Larian Studios Forums Baldur's Gate III Baldur's Gate III - Build Discussion my perfect party that kill everything

So here's an example of what objective analysis looks like:

https://anydice.com/program/32408

Your average damage right now is almost 29 on a normal hit and almost 44 on a crit, thanks to Savage Attacker. If you didn't have Savage Attacker, it would be a bit over 27 on a non-crit and roughly 41.5 on a crit, so Savage attacker is adding +2 to +2.5 average damage. You have at least +12 to hit, and advantage against large enemies, meaning you should hit a target with 19 AC like the one we're talking about 91% of the time (more if you have other bonuses I'm not seeing), and you should crit with advantage and a crit range of 16+ (25% as you put it) 43.75% of the time, thanks to all the effort you put into it.
That means your average damage per hit against the target we're discussing is 0 * .09 (Misses, dealing 0 damage) + 44 * .44 (crits) + 29 * .47 (normal attacks) = 0 + 19.5 + 14 = 33.5, rounding up in ways that are strictly favorable to your setup at every turn. On average, it takes you 15 attacks against it to kill it with your fighter set up in the way you have displayed. Given your 9 attacks in the first round will on average deal 301.5 damage, most of the time you need an extra 218 damage from other sources to finish him off.

Meanwhile, if you were using the elixir which grants 27 Str and were using Great Weapon Master instead of Savage Attacker, you would be dealing a bit over 41 damage on a non-crit, and a bit over 55.5 damage on a crit. This is driven by the extra 4 damage from the Str elixir and the extra 10 damage from GWM. You would have 1 less to hit, since you would get +4 to hit from the elixir and -5 from Great Weapon Master, and would crit on 17+ instead of 16+. This would mean in turn you would miss 12.25% of the time instead of 9%, and would crit 36% of the time instead of 43.75%. However, because getting much higher damage when you do crit at some point offers much better returns than improving your chance of a crit in the first place, and because Great Weapon Master is just much stronger than Savage Attacker, you would be doing significantly more average damage.
Against the target we're discussing your average damage per hit is 0 * .14 (misses) + 55.5 * .36 (crits) + 41 * .5 (normal attacks) = 0 + 20 + 20.5 = 40.5, rounding in ways that are generally unfavorable throughout. That's 7 more average damage per attack, or about 20% higher average damage, meaning on average it takes you 12 attacks to hill it with your fighter set up in this way, 12.34 to be more accurate. Of a note though less than 2% of the time do you fail to get a crit in the first round and thus you get 10 attacks in round 1 98+% of the time thanks to Great Weapon Master, so you're dealing over 404 average damage in round 1, leaving just an extra 114 damage to finish him off.

So how do I know that Great Weapon Master is stronger than Savage Attacker, and the Strength elixir is a better option for the character as described than the crit elixir? Because the numbers say so. Because I can proclaim with confidence that 100 different people could do this fight 10x each with your setup and 10x each with mine, and *all of them* would have better average results with my setup than with yours. That doesn't mean that even with the elixir and feat swap outs we've arrived at anything even close to an optimal setup of course. You can do better than this, much better, if you're not a Fighter 12. That's entirely ignoring how weak he is to hold monster and thus the superior options you have with a well designed party. But I don't rely on just how it feels to me as I run through it once. I don't just look at something that worked and assume it's the best option and everything else is trash. I can compare what sort of outcomes are expected with tweaks to the setup and see which one will be more favorable, and determine with absolute confidence that one is better, *and I can then back it up if questioned on it*.

still killed elder brain in 6 hits move along

giant form +6 dmg 41 + double str mod not showed in sword window lol total and this sword double str dmg bonus, your math is bad and you dont dont mechanics of the game, move along like i said again elder brain died on like 6 hits, can your monk do that ? nope

i do it english

Giantslayer: On a hit, double the damage from your Strength Modifier. This weapon grants you Advantage on Attack Rolls against Large, Huge, or Gargantuan creatures.

and btw great sword mastery dont do more damage if you miss all the time, accuary do more damage than raw damage, this feat is good only for barb with reckless attack xP sorry you talk to 15 year mmo veteran xD

5% crit rate also do more damage than 7 str in general

Giant Form is +1d6 not +6 and double Str mod is + 5 for less than +10 average damage lol. "elder brain died on like 6 hits" can your fighter do that? Nope. Outright impossible, blatant lies as demonstrated by your own screenshots. Just saying "your math is bad" doesn't actually replace providing any sort of evidence or analysis or rational arguments you know. Saying something is true doesn't make it true, no matter how much you want it to.

