D&D 5E Yet Another Arcane Trickster / Bladesinger Thread: The Concentration Problem

I've got an Eladrin (DMG variant) Arcane Trickster 6/Bladesinger 5 who uses Shadow Blade + Booming Blade for offense (mixing in the occasional Tasha's Hideous Laughter or Web in place of Shadow Blade), Mirror Image + Bladesong + Shield/Absorb Elements/Uncanny Dodge for defense, and is an exploration expert and utility caster out of combat.

My dilemma

In a nutshell, my dilemma is this: How to I build from here to be able to take full advantage of magical ambush without undercutting my strikerness too much (since at the moment I'm dependent on shadow blade for damage)?

Current Stats

Eladrin (DMG variant) Arcane Trickster 6/Bladesinger 5

STR 8
DEX 18 (Elven Accuracy)
CON 14 (Resilient CON)
INT 19 (Headband of Intellect; naturally 16)
WIS 12
CHA 8

Expertise: Perception, Stealth, Investigation and Arcana

Party Composition

The front line is me, a glaive-wielding totem barbarian and a mace and shield devotion paladin. There's a tempest cleric in the middle, and in back your standard agonizing blast sorlock and your standard debuff/control lore bard.

The sorlock frequently hastes me and the barbarian, which is a beautiful thing, since it synergizes wonderfully with shadow blade. (I don't usually double up on sneak attacks by readying an action, since I like to keep my reaction free for defensive uses; and booming blade replaces much of the damage from a second sneak attack anyway, so I'll generally booming blade on my turn with my main action and just make another attack with shadow blade using my haste action)

My Tentative Plan

What I'm thinking about to get around the shadow blade dependence problem is to start investing in ranged combat, since situations where magical ambush is applicable are by definition situations where hiding is available, making staying back and sniping more viable. I also suck at ranges greater than 20' (throwing shadow blade) right now, since I have less sneak attack than a pure rogue, and get neither booming blade nor shadow blade damage past that. Most likely my endpoint is AT 12 / BS 8, so I've got four ASIs/feats to come (We may not make it all the way to 20, but at the moment we are intending to do so)

The standard ranged rogue path is to take crossbow expert first, then sharpshooter after maxing DEX, but... I'm actually considering taking sharpshooter first (probably level 13) then crossbow expert (at 16 or 17), and some pair of +2 DEX, Skulker, Lucky, Tough, War Caster at 18 and 20. I might not be thinking about this properly though, so would love the community's input. Here's my thought process.

Milestone 1: AT 9 / BS 5 (Sharpshooter and Magical Ambush)

I think I'm going to advance rogue all the way to 9 from here before picking up extra attack with Bladesinger 6. So at Rogue 8 I'd like to do something to improve my non-concentration-dependent offensive capability. That something should be useful for two levels with a single attack and often a second via Haste from the sorcerer.

At this stage I'm not after the power attack from sharpshooter, which is at best a wash vs a typical AC for the level in most circumstances (though with bardic inspiration it might be worth it). The initial draw is removing disadvantage at long range: even when trying to be in melee it will allow me to throw my shadow blade 60 ft (vs a dragon, say), or if I want to be concentrating on something else, I can still use bladesong and fire the hand crossbow a decent distance (at much reduced DPR compared to shadow blade, but it shouldn't be too hard for the value of the other spell to make up the difference).

Taking crossbow expert instead at this stage would yield somewhat better ranged DPR in most circumstances, but is only even with a rapier and booming blade with extra attack (and unlike the other options, eats my bonus action). And unlike sharpshooter, it doesn't really open up new situations (i.e., at longer range) to decent attacks. So it'd really only be useful when I'm too far for melee but still within 30 feet (overland distance won't generally be an issue after the first round between bladesong and cunning action dash; I'm more worried about the z axis).

In any case, any time dim light is available and range is not an issue, shadow blade will still be the go-to option, since even at a 3rd level slot the damage it provides is enormous (something like +30 average DPR after factoring in to-hit and to-crit chances).

