PDA

View Full Version : Optimal Intelligence/Wisdom Scores for an Arcane Trickster Rogue



Expected
2018-09-18, 09:15 PM
My character is completely planned out with the exception of his Int and Wis scores and my choice of subrace (High Elf for 16 Int, or Wood Elf for 16 Wis). I am using point-buy and will have 8-17-15-?-?-8 with 9 points available for Int and Wis. High Elf gives +1 Int and Wood Elf gives +1 Wis.

For RP reasons, I'd prefer to be High Elf if my character's Int score will be high and Wood Elf if not. The reason behind this is the Forgotten Realms lore--Wood Elves detest arcane magic and do not prioritize education and intelligence, whereas High Elves are talented in magic and value education.

What do you think? How important is a high spell save DC? I plan on engaging in melee attacking for Sneak Attack instead of casting spells, but I'd like to have the ability to cast spells without the DC being too low, thereby making spells with saves much less useful. Should I max one and dump the other? Balance both scores out?

Specter
2018-09-19, 07:58 AM
Certainly High Elf. A Wisdom score of 15 is already massive for a Rogue, and futurely you'll get proficiency is WIS saving throws. In the meantime, 16INT makes sure you have a good save DC for when Magical Ambush comes (you don't need to raise it further).

Bloodcloud
2018-09-19, 08:49 AM
High elf all the way. The extra cantrip could be helpfull too, look at greenflame blade and booming blade.

JellyPooga
2018-09-19, 09:00 AM
Wood Elf. The extra speed goes well with Cunning Action and you can never have enough Wisdom.

KillingTime
2018-09-19, 09:13 AM
High Elf
The extra cantrip is golden, and though Wis is certainly an important stat, an Arcane Trickster will definitely want decent Int.

Personally I'd drop that odd point in Con and go:
8-17-14-16-12-8
Or even swap Con and Wis entirely.

As a rogue you should be avoiding getting hit altogether, while the increased Int and Wis will aid all of your perception and investigation type utility, which is what you should be bringing to the party.

(I'm assuming you'll take elven accuracy (dex) at 4th)

Expected
2018-09-19, 09:53 AM
High Elf
The extra cantrip is golden, and though Wis is certainly an important stat, an Arcane Trickster will definitely want decent Int.

Personally I'd drop that odd point in Con and go:
8-17-14-16-12-8
Or even swap Con and Wis entirely.

As a rogue you should be avoiding getting hit altogether, while the increased Int and Wis will aid all of your perception and investigation type utility, which is what you should be bringing to the party.

(I'm assuming you'll take elven accuracy (dex) at 4th)

Yes, I will be taking the Elven Accuracy (Dex) feat at level 4. I was also considering Resilient (Con), hence the odd Con score.

I'm going to be taking Sentinel for off-turn Sneak Attacks so I need high Con to survive. Blur and Mirror Image will help mitigate damage.

Expected
2018-09-19, 07:23 PM
Then again, would a High Elf's 8-17-13-16-12-8 be the optimal ability score allocation? I'll take Resilient (Con) to get 14 Con. Will I be too squishy with 13 Con on the frontlines even though I'll have Blur/Mirror Image? I will be dipping 1 level into Fighter for shield proficiency and Fighting Style (Defense).

Galadhrim
2018-09-19, 10:12 PM
Then again, would a High Elf's 8-17-13-16-12-8 be the optimal ability score allocation? I'll take Resilient (Con) to get 14 Con. Will I be too squishy with 13 Con on the frontlines even though I'll have Blur/Mirror Image? I will be dipping 1 level into Fighter for shield proficiency and Fighting Style (Defense).

I played this build and did not find it to be too squishy. With a a starting AC of 18, moving to 20 when you max dex, you should have the AC for the front lines. Add to that uncanny dodge if you take a crit and evasion for dex saves, and you are plenty tanky. Not to mention if you get overrun you can use your cunning action to escape.

I think you will like having an int of 16, especially once you get magical ambush. Disadvantage to save against hypnotic pattern or fear can be encounter breaking even at high levels.

Once you get to rogue 9/fighter 1 you might consider a second level in fighter for action surge. Hold person (they save at disadvantage), action surge, crit sneak attack.

