords, for example. You'd actually be a more effective tank against such things than a Barbarian. This is not a flashy feature, but it's a solid one.

Overall, of course, this is a massive improvement over the original PDK. I do think it steps on the Drakewarden Ranger's toes, feet, and ankles. I'd prefer a version that's not so reliant on Intelligence, but then, to be fair, Beast Masters need to heavily invest in Wisdom. The main thing is that I think I could actually imagine taking this subclass, even if realistically I'm unlikely to ever play a Fighter that isn't an Eldritch Knight or Battle Master (or maybe Rune Knight).

Oath of Noble Genies

If you want an elemental-themed Paladin, this one is all about the glory and splendor of the most powerful Genies. A paladin oath that respects the power of the elements, here's what you get:

Your Oath spells are the following:

Cantrip: Elementalism

1st: Chromatic Orb, Thunderous Smite

2nd: Mirror Image, Phantasmal Force

3rd: Fly, Gaseous Form

4th: Conjure Minor Elementals, Summon Elemental

5th: Banishing Smite, Contact Other Plane

Whoo boy, you know what I'm seeing there: even if a Paladin won't get the infamous Conjure Minor Elementals until they're 13th level, that is a spell that works quite well for characters who have Extra Attack. Upcasting it to 5th level once in tier 4 will also be quite strong (though it is still better for someone like a Valor Bard, who can upcast it to higher levels). But even if we set aside that arguably busted spell, a Paladin with Fly or Mirror Image (especially the new version of Mirror Image, which gets better if you have high AC) is not bad.

At 3rd level, you get Elemental Smite: When you cast Divine Smite, you can also expend a Channel Divinity use to cause one of four additional effects: 

Dao's Crush grapples the target (DC is equal to your spell save DC.) 

Djinni's Escape allows you to teleport up to 30 feet and take on a mist-like form, which gives you immunity to being Grappled, Prone, or Restrained until the end of your next turn.

Efreeti's Fury adds 2d4 Fire damage on top of your Smite.

Marid's Surge causes the target and each creature of your choice within a 10 foot emanation coming from you to have to make a Strength save or get knocked back 15 feet and fall Prone.

This is a bit resource-heavy, but damn if that isn't some excellent utility. Efreeti's Fury of course you'll want to save for a Crit, but I can see Marid's Surge doing a great job of opening up a hole to retreat, and Djinni's Escape is going to be a great way to get away from some sticky monster.

Also at 3rd level, when you aren't wearing Medium or Heavy Armor, you can add your Charisma modifier to your AC. You also gain proficiency in your choice of Acrobatics, Intimidation, Performance, or Persuasion. So, that's really interesting: if ever you wanted to make a Dex Paladin, this subclass is 100% the one you want to choose. Now, the Paladin I've played had a -2 to Dex, so if I had this subclass, I'd still definitely take heavy armor. But, for example, in my Witchlight campaign, we have a Fairy Paladin, who can't wear heavy armor and fly. Indeed, now that dual-wielding Paladins are quite viable, I could very easily imagine a Dex-based one using Studded Leather and a Shortsword and Scimitar. You could easily start a Paladin character with a +3 to Dex and +2 to Charisma (or even +3, but we'll be conservative,) which would mean a 16 AC in Leather Armor. At high levels, if you maxed out Dex and got Charisma to, say, +4, even in just Studded Leather with no shield, you could have an AC 21. And you would actually stand a good chance at decent stealth checks! I will say that I prefer the aesthetic of heavy armor, but I could very much imagine this being a fun way to build.

At 7th level, you get Aura of Elemental Shielding. At the start of each of your turns, you can choose for creatures affected by your Aura of Protection to have resistance to your choice of Acid, Cold, Fire, Lightning, or Thunder damage (I've rephrased this: the aura remains in effect at all times the Aura is, you can just swap the damage type each turn with no action required).