Yikes

Be careful guys, veteran here



Lol, your every comment is a giga fail

Let’s say without crit and without +7 str, you do 10 dmg.

After 100 hits you will do 100*10=1000 dmg

With 5% crit, you will do (95*10)+(5*20)=1050 dmg

With +7 STR, which is flat +3dmg, you will do 100*13=1300 dmg

Maybe you should’ve gone to school instead of playing WoW but even in wow, crit is the worst stat. So you didn’t learn anything

Crit is fine, but to know whether base damage increase or crit increase will do more damage, you have to actually run the numbers. There are certainly times in D&D where more crit is better than more damage, depending on the situation and the figures in question. But that's why we ran the numbers, and mathematically proved that while sometimes "accuracy do more damage than raw damage" in this case it definitely does not, thus the massive increase in average damage with the other config. You can say "great sword mastery don't do more damage", I can *prove* that in this case it does, substantially more, and have already done so. It also on average adds far more damage than Savage Attacker *even if you never have the passive turned on* due to gaining the bonus action attack on a crit, which is just fundamentally more valuable than +2.5 damage on a crit. When getting attacks with an average damage in the 30s, you can have 3, 6, or 9 attacks in a turn, it doesn't matter. Getting an extra one will be more valuable than an extra 2-2.5 damage per attack, and the more attacks you get the more likely that crit becomes.

It's not complicated to figure out when it should be on or off, either - if (percentage to hit) * (damage with it off) > (percentage to hit with it on) * (damage with it on), then don't use it, otherwise, do. In this case it's a matter of going from a 97.75% chance to hit if it's off, but dealing 31 damage (45.5 on a crit) for an average damage of 0 + 45.5 * .36 (16.36) + 31 * .62 (19.22) = 35.5 with it off, vs the 40.5 with it on. Meaning it provides an extra 5 average damage (15% more), *even without considering the bonus action attack*. So in this fight, it should be on. But you know, you can only really figure it out if you understand and care about the numbers.

The Accuracy valuation over Crit is very much a Vanilla/Classic World of Warcraft meta - and *for THAT game* it turned out to be very true. In Vanilla WoW anyway you always prioritize hit rating over everything else until you reach the 9% cap.

I don't know if it really works for BG3/5E as a meta since your hitting against a much smaller sample of more variable opponents with wildly different AC. in WoW bosses all had the same defense values and you are talking thousands of hits over the course of a fight.

Other factors which skew it in BG3:

- Crits only increase dice damage not all damage, so they’re more valuable on setups where damage dice represent a large portion of the damage, like Rogues and Paladins, and are less valuable on setups where high static numbers (such as double strength bonus) are the majority of your damage, as is often the case with Fighter and to an extent Monk

- You auto crit paralyzed targets, so your crit range becomes entirely moot once you land a hold person or similar effect. This both makes crit enhancing effects (like the half orc racial) more valuable since you have a reliable trigger, and makes crit range increases less valuable. As an auto crit is also an auto hit, it also skews the effect of accuracy boosts, though of course these effects are situational

- Advantage conversely makes crit range bonuses more impactful than they otherwise would be. With advantage, +2 crit range isn’t taking you from 5% to 15%, it’s taking you from 9.75% to 36%, a far higher percentage of your attacks and more importantly a far higher percentage of your hits. This means you need more damage, proportionally, than you otherwise would for raw damage to be better than a crit range increase if you have a reliable source of advantage (which with a crit build, you certainly should).

Of course these factors (and all the normal ones) play out differently for different enemies across different ACs with different initial damage amounts and damage makeups. But while a choice might be right for one situation and wrong for another (increased crit range might result in a damage increase generally but a damage decrease Vs paralyzed opponents as an obvious example), it shouldn’t be too hard to figure out generally which is better and what situations you need to watch out for, and how common those situations might be.

But yeah, in this case the strength elixir and GWM is *always* better than the crit elixir and savage attacker, against all enemies and Vs all AC targets. This is without even having properly taken into account the double strength damage, which only swings it more in favor of that setup of course. Whether Vs enemies who are easy to hit, in which case +16 damage is better than +2 to hit and +1 crit rate, or against enemies which are hard to hit, in which case +3 to hit and +6 damage is better than +1 crit rate. There is simply no situation whatsoever where the crit elixir and savage attacker is better.