Edit: recommend mirror image over blur and would not take both as you have very limited spells.

LudicSavant
2018-09-20, 01:23 AM
How important is a high spell save DC?

Let's say a target has +5 to their saving throw. In that case, the chance for them to fail their save based on your DC looks like this:

https://i.postimg.cc/vHY4kks4/Arcane_Trickster5save.png

Now, to someone who doesn't understand probability math, they might think "eh, getting hit 5% more or less often doesn't seem like too much." These people are of course wrong, because here's how effective durability really works:

https://i.postimg.cc/QtnWqqF6/Arcane_Trickster5save_HP.png
Note: Chart is relevant to "Save Negates" type effects. "Save for Half" graph is a little different

Why is this? Well consider the difference between a character who only fails a save on a 1 or 2 vs another who only fails on a 1. The first one doesn't get hit "5% more often" they get hit twice as often. 2x as many numbers hit them. They are twice as vulnerable to this attack, status effect, whatever.

Basically, to double your effective average durability you need to raise the enemy's chance to miss from 0 to 50%. To double it again you only need to raise it from 50 to 75%. To double it again you only need to raise it from 75 to 87.5%. To double it again you only need to raise it from 87.5% to 93.75%. And so on and so forth. Basically, the moral of the story is that raising your AC from 20 to 21 is worth a lot more than raising it from 10 to 11.

Corran
2018-09-20, 06:33 AM
]I'm going to be taking Sentinel for off-turn Sneak Attacks so I need high Con to survive. Blur and Mirror Image will help mitigate damage.
Multiclassing in battlemaster fighter is very important to a rogue that wants to play as a dedicated frontliner and make good use of sentinel. Very briefly, the reason is that to make good use of sentinel, you need: 1) a high AC (the fighter helps a lot in that respect) and 2) the riposte maneuver. No matter how counterintuitive it might seem to some, you really need riposte and a good AC to make good use of sentinel. Taking fighter levels also lands you many other good stuff but I wont go into that in this reply.

And if you are planning on playing a fighter/rogue multiclass, I would say that starting as fighter is better as it lands you better save proficiencies than by starting rogue (CON save proficiency is needed because you will want to use a few concentration spells (mainly blur and protection from evil when it applies, enlarge is another candidate if you went with the shield master feat), and it is also important to be proficient in wisdom saves. So starting fighter lets you get both (at the cost of losing dex; but with a DEX score of 14-20, depending if you went str-based or dex-based -both approaches have their con's and pro's-, and with evasion eventually, it's a loss worth suffering for the gain you get).

Mirror image is not all that good on characters that want to stay in the fray, and sentinel rogues do want to stay in the fray as opposed to the skirmishing of a more traditional rogue (sentinel rogue plays more like a fighter than a rogue, combat tactics wise). That is because it will last very briefly (only good as panic button, and you don't want a panic button, you want a reliable way to boost your effective AC -enter blur and protection from evil-, which indirectly but very surely, leads to a much more reliable use of the sentinel reactionary attack).

Regardless if you are going dex or str based, shield master (without the completely unnecessary nerf) is a very good way to optimize your action economy by getting access to a reliable use of your bonus action (remember, the sentinel rogue is not using cunning action to hit and run, he wants to stay up front within 5' of his enemies to make good use of the sentinel feat.