This is, just, you know, good. And fighting Fire Elementals? Now you and nearby friends are resistant. It's responsive and applies to the most common magical damage types.

At 15th level, you get Elemental Rebuke. When you are hit with an attack roll, you can take a Reaction to halve the damage (rounded down) and then you can force the attacker (there does not appear to be a restriction on being able to see them or any distance limit) to make a Dexterity saving throw (against your spell save DC). On a failure, the attacker takes 4d10+Cha of your choice of Acid, Cold, Fire, Lightning, or Thunder damage. They take half damage on a success. You can do this a number of times equal to your Charisma modifier per long rest.

The retaliatory damage is fine and good, but the enormous thing here is that you have an amazing defensive feature to use against a massive hit - use this when you get crit, for example. And 4d10+Cha (likely 25 or so) is not a small amount of damage on a reaction.

Finally, your level 20 Ult is Noble Scion. As a bonus action, for ten minutes, you get a 60 foot flying speed (with hover) and when you or an ally in your Aura of Protection fails a d20 test, you can use a reaction to cause the test to succeed instead. Like other Ults, you can use this again by expending a 5th level spell slot.

That is damned good. At level 20, a failed saving throw can be massive, so being able to just, as a reaction, say "no, it succeeded," is hugely powerful. Sure, you might use this on an attack to make sure that you're damaging the final boss as much as possible, but I think the "no, I don't think I will get swallowed by the Tarrasque" reaction, for example, is really, really big.

I've got to say, of the four subclasses we've been through so far, this is the one that has me the most excited. This feels good. Possibly too good, but I'm ok with that.

Winter Walker

Gotta say, just from the name I'm already pretty excited about this one. While I've happily moved to Southern California where it's nice and warm basically all the time, in a fantasy tale, the frigid cold is a good setting for grim fantasy tales (naturally, the name evokes the White Walkers of A Song of Ice and Fire). So, let's look at what it is, mechanically:

At 3rd level, you get Frigid Explorer. This gives you the following: You have resistance to Cold damage (which, remember, also means that you don't get Exhaustion from being exposed to extreme cold) and also, when you hit a creature with a weapon, you can deal an extra 1d4 Cold damage. The target can take this damage only once per turn. The damage ignores resistance to Cold damage and goes up to 1d6 at level 11.

Couple nuances: first of all, this is just free damage. It's not a ton, but it's there, and should stack with things like Hunter's Mark or other boosts. Second, it's the creature that can only take this extra damage once per turn, meaning that if you are, say, a Dual Wielder, you could potentially hit four separate monsters and deal this damage to all of them (if you've got main hand 1, main hand 2, off hand Nick, and then off hand Bonus Action). Not enormous, but a pretty solid boost for a 3rd level feature. Ignoring Cold resistance also really helps it fit into the themes of the subclass without paradoxically hamstringing you when fighting in your preferred envrionment.

Also at 3, you get Hunter's Rime. When you cast Hunter's Mark, you gain Temp HP equal to 1d10+your Ranger level. Also, while a creature is marked by Hunter's Mark, it cannot take the Disengage action.

While I'm not always happy about features that only work when the Ranger is using Hunter's Mark, this is nice if you are using it. (I really wish the spell were just a class feature that didn't interact with the rest of your spellcasting abilities).

Finally, you have your subclass spells. These are:

Ice Knife, Pass without Trace, Remove Curse, Ice Storm, and Cone of Cold.

Nothing terrible there. Pass without Trace is a no-brainer for any Ranger, but I guess this frees up a spell option. All seems on-theme.

At 7, you get Fortifying Soul. When you finish a Short Rest, you can choose a number of creatures up to your Wisdom modifier (minimum 1) and they regain HP equal to 1d10 + your Ranger level and have Advantage against being Frightened for 1 hour. You can use this once per Long rest.