Milestone 2: AT 9 / BS 8 (Extra Attack, 4th level spells, Crossbow Expert)

At levels 16 and 17 I'll get 4th level spells and another feat. If I've got a target restrained or I'm invisible I will have (super)advantage on all attacks. Opening up two more power attacks with the hand crossbow, frequently at trivantage, is, of course, massive, and pulls ahead of a 3rd level shadow blade that gets advantage on only the first attack. Shadow blade still reigns supreme in dim light, and ekes slightly ahead even without full turn advantage if using a 5th level slot, but that doesn't factor in the benefits of those other spells to defense or to allies', so (assuming a magic hand crossbow), it likely means largely retiring the shadow blade outside scenarios when those other spells would not be effective, using a hand crossbow for both melee and range as needed.

End Game: AT 12 / BS 8 (probably)

ASI/Feat considerations in this last tier include +2 DEX, Skulker, Lucky, War Caster, Tough. There's a little bit of anti-synergy with Skulker earlier in the build, since the hiding benefit is most useful in dim light, where I'm least likely to be staying at range. But that's less true in the end game since SS/XBE will make hand crossbow attacks competitive with shadow blade even in dim conditions.

Alternative Considerations

I could just suck up the fact that I'm no good at range any more, just max DEX at 13, maybe take Warcaster and Sentinel at 17 and 18 (nice albeit a bit cheesy synergy with the fear spell), and do the canonical melee rogue thing. But as I said, I'm hesitant to devote reactions to attacking since I have so many nice defensive uses for it, and it means I have a hole in my offense if I can't close for melee.<

So... what do y'all think?

Would you go this way, or is it too much feat investment for an M.O. I'll only be using sometimes? What would you tweak or reorder?

Nice write-up. I completely understand the dilemma you're in. To reiterate your problem, you are level 11 and your melee damage is dropping off because you cant keep shadowblade up due to concentration checks. My AT/BS build ends up Rogue 10/Wizard 10, with the first ASI going towards Dex and the second picking up War Caster to help bolster melee effectiveness. My level progression gets to Rogue 4/Rogue 4, and since I dont end up getting shadowblade until 5-6th level, I postponed War Caster to level 8.

When you think about going ranged with this build, it's either something you start towards from the beginning or you are going to spread yourself too thin, IMHO. Shadow Blade is amazing and is the focal point of a bulk of the build. You can bolster your defenses (and hence concentration) by keeping Bladesinging and Mirror Image up as much as possible. Without the use of Shield, you should have around 20-22 AC from levels 3-10. You will still take damage, which is why having War Caster is so helpful, more so than the Con saving throw from Resilient, IMO. War Caster also gives us the ability to dual wield and still cast somatic spells without breaking the rules. It also makes booming blade opportunity attacks possible, in lieu of your defensive reaction, but definitely adding more offense as you continue to add to your non-concentration defensive spells.

So what should you do now? Get blink if you haven't already, and pick up War Caster for advantages on Concentration checks. Also don't forget that when you gain a level in either Wizard or Rogue, you gain spells (or can trade them out) of any spell level for which you have slots. Meaning 4th level at this point.

The way I envision the AT/BS Shadowblade/Disabler playing out at higher levels is mainly as a melee rogue with Bladesinging and War Caster giving you +3-5 on Con saves, with advantage. Ideally you wont get hit at all, but AOE damage will happen and require a lot of DC 10 concentration checks. Mirror Image and Blink are some of the most important buffs you can have up otherwise. Mirror Image is great against the more numerous opponents - drawing out attacks and essentially nullifying your opponents offense, while Blink excels in fights vs hard hitting opponents - by keeping you in the fight and able to use dispel magic, counterspell, blind, slow, or even use extremely powerful defensive spells like Otiluke's Resilient Sphere, Wall of Force, or Polymorph when in a bind.

The ranged drawback does suck, which is why its good to just bite the bullet and pick up a decent ranged cantrip. Cantrips scale based on your overall level, so at 11th level, your Firebolt could be 3d10. Toll the Dead is pretty good too, although shorter range. I personally am a Chill Touch kind of guy. When you need a ranged spell with 120', 1d8 increments isn't bad. And when you need to stop someone from healing, there's nothing quite as good as Chill Touch.

Thanks for the feedback, [MENTION=6992717]Irregular_Gnoll[/MENTION]!

I think you misunderstood the core of my dilemma, though. The issue isn't so much maintaining concentration on shadow blade as not being able to devote concentration to magical ambush spells and still do decent damage.