I mostly agree, but obviously with Ranged attackers like, well Rangers using say Gontr Mael instead of Titanstring Bow it makes sense to use the Elixir of Viciousness OR Elixir of Bloodlust + Sharpshooter depending on what the rest of your loadout looks like. If you are already stacking Crit, then Viciousness' makes a bit more sense especially for the Gloomstalker Ranger build or even a Dex Fighter build. I have minsc specced out this way and am currently testing between different build types to maximize that ranged attack ability.

I was planning to look at Dex fighter because I think the Improved extra attack is more consistent damage in a longer fight. (not that any fight lasts that long with Jaheira as a Monk/Thief murdering everyone in round 1)

you have 0 idea about how probability works the more crit you stack the more you get from it, 5% its even more wroth in this game where you have little sources for crit chance while you gave shitloads sources for raw damage

your math is wrong cause you dont count the whole crit chance that you get from 15% to 20% for example

if you are lucky you can get even more damage than raw damage if not you do less

you do less damage with -25% accuary and 10 raw damage that do normal damage and 25% more accuary

its always better to have 95% accuary than 70% its even more you lost with bosses if you drop for 60% accuary orceven less if you dont have support like reckless attack

this is common knowleage from mmo and how probability works cause when you miss you do 0 dmg and even you can get parry or riposte from it and lost even more in extreme causes

Yeah see this is the trouble, you're just not credible when you make statements like this. If you have 10 average base damage, and hit 95% of the time, that's 9.5 average damage per hit. If you hit 70% of the time, and deal 20 average damage, that's 14 average damage per hit.

So clearly it's not "always better to have 95% accuracy than 70%", you deal more damage with the increased damage and reduced accuracy even in the most basic of examples. Now, if you deal 40 base damage and have 65% accuracy that means you have 26 average damage, but if you reduce that to 40% to deal 50 damage then your average damage is only 20, so it's not Universally better, but the fact is, it's simply not universally true one way or another - you need to actually understand the numbers and to be able to run the calculations to determine when it is and is not better. Your "common knowledge from mmo" may well be universally true in the MMO you learned it, but that is *not* the same as it being universally true.

It's also not universally -25% to hit! In your own example, you're attacking a large creature with auto advantage due to your weapon. In case anyone here doesn't understand how advantage impacts your odds, here:

https://anydice.com/program/2345

Let's focus on the "at least" tab. To be -25% when you have advantage, you need to go from a 2 (99.75%) to an 11 (75%), or a -9, not a -5. Going from needing a 3 to an 8 is going from 99% to 87.75%, a drop of only 11.25%. This is partly why it's so good for barbarians since with reckless attacks you can get universal at will advantage. Monks can also get advantage at will (not that they need it due to tavern brawler), but there's plenty of ways to obtain it, whether from the risky ring (better on ranged characters imho) or the item which gives you advantage when surrounded by 3 or more, or knocking enemies prone, or *using the sword you're using* against large enemies. Just to name a few. And advantage is potent stuff, too - if you would be at 60%, meaning you need an 8, instead you could be at 87.75%. That's a nearly 50% increase in hit rate (.9 is equal to .6*1.5), so advantage is always worth pursuing in general.

There's also a cap on how accurate you can be - if you're someone with Str 27 (+8), and a +3 weapon (+3) they're proficient in at level 12 (+4), you have +15 to hit. If the enemy AC is 12, you hit on a 2+, because a 1 *always* misses. But if you took a -5, giving you +10, you still hit on a 2+! This means you go from 95% to 95%, haven't lost any accuracy *at all*, and gained 10 damage. Since you want high to hit bonuses to help out against the AC 27 creatures of the world, it's nice to have a way to shed some of that "overkill" accuracy against weaker foes, to ensure you take them out faster.