This feels... it feels very niche. The HP boost is cool, but the 1 hour of advantage against Fear effects feels like it's such a roll of the dice. If you know you're about to face something that will frighten party members, are you really able to take a rest before such a fight? And so soon before? And why can't we give this bonus at the end of a long rest?

At 11, we get Chilling Retribution. When a creature hits you with an attack roll, you can take a Reaction to make them make a Wisdom save. If they fail, the creature is Frightened until the end of your next turn and while Frightened, their speed is 0. You can use this a number of times equal to your Wisdom modifier.

This is cool, but I'd prefer a slight change: let you use this in response to an attack, rather than a hit. That way the reaction can potentially prevent the hit coming for you.

At 15, you get Frozen Haunt. When you cast Hunter's Mark, you can adopt a ghostly, snowy form. The form lasts until the spell ends. While it's up, you have Immunity to Cold damage, and when you adopt the form and at the start of each subsequent turn, each creature of your choice within a 15 foot emanation takes 2d4 Cold damage. You also have immunity to being Grappled, Prone, and Restrained, and can pass through creatures and objects as if they were difficult terrain, taking 1d10 Force damage if you stop inside a creature or object, and getting shunted to the nearest unoccupied space if the feature ends while inside a creature or object.

You can do this once per long rest, or you can spend a 4th level spell slot to do it again. (As far as I can tell, this is in addition to your casting of Hunter's Mark).

2d4 isn't a ton of damage at 15th level, unfortunately, even if it's guaranteed damage (and AoE damage... maybe it is fair?)

I'm mixed on this subclass: flavor-wise, I adore it. It's dripping with vibes, and those vibes are extremely grim. It would be the perfect subclass for playing in Icewind Dale: Rime of the Frostmaiden. Like so many Ranger things, including the class in general, I just worry that it's a little undertuned.

Scion of the Three:

Do you want to be the party's token evil teammate? Ok, sure, you don't have to be a bad guy, but this is the subclass related to the Dead Three. Already a pretty strong aesthetic is presenting itself, but let's look at its mechanics.

At 3rd level, you get Bloodthirst. If your Sneak Attack hits a Bloodied creature, the creature takes extra damage equal to half your Rogue level (rounded up). When an enemy you can see is reduced to 0 hit points, you can use a reaction to teleport to an unoccupied space within 30 feet of yourself, and then you can make one melee attack. You can do this a number of times equal to your Intelligence modifier (minimum 1) per long rest.

Ok, yes. Not only does this boost your damage by a bit, but so many team compositions are about finding ways to get the Rogue multiple Sneak Attacks per round. This has it built in.

Also at 3, you get Dread Allegiance. You pick between Bane, Bhaal, and Myrkul. Each god grants a damage resistance and a cantrip (Intelligence to cast). You can swap your allegiance after a long rest.

Bane grants Psychic resistance and Minor Illusion

Bhaal grants Poison resistance and Blade Ward

Myrkul grants Necrotic resistance and Chill Touch.

Of these, I think Minor Illusion will be the easiest to use in most scenarios, though you might find the Poison resistance more broadly helpful.

At 9th level, you get Strike Fear. You gain an additional Cunning Strike, which is "Terrify." This costs 1d6 of your Sneak Attack and forces the creature to make a Wisdom saving throw, becoming Frightened for 1 minute on a failure (they can repeat the save at the end of each of their turns).

This makes perfect sense for the vibes here. And Frightened is a potentially really powerful effect.

At 13th level, you gain Aura of Malevolence. At the start of each of your turns, each creature of your choice within a 10 foot Emanation takes damage equal to your Intelligence modifier (minimim 1), the damage being the same kind you have resistance to via your Dread Allegiance feature. The damage ignores resistance (though the aura is inactive if you're incapacitated).

This definitely encourages you to fight in melee, which can be a risky proposition for a Rogue, but it's free, guaranteed, AoE damage that is always on, so I can't complain much.