I've got Resilient (CON) already and so when bladesong is up I've got +10 to concentration checks, meaning I'm immune to losing concentration on anything below 22 damage on a hit. With studded leather, high DEX and INT and the shield spell (and often haste on top of that), attack rolls need to be 25 to hit me (27 if hasted). On top of that, I pretty much automatically cast mirror image if I have an action before the fight starts to do so. And then, if all that fails and I do get hit, I can often use uncanny dodge to get the damage down into a range where I can maintain concentration easily.

We recently had a long vampire fight where the vamp charmed two of our party and seemed to attack me every chance she got. That was a case where I actually used a combat turn to cast mirror image. And good thing too as one of my images took a crit for me. She did manage to hit me once with a grapple attack, but misty step solved that problem. But even as our paladin almost went down, I managed not to take a single point of damage.

I'm considering war caster, but dual wielding isn't that helpful for me since I so frequently have superadvantage on my first attack from some source or other (typically owl help). I do have a +1 dagger that I can put in my offhand (or use in place of a rapier when not using shadow blade), but since I don't have the TWF fighting style that would only add three or four damage a round, even taking into account another chance to land sneak attack. Just boosting DEX would be worth more to damage than dual wielding. The chance to add booming blade damage to an AoO is nice, but those don't come up much, since I'm usually in melee with at least the barbarian, so creatures don't tend to provoke AoOs from me unless the bard lands a dissonant whispers. And if I do happen to be in solo melee, I only get sneak attack on an AoO if I have some source of advantage (e.g., dim light or faerie fire from the bard). Absent that I usually won't take the attack in the interest of conserving my reaction.

I didn't take Blink because I think mainly that would just serve to open up my allies to getting attacked more, and since I'm so much harder to hit than anyone else, that's not usually desirable.

A ranged cantrip is a possibility (and definitely would have liked for someone to have had chill touch vs the vamp, though preferably someone else as I was doing a lot of damage with shadow blade there). 3d8 is only 13.5 damage though, vs 1d6+4d6+4 = 21.5 from a single hand crossbow hit. And haste doesn't allow two cantrips in a turn, whereas using a crossbow would be another 1d6+4, for a total of 29. Note also that the range benefit of sharpshooter applies to a thrown shadow blade as well, which is worth even more damage. Unlocking that damage at longer range (and eventually comboing it with crossbow expert for an additional two attacks a turn) seems worth more to me than what war caster buys, though I guess that comes down to how often I make AoOs vs how often I need to attack from more than 20-30'. It feels like the bigger opportunity cost of taking sharpshooter is the DEX boost (or even something like Lucky).

Note that I can't get 4th level spells until I have a 7th level in wizard. Spell picks are based on your class level, not your slots available. I tend to think that some kind of exception should be made when you're multiclassed into a class that uses the same list, but there's nothing like that in the rules.

Nice write up.

Would you consider losing that last ASI and throwing in one (maybe 2) fighter levels?

Archery fighting style helps so much if you plan on using that hand crossbow.

Minsc said:

Nice write up.

Would you consider losing that last ASI and throwing in one (maybe 2) fighter levels?

Archery fighting style helps so much if you plan on using that hand crossbow.

Click to expand...

Thanks! Yeah, it's something I've thought about, for sure. If it would come at the cost of a +2 DEX though it's only a net +1 to hit, at the cost of a bit of damage per hit, +1 AC, etc.

Let's assume 18 DEX, sharpshooter + crossbow expert at Rogue 11 (so 6d6 sneak attack), vs an enemy with AC 20, and compare archery fighting style to +2 DEX. We'll do three attacks and assume no plus modifier on the weapon.

* Baseline with no advantage: it's slightly better not to use power attacks since the chance that they would all miss is too high. 33.7 expected DPR.
* Baseline with advantage on the first shot: probably want to power attack only on the first shot; the mean damage is about the same either way, but I'd rather have the lower variance resulting from regular attacks. 43.7 expected DPR.
* Baseline with advantage on all attacks: power attacks all the way, for 60.5 expected DPR.

Next, assume we take Rogue 12 for +2 to DEX.
* No advantage: No power attacks, for 37.1 expected DPR. The DEX was worth 3.4 DPR.
* Advantage on 1st: Now you get a couple of points of DPR by using power attacks every time. Not clear it's worth the variance, but 49.3 expected DPR. The DEX was worth 5.6 DPR.
* Advantage on all: Power attacks all the way, for 66.4 eDPR. The DEX was worth 5.9.