As I said earlier, riposte is very necessary (I will not explain why in this post, so I can keep it relatively brief), so if you are already planning for 3 fighter levels, then you might as well go up to 5, for extra attack (very important, for giving you another chance to sneak attack -as the ideal setup of the sentinel rogue IMO is S&B -regardless if you take shield master or not-. That's cause as I said in the very beginning, you really need to boost your AC as much as possible. Having a high effetctive AC -and access to riposte- is what makes the sentinel rogue both viable and extremelly efficient, I cannot emphasize this enough! Also, extra attack is useful when you want to shove (all of SM, enlarge and athletics expertise help here), to cancel out potential disadvantage (other ways to try to cancel disadvantage -which shuts down your sneak attack, so having ways to cancel disadvantage is a very big deal- are through a familiar and through the feinting battlemaster maneuver (I would pick it as my 3rd maneuver, along with the very important riposte and precision; feinting complements your maneuvres in a very efficient way, as it is situationally golden, and you need a good situational maneuver since riposte but mainly precision will need all the superiority dice you can give them).
And if you take extra attack (ie fighrer 5), you don't want to aim for a fighter5/rogue15 split (unless you are starting at level 20), cause that would mean not having the very important wisdom save proficiency till you are 20th level (so your poor wisdom saves would be a huge weak point of your build). So 6/14 or 7/13 are good splits for a 20th level battlemaster sentinel rogue.
Edit: The feinting maneuver (and to a lesser extent uncanny dodge) play well with your back up tactics which you apply when you find your sentinel rogue fighting alone (eg you melee buddy -on whom you rely for triggering sentinel- drops unconscious, or your scouting failed). Then you want to play as a skirmisher, so you use feinting to give yourself advantage so that you can apply sneak attack damage -also to improve your chances of delivering a good payload of damage- and BB with your action (you move away taking OA's -since feinting uses your bonus action, so you can't use cunning action disengage- which is how uncanny dodge comes in).

Without giving any reasoning (to speed this up), fighter 1 -> fighter 1/rogue 5 -> fighter 5/rogue 5 -> fighter 5/rogue 7 is the progression I would follow till character level 12. I would take sentinel at 9th level and not sooner (too risky before that IMO; perhaps you can pull it off, depending on the difficulty of the battles in your campaign, but still, I would choose not to rush it). Other good feats are SM (as I mentioned earlier), resilient wisdom (also mentioned earlier), maxing attack stat (whether it is dex or str) and boosting con with the last (couple) ASI's.

If dex based, I would start with stats roughly (due to not determining the race) like this:
STR 12, DEX 16, CON 16, INT 8, WIS 13, CHA 8.

If str based, I would go with stats (roughly) like that:
STR 16, DEX 14, CON 16, INT 8, WIS 10, CHA 8.

Both approaches have advantages but I think that what weights a lot into this decision, is if you want your character to be stealthy. If you want both to be str based and to be stealthy, a nice compromise is selecting earth genasi as your race, as it has stats suited to the str based build, and it also allows you to cast pass without a trace once per day, which kind of cancels the disadvantage to stealth checks due to heavy armor (later on, stealth expertise and reliable talent help even more in that respect). Of course they miss darkvision (which IMO is very counterintuitive after having read their lore text), so it's far from a perfect fit.

Other good choices for a race is variant human and stout halfling (for the dex build). Both lack darkvision.



One could mess a bit with the starting stats, lowering con and boosting int (to make some better use of the trickster's spellcasting). I prefer spending the points on CON (as this character is a frontliner, and apart from self buff casting -paired with action surge so that you wont miss on sneak attack opportunities-, this build has no place for spellcasting in combat. But it would be viable.

So, the int version, would essentially lower CON to 14 and use the spare points for boosting INT (this works better for a dex based build, since the dex build has more spare points than the str build, due to wanting less points on a tertiary stat; the dex build wants 12 str, the str build wants 14 dex). In this case, any race pumping DEX and one of CON/INT is a good choice. Preferably one with darkvision, as scouts need darkvision, and since you went dex based, then you need to be a scout to make the most out of it (at the cost of 1 AC, important cost as I talked and talked a lot about the importance of AC for this build, but I think that stealth is worth it; else, why did you go with a dex build? Initiative and dex saves are not enough on their own to justify it, at least for this builds that has good uses for athletics and that needs every point of AC it can get -heavy armors grant you a better AC-).

PS: Obviously, to make the sentinel rogue work, you need an ally opting to fight next to you (also he should not take the sentinel feat). A durable GWM melee buddy is the ideal such ally, and you need him to make for a more valuable target than you to the enemy (we are doing our fair share optimization wise to work towards that direction -ie throwing off of us the aggro that sentinel draws on us, via boosting our effetcie AC and by taking riposte, but our melee ally buddy plays a part in that too, so if he is a GWM it helps, if he is a turtle it does not help). It's all about making sentinel trigger as often as we can, that's what we are truly optimizing here.