Finally, at 17, you get Dread Incarnate: You have Advantage on attack rolls against creatures with the Frightened condition, and when you roll Sneak Attack damage, you can treat a roll of 1 or 2 on the die as a 3.

Ok, the first part is going to be very nice if your party can toss fear effects around, but you also have your Terrify Cunning Strike, so you can set this up for yourself. Let's figure out how the damage boost treats you:

Normally, a d6's average damage is 3.5. But with this change, the average damage becomes 4. So, at level 17, our Sneak Attack does 9d6, meaning we're going from 31.5 to 36, or a boost of 5.5 damage. (And 6 at level 19). Much like the new Great Weapon Fighting style, this isn't enormous, but we're also rolling way more d6s than we would with a Greatsword, so the boost is multiplied. I don't think this is a massive game-changer, but it's nice.

Like the Winter Walker, I love the vibe of this subclass - truly wicked, you can imagine this Rogue is just utterly splattered with blood in the midst of combat. I think the best features are probably the first ones, especially Bloodthirst. It might need a tune-up, but I think it's probably in pretty good shape.

Spellfire Sorcery:

The Spellfire Sorcerer gets its power directly from the Weave (the FR explanation for how magic works is that there's essentially an underlying energy field to the universe call the Weave). Thematically, you have strange magic that can act in myriad ways - both harmful and helpful.

First off, you get a number of subclass spells. These are:

1st: Cure Wounds, Guiding Bolt

2nd: Lesser Restoration, Scorching Ray

3rd: Aura of Vitality, Dispel Magic

4th: Fire Shield, Wall of Fire

5th: Greater Restoration, Flame Strike

Naturally, the notable ones here are that you get various healing and restoration spells, usually not the purview of a Sorcerer. Otherwise, there's a "radiant/fire" theme, not unlike a Celestial Warlock.

At 3rd level, you get Spellfire Burst. When you expend at least 1 Sorcery Point as a Magic action or bonus action on your turn, you can unleash one of the following effects (max once per turn). Bolstering Flames grants you or a creature you see within 30 feet Temp HP equal to 1d4 + Cha. Radiant Fire forces a creature you can see within 30 feet to make a Dex save or take 1d6 Fire or Radiant damage (your choice).

So, this is nice mainly in that it doesn't expend any more resources than you're already doing: just using Sorcery Points, which Sorcerers do most of the time anyway, will get these little bonus effects. Neither are huge, but they can help.

At 6th level, you get Absorb Spells. In addition to your subclass spells, you also get Counterspell, and if a target fails its saving throw against Counterspell, you regain 1d4 Sorcery Points. A fun little efficiency boost, though it's very, very dependent on how many spellcasters you're fighting.

At 14th level, you get Honed Spellfire, which lets you add your Sorcerer level to the Temp HP from Bolstering Flame (a massive boost) and Radiant Fire is now 3d6, up from 1d6, more or less keeping it relevant at this higher level.

Finally, at 18, you get Crown of Spellfire. As a bonus action, you gain a buff for 1 minute. Once per turn while under this effect, if you are hit by an attack roll, you can spend a number of Hit Dice up to your Charisma modifier, and reduce the damage you take by an mount equal to the roll plus your Sorcerer level. (So, if you spend the max of, say, 5d6 + 18, that's 35.5 on average). Additionally, you gain a Fly speed of 60 feet and can hover. Finally, you gain Spell Avoidance - if you're subjected to a spell or magical effect (notably a Dragon's Breath does not count, as it's not magical) that allows you to save for half damage, you instead take no damage on a success and half on a failure, as long as you're not incapaciated.

You can do this once per long rest, or you can expend 7 Sorcery Points to do it again.

This is actually a pretty cool Ult move, like a Paladin's. The spend-hit-dice feature is probably not the most efficient use of hit dice, but if you're in a really tight spot, this could be a way to avoid or mitigate some truly deadly blast of damage.