Instead, suppose we take Fighter 1 for archery style:
* Without advantage: we get up to 40.0 eDPR, an additional 2.9 over the ASI.
* Advantage on 1st: 51.3 eDPR, 1.8 more than the ASI.
* Advantage on All: 67.2 eDPR, 0.8 more than the ASI.

So archery style by itself has pretty meager returns if we forego an ASI for it; not worth giving up the other benefits of dexterity, IMO. Now of course there was another feat in there. If that feat was Skulker, the benefits of archery go down further, since we will more often push advantage from the first attack to the second.

However, we could take +2 DEX at 18 and then a fighter level, and not take Skulker. In that case, archery nets about 7 extra eDPR across all advantage statuses. Now we want to compare that to the value of Skulker.

It's hard to measure the DPR value of Skulker, since its benefit is situational: if I can't get owl help, there would otherwise be no place to hide, but there is a source of light obscurement, turning that first shot from no advantage to advantage is worth just as much all by itself as archery style (about 7 extra eDPR). But that's probably not very common.

If there would have been hiding available anyway, then Skulker gives advantage on the second shot if the first one misses. We can math that:

* With advantage on the first shot only, there's a 27% chance the first shot misses, and a 12% chance of hitting zero times, meaning the expected damage from sneak attack is 22.
* Going from no advantage to advantage is worth about 7 eDPR. That means skulker increases the expected damage on the second shot by 0.27 * 7 = 1.9.
* The chance that the first two shots both miss is 8%. So skulker adds 0.08 * 7 = 0.6 expected damage to the third shot.
* The chance that we hit zero times goes down to 5% with skulker, so the expected damage from sneak attack goes up by 1.9.
* So all together, if Skulker enables us to hide when we couldn't otherwise, it's worth about 11.5 damage. If we could have hidden anyway we've gained about 4.5 expected damage from Skulker.
* All of this assumes we always qualify for sneak attack on every shot. In situations where the target doesn't have an adjacent enemy, skulker helps qualify for sneak attack not just land it; archery style doesn't do that.

So I think the bottom line is that getting archery style may be a bit better than skulker from a DPR perspective alone if we don't expect it to enable hiding very often and if we're not worried about not qualifying for sneak attack. But skulker also can enable magical ambush sometimes, which is worth a lot on its own. All together, this leads me to prefer picking up skulker to taking that single fighter level.

Of course, a second level for action surge is a whole other ball of wax.

By the way, all the above eDPR calculations factored in crits, in case you're puzzled about where some of the numbers are coming from.

I just have to say that I love this build idea.

I had a similar flavor idea (we’re not hitting level20 though) for an upcoming campaign. All my theorycrafting went to a maximum of 3 levels in either BS or AT to create my “spell-thief” but now I think I might actually be better off mixing that up a bit.

Thank you.

(Sorry I don’t have much to add other than bumping your thread LOL)

Apologies for the necro, but I'm quite curious about how you ended going with this character's build, since I'm pursuing the exact same concept with my next character.
That said, I'm definitely going to take Mobile on first ASI and Resilient (CON) or Warcaster for my second, then maybe Elven Accuracy for my 3rd.
Mobile seems really important to me for freeing up your bonus actions.
I'm also considering just a two or three level dip into Blade Singer so I wouldn't be as reliant on Shadow Blade as you said, but this does mean less spells and delaying ASI two levels.

We've only leveled once since this post, so I haven't gotten to that ASI yet, but I'll keep you posted.

I could certainly see just a two-level dip making sense, though right, it only ends up advancing your spell slots by one caster level due to the weird round down rule kicking in once you have spell casting from two classes. And delaying those ASIs hurts for sure. Plus it means you don't get Shadow Blade until level 9.

I also really like Mobile on this build, since between Bladesong and Shadow Blade, you usually don't have a bonus action free to disengage until the third round. But it's also worth noting that between bladesong, shield, and uncanny dodge, you're likely to be one of the tankiest party members (low HP aside), so you may not necessarily want to be dipping away if it means one of your allies gets whacked instead of you. Really depends on the rest of the party; in a party full of ranged characters or where all the melee characters have some kind of AoO avoidance mechanism, Mobile paired with booming blade becomes amazing, since if you can leave the enemy with nobody in melee range, it makes triggering the extra damage from BB that much more likely.

Alright, I'll stay posted then
I am curious if there's anything you would have done differently, and how you would build your level progression now.