D&D 5E How do I optimize this character?

krunchyfrogg · May 23, 2019

I’m playing an elf arcane trickster (monoclass so far) who is level 3 with the following stats (standard array): 8/17/14/14/12/10.

Now, I can rearrange these if I wanted, but I think this is pretty optimal.

The character is played as an archer who uses an owl familiar to pretty much always gain advantage on attacks with the familiar granting the help action and my elf sniping with his longbow.

Next level I get my first ASI.

I’m tempted to grab sharpshooter, but I’m afraid of the -5 to hit (even though I’m often gaining advantage, which should nullify that). Also, I’m not sure if that -5 is worth the risk of losing the 2d6 sneak attack dice with the possible payoff of +10 damage.

I think a better option is just pumping my DEX (+1 hit/damage/initiative/most of my skills). Now, with a DEX of 17, As well as all even stats, I’m better off getting a feat that adds +1 to DEX and later bumping my DEX from 18 to 20.

The question is, which feat is worth it?

I could go elven accuracy, which would later really help with that sharpshooter feat that I could grab down the road. Is there a better choice?

Edit: if there isn’t really a decent +1 DEX feat, I could start with a 16 DEX and 15 CON, and I’ll just add resilient CON at some point. This won’t happen for a while, but it is good.

Esker · May 23, 2019

Given your current stats I think elven accuracy is the clear winner given how often you are getting advantage, unless you want to plan for Resilient CON. Even with constant advantage, unless shooting into cover, using the sharpshooter power attack option is only a net average DPR gain vs elven accuracy against ACs of 12 and below (coming out even at AC 13 and behind above that).

Once you have elven accuracy, at level 8, sharpshooter gives you a bit of a DPR edge compared to boosting DEX again up through AC 16, but important enemies will often be above that at that level, and you'd be giving up the AC, initiative, and skill boost.

The other ranged feat to consider is crossbow expert, obviously. Since you're getting advantage from your owl rather than from hiding, you have your bonus action available, and can still often get advantage on your first shot. So this is better for you than many other rogues. And DPR-wise it does beat elven accuracy at level 4 up through AC 16, tying at AC 17. It also has the advantage of allowing you to redirect some of that damage to a new target if your first shot is a kill, but the disadvantage of a short range. And the DPR gain is less than 2 points on average, above AC 13. So you'd probably need to value that two target option a fair bit to make it worth giving up the other benefits of higher DEX.

Esker · May 23, 2019

(Happy to provide calculations behind the above claims if you want them --- all of this is based on average DPR factoring in crits and assuming that if making more than one attack, you take sneak attack on the first one that hits.)

krunchyfrogg · May 23, 2019

Thanks, I’d love that!

Esker · May 23, 2019

Here's an example:

Path 1: Level 4 Rogue with 18 DEX and Elven Accuracy, one attack with a longbow
Path 2: Level 4 Rogue with 17 DEX and Sharpshooter, one "power attack" with a longbow
Path 3: Level 4 Rogue with Crossbow Expert and 17 DEX, two attacks with a hand crossbow

Both have: 2d6 sneak attack, +2 proficiency, assumed advantage on 1st attack, and for path 3, assume an ally adjacent to the second target to proc sneak attack if the first attack misses

Path 1 (Elven Accuracy):

Damage on Hit: 1d8+2d6+4
Average Damage on Hit: 4.5 + 2 * 3.5 + 4 = 15.5
Avg Extra Damage on Crit: 1d8+2d6 = 11.5
Attack Mod: +6
Base To-Hit vs AC 15: 0.60
To-Hit (Advantage): (1 - (1 - 0.60)^3) = 0.936
To-Crit (Advantage): (1 - (1 - 0.05)^3) = 0.143
Total Expected Damage: (to-hit-adv) * (dmg-on-hit) + (to-crit-adv) * (extra-dmg-on-crit)
= 0.936*15.5 + 0.143 * 11.5
= 16.2

Path 2 (Sharpshooter):

Damage on Hit: 1d8+2d6+13
Average Damage on Hit: 4.5 + 2 * 3.5 + 3 = 24.5
Avg Extra Damage on Crit: 1d8+2d6 = 11.5
Attack Mod: +0
Base To-Hit vs AC 15: 0.30
To-Hit (Advantage): (1 - (1 - 0.30)^2) = 0.51
To-Crit (Advantage): (1 - (1 - 0.05)^2) = 0.0975
Total Expected Damage: (to-hit-adv) * (dmg-on-hit) + (to-crit-adv) * (extra-dmg-on-crit)
= 0.51*24.5 + 0.0975 * 11.5
= 13.6

Path 3 (Xbow Expert):

This one is a bit complicated because the damage that the second attack does depends on whether the first attack hit or not.

Base Damage per Hit: 1d6+3
Average Base per Hit: 3.5 + 3 = 6.5
Avg Sneak Damage (once per turn): 2d6 = 7
Avg Extra Damage 1st Crit: 3d6 = 10.5
Avg Extra Damage 2nd Crit: 1d6 = 3.5
Attack Mod: +5
Base To-Hit vs AC 15: 0.55
To-Hit (Advantage): (1 - (1 - 0.55)^2) = 0.798
To-Crit (Advantage): (1 - (1 - 0.05)^2) = 0.0975

Total Expected Damage:

(to-hit-advantage) * (base-damage + sneak-damage) + (to-crit-advantage) * (full-crit-damage) + (1st-hits)*[(base-to-hit)*(base-damage) + (base-to-crit)*(crit-damage-no-sneak)] + (1st-misses)*[(base-to-hit)*(base-damage + sneak-damage) + (base-to-crit)*(full-crit-damage)]