When assessing the choices in a game, you need to understand the mechanics of how *that game* works, not how a different game entirely works. In this game, with the character you provided, you're better off with a Str Elixir and Great Weapon Master being on in the fight in question, and against all targets with an AC of 20 or less. You do MORE damage that way than you are currently doing with the crit elixir and without Great Weapon Master. The math has already been provided, if you have questions about it or don't understand something I'd be willing to clarify, no problem. But until then just know that the statements you made here are blatantly, demonstrably false, in multiple ways. Which unfortunately has been a trend.

i dont even click on that link

i just loled, so you telling me math and probality work diffrent in bg3 than in normal world ? xD xD xD

dude i use math that is univeral

25% accuary is wroth more on 20str cause 20 str is soft cap in this game
crit rate is better if you stack it, more you stack more you get from it and more str you more you stack you get less from it

this is something what is called diminishing returns

0-5% crit rate is use less
5-10% too

but if you stack it to 20 or 25% is wroth cause this how nature of RNG works

also there is better get str from feat than gwm cause with this move you have 30% accuary more and this give much more damage than waste this point on gwm

now guess why crit builds are much popular than stack it to raw damage, this is that you can do burst damage with crits that how i rekt elder brain in one turn by my champion

Yeah, well, if someone doesn't want to learn or try to understand at some point it ceases being worth engaging in conversation with them.

Crit has diminishing returns, it is *not* better the more you stack it. Something like AC or Damage Reduction is better the more you stack it, because you're effectively subtracting from an ever smaller pool. Taking them from needing a 12 to hit to needing an 12 to hit is reducing their hit percentage from 50% to 45%, it's a 10% reduction in incoming damage. Taking them from needing an 18 to hit to a 19 to hit is reducing their hit percentage from 15% to 10%, a 33% reduction in incoming damage. Crit doesn't have anything like that, it's a multiplier for a segment of your damage only, and like any multiplier you need to make sure that it's sufficiently large and thus multiplies effectively, *and that there's something worth multiplying coming in*.

The crit percentages aren't 5% or 10% or anything either, had you clicked the link or if you understood how advantage worked you would know that, but I can have a crit on an 18+ and crit 27% of the time while you're getting a crit on a 16+ and only critting 25% of the time.

Getting Str from the feat is never worth it since you can just set your Str to 28. Investments in Str on Str based characters are literally wasted as a result of that. Or gloves to set it to 23 if you want to go that route. If you dumped feats into Str and Savage Attacker, you're making an absolute trash character compared to what they could be.

You didn't rekt the elder brain in one turn, you shared screen shots and showed you were lying and didn't have the damage to do the thing you claimed to do, as usual.

But the funniest part of all of it, is that generally stacking crit percentage is a total waste. If I wanted to crit all over the place, I'd play an assassin, and auto-crit surprised enemies, with a huge pool of bonus dice to make that crit actually mean something. Or better yet, have someone with hold person or hold monster in the party, and paralyze the enemy. Then *all* of my characters auto-crit the enemy, and I have 100% crit chance across the board. At which point a character who has invested in stacking to 25% has completely wasted that effort, and if they had instead focused on dealing greater base damage their crits would hit harder.

Basically nothing you have said is true. As usual you're wrong, in like 5 different ways.

While it is a trivialistic d20 rulesystem, I love how in SWKotoR you can stack Master Critical Strike and Keen on dual lightsabers (need the appropiate endgame crystals though) and get a critical hit on 11-20, i.e. over 50% of the time (since rolling 1 is a critical miss):

Lightsabers crits on 19-20 (thus one uses dual lightsabers and not a twohanded lightsaber, which only crits on 20)

Keen Lightsaber would double the base crit range, thus crit 17-20 (while keen twohanded lightsaber would only crit 19-20)

"Regular" Critical Strike would add another extension: 15-20

Improved Critical Strike would do the same again: 13-10

Finally Master Critical Strike did this the third time: 11-20

That was pretty wicked. Some opponents would unfortunately be immune to critical hits, though.

Even worse, if you made a Strength focused build, well Strength would decide how hard the DC to also stun opponents would be. Again certain opponents would be immune to that, too.

All in all Critical Strike was so overpowered, and definitely better than the other two choices for weapon attack style, except if you faced opponents with immunities (not that often).

Flurry instead would give you another attack. Not impressive with Dualwielding and Master Speed (4 attacks total).

Power Attack gave you extra damage per attack. With the best gear and Master Speed, it was about as good as Flurry.

Never seen anything like this in any other game.



In BG3 and D&D5 unfortunately all weapons naturally only crit on a 20, and I dont think there is even a single keen weapon in the game (?). Fortunately as people mentioned certain states such as paralyzed allow you to always autocrit anyway.