I think overall that the Spellfire Sorcerer is probably all right, and definitely letting Sorcerers use their tricksy metamagic on healing spells has some potential for fun. Thematically it's a little dull, though.

Bladesinger:

It feels like this was only just reprinted, but then I guess Tasha's was actually a while ago at this point.

Now, we've seen a lot of versions of this, but I'm going to go through each feature individually to make sure we're catching it all:

At 3rd level, we get Bladesong. As a bonus action, we enter our Bladesong, as long as we aren't wearign armor or using a shield (interesting: no armor whatsoever - this used to allow light armor). The Bladesong lasts 1 minute and ends if you're incapacitated or don armor or a shield, or if you use two hands to make an attack with a weapon. Or you can dismiss it, no action required. You can invoke the Bladesong Int times per long rest.

While active, the Bladesong does the following:

You gain a bonus to your AC equal to your Int modifier and your speed increases by 10 feet.

When you attack with a weapon with which you have proficiency, you can add Intelligence rather than Strength or Dexterity to the attack and damage rolls.

You can add your Intelligence modifier to your Concentration saving throws.

Let's back up: the enormous change here is that you can now use Intelligence to make your weapon attacks. I believe the earlier version allowed you to add Int to your damage rolls, but this will allow you to fully focus on your spellcasting ability and keep full pace on your weapon ability. Essentially, you get the benefits that Bladelock or an Armorer/Battle Smith does.

Also at 3rd level, you get Training in War and Song. You gain proficiency with all Melee Martial Weapons that don't have the Two-Handed or Heavy properties. You can use a Melee weapon you have proficiency with as a spellcasting focus for Wizard spells. You also gain your choice of Acrobatics, Athletics, Performance, or Persuasion.

So! While you won't be able to use a Greatsword or a Halberd, you can easily use a Longsword or a War Pick. That said, you won't be able to get a d10 die, as you still can't go Versatile. Thus, a Rapier is probably one of your best picks still. Or is it? You could also dual-wield d6 weapons, as nothing says you have to keep one hand free (and given that your weapons can be spell foci, you can still cast just fine with them).

With a reasonable +2 to Dex, and then with Mage Armor, you shouldn't have too much trouble pushing your AC up to Heavy Armor levels (with +5 to Int, that's now at 20, and 25 with the Shield spell).

At 6th level, you get Extra Attack. Much as in Tasha's, as well as what the Eldritch Knight and Valor Bard have inherited, the Bladesinger can replace one attack with a Cantrip. I think 99% of Bladesingers should pick up the new True Strike, which will work perfectly for them given that they can use Intelligence for attacks anyway.

At 10th level, you get Song of Defense. As a reaction, when you take damage, you can expend a spell slot and reduce the damage by an amount equal to 5 times the spell slot's level. So, if Shield doesn't do it, you can pick the spell slot to mitigate the ideal amount of damage. I think this is the same as from Tasha's.

At 14th level, you gain Song of Victory. If you cast a spell that has a casting time of one action, you can make an attack with a weapon as a bonus action. Note that this interacts nicely with your Extra Attack - if you cast True Strike and then make a normal attack as your Attack Action, True Strike triggers this and gives you a third attack.

The Bladesinger is a popular and powerful subclass (when I rolled really good stats for my Scribes Wizard I actually considered going this way because I could). I think that this version cuts down on some of the doubling-down that can let their numbers go crazy, but in exchange, the whole thing gets more user friendly. I think my only possible change would be to let a Wizard expend some level of spell slot to get additional uses of Bladesong. Even outside of Bladesong, the subclass has a lot going on, though.

    And there we have it! A whole bunch of subclasses reviewed.

I think my favorite is probably the Oath of the Noble Genie. In terms of flavor, I like the Winter Walker and Scion of the Three best. Most improved has to go to the Purple Dragon Knight (though it's only one of three revisions).

I'll be curious to see what others think of these subclasses, but I think it's a decent crop.