Larian Studios Forums Baldur's Gate III Baldur's Gate III - General Dice Rolls Do Feel..Off?

I mean, I made a Cleric of the Light Domain. I have a +4 to hit..
..And I ended up missing FIVE 78% attacks in a row.

So I get to the beach. I cast Sacred Flame; a 41% chance to hit while I have 14 Wisdom, and Clerics cast off of Wisdom. That feels AMAZINGLY low.

Meanwhile, the Intellect Devourers on the beach did 11 damage in one hit to my Cleric and 10 to Shadowheart. This instantly downed both of them in a single round of combat.


I'm against save scumming, but I tried it to save the little Tiefling girl; it took me FIFTEEN MINUTES of rolling and reloading before I succeeded a DC 11 check. I mostly rolled 1-4 with a few 6-9 sprinkled in. Assuming 1 roll every 30 seconds (to allow for reloading, triggering the dialogue, picking the proper options, and seeing the roll):
That was THIRTY ROLLS in which I never hit or broke 10 on a 20-sided dice.

Someone who knows probability, please tell me what the odds are of only rolling 1-9, 30 times, on a 20 sided dice?

I don't know what difficulty the early access defaults to, but all player dice rolls feel weighted towards low numbers, while the monsters are getting all the higher rolls.

I've been having the same issues with my spellcasters. My Warlock has as much Charisma as is possible at this stage and I'm still consistently seeing very low chances to hit, even at the appropriate range, and even with height and surprise advantages without any enemies near me. Meanwhile, Astarion will be an entire screen length away with a 50% chance to hit and nail 8 out of every 10 arrows while I'm missing with Eldritch Blast a grand majority of the time.

I legitimately don't understand what else I can do to get these percentages higher, and it's made my choice to do a 3 spellcaster playthrough (Gale being the third) for my first run an absolute nightmare and barely viable.

I play D&D with a virtual tabletop program where a computer does all the rolling. Without fail, some players always feel like the dice rolling is “broken” because they keep getting low rolls. I even had a player go as far as to record like 100 rolls to see if there was some kind of “bias.”

But really, it’s just bad luck. There is no gremlin in the machine making your rolls suck. It happens with real dice, and it happens with a computer generating numbers. Sometimes, you just roll badly.

The big goblin fight outside the gate was a gigantic missfest. So many people, on all sides, missing like crazy. The upside of it was that none of the mercenaries died, thanks to all the goblins missing often as well.

Intellect Devourers have AC 12 / higher saves than in the game: https://5e.tools/bestiary/intellect-devourer-mm.html

BG3 currently runs an AC7 intellect devourer for their intro /1st act.

I feel like I'd rather see people hitting more across the board than missing. All of my care and strategy for setting up proper positioning and win conditions goes out the window when everyone's missing, and it doesn't feel particularly rewarding to set up intelligently before fights.

I've managed with 3 spellcasters on my team, but I imagine that people who took Lae'zel and Shadowheart instead of Gale and Wyll are having a much easier time breezing through.

I know that; I've programmed a number of random number generators while practicing in Python 3.

But given the sheer number of people reporting low rolls and the fact that the game outright tells you your to-hit chance, it is also possible Larian misplaced a decimal or added an extra 0 somewhere and skewed everything.
Heck, not only possible, but probable. There's always that time you hit an extra key, or your keyboard did not record a keystroke, and you just missed it, because it did not break the code.

1 or 10 people having low rolls is 'just bad luck' but I have seen a number of people, between this forum and Steam's forum, mentioning low conversation rolls, and the game outright SHOWS that spellcasters to-hits are sub-50% in almost every fight.

This, to me, hints that something may have been missed in the Early Access.
And since it's EA, it's our duty to report this so Larian will know if enough people have the problem, which changes from 'just back luck' to 'We missed something'.

And even if it IS just back luck, the fact that spellcasters are frequently getting upwards of 30%-40% less to-hit than physical attackers is a balance problem.

It's actually a pretty simple calculation. The odds of rolling anything on a die are (c/s)ⁿ where c is the number of sides that satisfy your condition, s is the total number of sides on the die, and n is the number of times you roll the die.

Given you roll a 20 sided die 30 times in a row and that each roll is independent, the odds you roll nine or below on each are P = (c/s)ⁿ = (9/20)³⁰ = 0.000000000039479843

You have a 00.000000003948% chance of rolling nine or below 30 times in a row.

Thank you. Yeah, that goes beyond 'just bad luck'.

Computer die rollers are wonky, I run games on Roll20 and see huge strings of bad rolls in every session. Thankfully the bad guys often miss as well.

I haven’t noticed this at all. I seem to hit and miss in a way that doesn’t seem to disagree with the stated percent chances.

If you’re getting low chances to hit enemies, there may be a reason. If you’re using a ranged attack and an enemy is threatening you (next to you), you will roll with disadvantage. The game also accounts for concealment, so that can hurt your chances too.

Finally, I’m pretty sure you can see your rolls in the combat log. If you think the percentages are wrong, it’s not hard to check.

Just wanted to point out that Sacred Flame's save is based on the target's dex, not their AC. A save negates the spell (or causes it to miss in BG3). The saving throw is the target's and not yours so your wisdom score doesn't factor in.

I assume the target was an intellect devourer. I found a dead intellect devourer on the beach and examined it. They have 14 dex which is +2 on their saving throw so 41% doesn't sound crazy to me.

IMO the problem here (regarding sacred flame) comes from lack of explanation of the dnd mechanics and arguably misleading default stat attributions. 14 Wis on a cleric is somewhat low, which really doesnt help. To be fair, having optimal stats (16 or 17) in the situation you're describing would result in only a 5% better chance of hitting the sacred flame, because intellect devourers have a pretty decent dex save. But that case in particular feels like a design choice. Can't make it with dex saves? how about regular attacks or other saves? Coming from a pure tabletop perspective, monsters usually have a bad stat that you can exploit.

Hitting dex saves abilities might also be a case of the devs wanting you to cheese with ray of frost (if that's the case, i despise that fact), because it creates an ice surface which might make the target prone. Why is this important? because prone targets automatically fail dex saves, which would give you 100% chance to hit your sacred flame. This, imo, makes the game feel way too much like a DOS.

That said i totally feel you on the 78% and still miss, I had the same feeling of having the rng stacked against me for a solid portion of my playthrough.

I've seen a Crab do an attack of 2d4 deal 1 damage on board of the Nautiloid

For the brain devourers, shadowheart tells you 'keep your distance or this will be a short fight'.

Maybe try actually doing that?

Things are definitely broken inside and outside of combat.
I've failed persuasion checks 9 times in a row with reloads, every time I have more than a 50% chance of meeting the check. .5^9= 1/500

I've missed in combat several times in a row despite having a 90+ chance to hit. The tooltips are either 100% wrong/meaningless or the roller is biased.

I suspect if somebody were to collect data you would find a pretty extreme bias.

Maybe I missed something on how Dice Rolls work on this game but when I have a +2 from ability score and a +3 from proficiency how is it possible for me to roll a 1?

The +2 and +3 are subtracted from the DC so all you see is the net DC as your objective roll to beat with your D20. The math is the same either way.

That's what I figured. Wish they'd show the explicit calculation in the chat window like they did in the Neverwinter games.

[quote=GunTheHuman]I've seen a Crab do an attack of 2d4 deal 1 damage on board of the Nautiloid :)[/quote]

Yes. If you open the combat log, you'll see that the Imps you were fighting have resistance to Slashing and Piercing damage - thus, they take half damage from these types of attacks. 2d4 / 2 can turn out as 1.