Unseen Menace and Knife of the Undermountain King both crit on 19, though the Knife actually applies that to all weapon attacks you make so it's effectively +1 crit range. Other weapons like Dead Shot or Bloodthirst can improve your crit range, and since they stack you can get +3 crit from your weapons, +1 from Champion, +1 from the Elixir, +1 from your helm, and a conditional +1 from your cloak, letting you crit on a 13+ if hidden. With advantage this is a 64% crit rate. That doesn't make it a good build of course, absolutely none of those things help with the damage you do when you crit so baseline the only benefit of all of that is an extra d4 or d8 damage 64% of the time meaning it's worse than Hunter's Mark. But if you want to crit fish, you can do it.

Sword of Life Stealing gives a return on getting crits, as do the Craterflesh gloves, and of course anything that gives bonus dice of damage in general. A good crit build can certainly be built by balancing crit enablers and payoffs of course, but yeah, all of that effort and it'll still be less reliable than just paralyzing them.

crits have diminishing returns over like 60% crit rate xD, yes crit chance is better when you stack it, 5% to 20% crit rate give more than 60% to 80% in terms of probality, again this is math and you have no idea how probality works

again crit stack is better when you reach soft cap of raw damage, you show again you dont know mechanic of this game, crit needed for crit for dice 19-20 insted just 20 its 5% more crit chance same with accuary, 2 str gain 5 % accuary for melee

you have 5% crit rate just starting game when you roll 20
champion lower this to gain crit you need roll 19 or 20 its 10% rate
potion stack give 18-20 in roll dice its 15% rate
rare bow stack it to 17-20 its 20% rate
helm stack it to 16-20 its 25% rate

25% rate give more damage with 20 str then you do damage without crit with 27 str, fighter dont need other stats then str and con, even you stack potion to 27 other stats for fighter are useless trash

omfg yes i rekt elder brain in one fucking turn, not my fault you dont know game mechanic and you are not able to count fucking fighter can do 9 attack in one turn when its hasted, so stop telling me i lie fucking scrub who dont even know basic of math and use some shity orogram that also count wrong

your brain is trash cause you dont know how probality works and how savage attacker works that you never deal minimal damage from great sword and this is what rekt with crits elder brain in one turn

my sorc,paladin or barb in his place, fighter and life cleric make this game joke even on hard mode and they should add difficulty harder than hard cause this game is way too easy

The trouble isn’t the minimum damage, it’s the maximum- even if you hit it (which with your damage pool there’s a less than one percent chance of that) with every attack for the claimed 6 attacks to take it out, you’re nowhere close to enough damage, and that’s based on the screenshot you yourself posted. There’s literally no options other than you’re lying, it’s plain to see to anyone who looks at the information you yourself presented.

Speaking of probability, yep you spend all that effort, get to 16-20 and have a 25% crit rate. Meanwhile, I ensure I get advantage on attacks, because I know how valuable it is. I thus start with 9.75% crit.
19-20 gives 19% crit
18-20 gives 27.75% crit

As promised, I’ve got more crit chance with 18-20 than you’ve got with 16-20. Nothing inherently makes crit more valuable if you stack it. I understand perfectly well how probability works, as anyone would know had they followed the links, assuming they’re capable of understanding the data displayed.

We even specifically covered the damage of 20 Str, crit elixir and savage attacker vs 27 strength, GWM and strength elixir. The latter setup was doing nearly as much on a non-crit (41) as you were doing on a crit (44). Overall average damage was calculated, and it was way higher with the GWM setup. You never offered any counter arguments or math of your own, you just lol’d.

you cant even count how much is 1 from 20 ?xD your math is wrong and you show it all the time

1 from 20 is 5%

19-20 its 10% crit rate,

again you show that you have no clue how probality works

ofc i lie rofl, nope your math is just shit go back to elementary school and learn how to count

its took me 1 action to kill elder brain if it was not 6 then maybe 7 attacks, i know i dont used whole 9 attacks to kill it, and if your shity math dont match its not my problem, this is my personal achievement that your monk just cant do unless its astorian as high vampire

you dont even know how savage attacks works and how good its rolling for damage 2 dices insted of one and how good it is if you pair it with great sword fighting style

with 95% accuary its anti rng build cause rng is shit and its non skill tatican crap if you can lost cause you have bad luck which is ridiculas

all my build must have 95% chance to hit, for me 90% accuary is miss in my tatican brain and for 90 % i know this shity game give me miss xD

my meles have 95%, my ranger if i have one and my sorc have all magic accuary items like 25% acc from items or 20% dont remember

Hey, nice. Thanks for the information.