= 0.798 * (6.5 + 7) + 0.0975 * (10.5) + 0.798 * [0.55 * 6.5 + 0.05 * 3.5] + (1 - 0.798) * [0.55 * (6.5 + 7) + 0.05 * 10.5]
= 16.4

Esker · May 23, 2019

For reference, if you can't get advantage at all, then Xbow Expert wins in DPR against every AC, since a second attack is worth more than a +1 to-hit and damage on the first attack, and power attacking is pretty much never a good option. But, if you can't get advantage from your owl, then paths 1 and 2 allow cunning action hide, whereas Xbow expert is using their bonus action to attack. So if hiding is available, and no owl help, then elven accuracy wins by even more than when owl is helping. And of course, it should be pointed out that if your DM uses cover a lot, sharpshooter gets a lot better.

krunchyfrogg · May 23, 2019

Well, I’ve been using a longbow so far. I really like the idea of getting the sharpshooter feat at some point, but I’m pretty sure when I hit level 4, I’m getting elven accuracy first. It boosts my DEX to 18 and with my owl familiars flyby/help ability, I’m going to be able to use elven accuracy a lot.

Maybe I’ll nab sharpshooter at level 8. I am a little bummed that I’d only be regularly using half the features of the feat. Adding a fighter level does make the power attack from sharpshooter a little more viable.

I had never really considered crossbow expert for this character, but it is a really powerful feat. It was a long time ago, but I had a fighter who concentrated on using that feat and he dealt out huge damage. Stacking that on top of a huge sneak attack could be devastating.

Thank you for your help and input.

ON EDIT: I cleaned up my language a little bit (I was distracted earlier). I don’t think grabbing a feat is a good move at level 8. I think my best move at that time is to improve my DEX to 20.

As I am about to advance to level 10, I might be best off adding a fighter level at this point. The archery fighting style is really a great thing to have in preparation for the ASI I will get at rogue 10 (character level 11). I know fighter 2 and 3 are very attractive, but nothing is better than more sneak attack dice.

Rogue 12 (character level 13) will be tricky. Do I spend that ASI on crossbow mastery and change the flavor of this character? Mathematically, I think an extra attack would probably be pretty darn good, especially since ranged rogues don’t use that bonus action as much as melee rogues do. It’ll be something to think about if I ever get that far. Heck, it might be a good time to add wizard levels here too. Bladesinger 2 would grant me a lot of bonuses, although I probably wouldn’t be as tempted to use this unless I were lucky enough to find a headband of intellect (natural INT is 14).

krunchyfrogg · May 24, 2019

Esker said:
For reference, if you can't get advantage at all, then Xbow Expert wins in DPR against every AC, since a second attack is worth more than a +1 to-hit and damage on the first attack, and power attacking is pretty much never a good option. But, if you can't get advantage from your owl, then paths 1 and 2 allow cunning action hide, whereas Xbow expert is using their bonus action to attack. So if hiding is available, and no owl help, then elven accuracy wins by even more than when owl is helping. And of course, it should be pointed out that if your DM uses cover a lot, sharpshooter gets a lot better.

The only thing I can think of to make to power attack portion of sharpshooter worth is to multiclass that one level of fighter. As I detailed in my previous post, if I’m going to take sharpshooter, I’m going to get that archery fighting style.

Esker · May 24, 2019

krunchyfrogg said:
I cleaned up my language a little bit (I was distracted earlier). I don’t think grabbing a feat is a good move at level 8. I think my best move at that time is to improve my DEX to 20.

As I am about to advance to level 10, I might be best off adding a fighter level at this point. The archery fighting style is really a great thing to have in preparation for the ASI I will get at rogue 10 (character level 11). I know fighter 2 and 3 are very attractive, but nothing is better than more sneak attack dice.

There's no doubt that archery fighting style is useful, but I think getting an extra attack has a better synergy with sharpshooter. Obviously it'd be great to have both, but in terms of priorities...

At level 10, let's compare Xbow expert and Rogue 10 to archery style Rogue 9 / Fighter 1, assuming elven accuracy, maxed DEX, and advantage on one attack only:

DPR by enemy AC: (Xbow / archery)

AC 14: 35.9 / 30.1
AC 15: 35.4 / 30.0
AC 16: 34.9 / 29.9
AC 17: 34.1 / 29.7
AC 18: 33.3 / 29.4
AC 19: 32.4 / 29.0
AC 20: 31.4 / 28.4

So Xbow expert provides more damage at level 10 and allows two targets, but has shorter range and prevents cunning action hide. The damage is close enough that you could well be better off with archery style if you find yourself needing a longer range and/or find that your owl is going down often.

Now let's go to level 11, and compare Xbow Rogue 11 to Archery+SS Rogue 10 / Fighter 1. The single classed build gets an extra sneak attack die (and reliable talent!), but the multiclass build is in a good position to get a net gain from SS power attacks.

DPR by enemy AC: (Xbow / archery + SS)

AC 14: 39.9 / 38.6
AC 15: 39.3 / 37.8
AC 16: 38.7 / 36.8
AC 17: 38.1 / 35.5
AC 18: 37.3 / 34.0
AC 19: 36.3 / 32.1
AC 20: 35.2 / 30.0

(by the way, power attack is a gain at all of these ACs so it's not that the SS character can do better by switching to a regular attack against tough-to-hit enemies)

So again, there are tradeoffs other than DPR here, but Xbow expert and an extra sneak attack die wins over archery+SS.

Gotta step away now, but will pick this up later with further progression analysis...